1DBI(3)                User Contributed Perl Documentation               DBI(3)
2
3
4

NAME

6       DBI - Database independent interface for Perl
7

SYNOPSIS

9         use DBI;
10
11         @driver_names = DBI->available_drivers;
12         %drivers      = DBI->installed_drivers;
13         @data_sources = DBI->data_sources($driver_name, \%attr);
14
15         $dbh = DBI->connect($data_source, $username, $auth, \%attr);
16
17         $rv  = $dbh->do($statement);
18         $rv  = $dbh->do($statement, \%attr);
19         $rv  = $dbh->do($statement, \%attr, @bind_values);
20
21         $ary_ref  = $dbh->selectall_arrayref($statement);
22         $hash_ref = $dbh->selectall_hashref($statement, $key_field);
23
24         $ary_ref  = $dbh->selectcol_arrayref($statement);
25         $ary_ref  = $dbh->selectcol_arrayref($statement, \%attr);
26
27         @row_ary  = $dbh->selectrow_array($statement);
28         $ary_ref  = $dbh->selectrow_arrayref($statement);
29         $hash_ref = $dbh->selectrow_hashref($statement);
30
31         $sth = $dbh->prepare($statement);
32         $sth = $dbh->prepare_cached($statement);
33
34         $rc = $sth->bind_param($p_num, $bind_value);
35         $rc = $sth->bind_param($p_num, $bind_value, $bind_type);
36         $rc = $sth->bind_param($p_num, $bind_value, \%attr);
37
38         $rv = $sth->execute;
39         $rv = $sth->execute(@bind_values);
40         $rv = $sth->execute_array(\%attr, ...);
41
42         $rc = $sth->bind_col($col_num, \$col_variable);
43         $rc = $sth->bind_columns(@list_of_refs_to_vars_to_bind);
44
45         @row_ary  = $sth->fetchrow_array;
46         $ary_ref  = $sth->fetchrow_arrayref;
47         $hash_ref = $sth->fetchrow_hashref;
48
49         $ary_ref  = $sth->fetchall_arrayref;
50         $ary_ref  = $sth->fetchall_arrayref( $slice, $max_rows );
51
52         $hash_ref = $sth->fetchall_hashref( $key_field );
53
54         $rv  = $sth->rows;
55
56         $rc  = $dbh->begin_work;
57         $rc  = $dbh->commit;
58         $rc  = $dbh->rollback;
59
60         $quoted_string = $dbh->quote($string);
61
62         $rc  = $h->err;
63         $str = $h->errstr;
64         $rv  = $h->state;
65
66         $rc  = $dbh->disconnect;
67
68       The synopsis above only lists the major methods and parameters.
69
70   GETTING HELP
71       General
72
73       Before asking any questions, reread this document, consult the archives
74       and read the DBI FAQ. The archives are listed at the end of this
75       document and on the DBI home page <http://dbi.perl.org/support/>
76
77       You might also like to read the Advanced DBI Tutorial at
78       <http://www.slideshare.net/Tim.Bunce/dbi-advanced-tutorial-2007>
79
80       To help you make the best use of the dbi-users mailing list, and any
81       other lists or forums you may use, I recommend that you read "Getting
82       Answers" by Mike Ash: <http://mikeash.com/getting_answers.html>.
83
84       Mailing Lists
85
86       If you have questions about DBI, or DBD driver modules, you can get
87       help from the dbi-users@perl.org mailing list. This is the best way to
88       get help. You don't have to subscribe to the list in order to post,
89       though I'd recommend it. You can get help on subscribing and using the
90       list by emailing dbi-users-help@perl.org.
91
92       Please note that Tim Bunce does not maintain the mailing lists or the
93       web pages (generous volunteers do that).  So please don't send mail
94       directly to him; he just doesn't have the time to answer questions
95       personally. The dbi-users mailing list has lots of experienced people
96       who should be able to help you if you need it. If you do email Tim he
97       is very likely to just forward it to the mailing list.
98
99       Online
100
101       StackOverflow has a DBI tag
102       <http://stackoverflow.com/questions/tagged/dbi> with over 400
103       questions.
104
105       The DBI home page at <http://dbi.perl.org/> and the DBI FAQ at
106       <http://faq.dbi-support.com/> may be worth a visit.  They include links
107       to other resources, but are rather out-dated.
108
109       I don't recommend the DBI cpanforum (at
110       http://www.cpanforum.com/dist/DBI) because relatively few people read
111       it compared with dbi-users@perl.org.
112
113       Reporting a Bug
114
115       If you think you've found a bug then please read "How to Report Bugs
116       Effectively" by Simon Tatham:
117       <http://www.chiark.greenend.org.uk/~sgtatham/bugs.html>.
118
119       Your problem is most likely related to the specific DBD driver module
120       you're using. If that's the case then click on the 'Bugs' link on the
121       <http://metacpan.org> page for your driver. Only submit a bug report
122       against the DBI itself if you're sure that your issue isn't related to
123       the driver you're using.
124
125   NOTES
126       This is the DBI specification that corresponds to DBI version 1.627
127       (see DBI::Changes for details).
128
129       The DBI is evolving at a steady pace, so it's good to check that you
130       have the latest copy.
131
132       The significant user-visible changes in each release are documented in
133       the DBI::Changes module so you can read them by executing "perldoc
134       DBI::Changes".
135
136       Some DBI changes require changes in the drivers, but the drivers can
137       take some time to catch up. Newer versions of the DBI have added
138       features that may not yet be supported by the drivers you use.  Talk to
139       the authors of your drivers if you need a new feature that is not yet
140       supported.
141
142       Features added after DBI 1.21 (February 2002) are marked in the text
143       with the version number of the DBI release they first appeared in.
144
145       Extensions to the DBI API often use the "DBIx::*" namespace.  See
146       "Naming Conventions and Name Space". DBI extension modules can be found
147       at <http://search.cpan.org/search?mode=module&query=DBIx>.  And all
148       modules related to the DBI can be found at
149       <http://search.cpan.org/search?query=DBI&mode=all>.
150

DESCRIPTION

152       The DBI is a database access module for the Perl programming language.
153       It defines a set of methods, variables, and conventions that provide a
154       consistent database interface, independent of the actual database being
155       used.
156
157       It is important to remember that the DBI is just an interface.  The DBI
158       is a layer of "glue" between an application and one or more database
159       driver modules.  It is the driver modules which do most of the real
160       work. The DBI provides a standard interface and framework for the
161       drivers to operate within.
162
163       This document often uses terms like references, objects, methods.  If
164       you're not familiar with those terms then it would be a good idea to
165       read at least the following perl manuals first: perlreftut, perldsc,
166       perllol, and perlboot.
167
168   Architecture of a DBI Application
169                    |<- Scope of DBI ->|
170                         .-.   .--------------.   .-------------.
171         .-------.       | |---| XYZ Driver   |---| XYZ Engine  |
172         | Perl  |       | |   `--------------'   `-------------'
173         | script|  |A|  |D|   .--------------.   .-------------.
174         | using |--|P|--|B|---|Oracle Driver |---|Oracle Engine|
175         | DBI   |  |I|  |I|   `--------------'   `-------------'
176         | API   |       | |...
177         |methods|       | |... Other drivers
178         `-------'       | |...
179                         `-'
180
181       The API, or Application Programming Interface, defines the call
182       interface and variables for Perl scripts to use. The API is implemented
183       by the Perl DBI extension.
184
185       The DBI "dispatches" the method calls to the appropriate driver for
186       actual execution.  The DBI is also responsible for the dynamic loading
187       of drivers, error checking and handling, providing default
188       implementations for methods, and many other non-database specific
189       duties.
190
191       Each driver contains implementations of the DBI methods using the
192       private interface functions of the corresponding database engine.  Only
193       authors of sophisticated/multi-database applications or generic library
194       functions need be concerned with drivers.
195
196   Notation and Conventions
197       The following conventions are used in this document:
198
199         $dbh    Database handle object
200         $sth    Statement handle object
201         $drh    Driver handle object (rarely seen or used in applications)
202         $h      Any of the handle types above ($dbh, $sth, or $drh)
203         $rc     General Return Code  (boolean: true=ok, false=error)
204         $rv     General Return Value (typically an integer)
205         @ary    List of values returned from the database, typically a row of data
206         $rows   Number of rows processed (if available, else -1)
207         $fh     A filehandle
208         undef   NULL values are represented by undefined values in Perl
209         \%attr  Reference to a hash of attribute values passed to methods
210
211       Note that Perl will automatically destroy database and statement handle
212       objects if all references to them are deleted.
213
214   Outline Usage
215       To use DBI, first you need to load the DBI module:
216
217         use DBI;
218         use strict;
219
220       (The "use strict;" isn't required but is strongly recommended.)
221
222       Then you need to "connect" to your data source and get a handle for
223       that connection:
224
225         $dbh = DBI->connect($dsn, $user, $password,
226                             { RaiseError => 1, AutoCommit => 0 });
227
228       Since connecting can be expensive, you generally just connect at the
229       start of your program and disconnect at the end.
230
231       Explicitly defining the required "AutoCommit" behaviour is strongly
232       recommended and may become mandatory in a later version.  This
233       determines whether changes are automatically committed to the database
234       when executed, or need to be explicitly committed later.
235
236       The DBI allows an application to "prepare" statements for later
237       execution.  A prepared statement is identified by a statement handle
238       held in a Perl variable.  We'll call the Perl variable $sth in our
239       examples.
240
241       The typical method call sequence for a "SELECT" statement is:
242
243         prepare,
244           execute, fetch, fetch, ...
245           execute, fetch, fetch, ...
246           execute, fetch, fetch, ...
247
248       for example:
249
250         $sth = $dbh->prepare("SELECT foo, bar FROM table WHERE baz=?");
251
252         $sth->execute( $baz );
253
254         while ( @row = $sth->fetchrow_array ) {
255           print "@row\n";
256         }
257
258       The typical method call sequence for a non-"SELECT" statement is:
259
260         prepare,
261           execute,
262           execute,
263           execute.
264
265       for example:
266
267         $sth = $dbh->prepare("INSERT INTO table(foo,bar,baz) VALUES (?,?,?)");
268
269         while(<CSV>) {
270           chomp;
271           my ($foo,$bar,$baz) = split /,/;
272               $sth->execute( $foo, $bar, $baz );
273         }
274
275       The "do()" method can be used for non repeated non-"SELECT" statement
276       (or with drivers that don't support placeholders):
277
278         $rows_affected = $dbh->do("UPDATE your_table SET foo = foo + 1");
279
280       To commit your changes to the database (when "AutoCommit" is off):
281
282         $dbh->commit;  # or call $dbh->rollback; to undo changes
283
284       Finally, when you have finished working with the data source, you
285       should "disconnect" from it:
286
287         $dbh->disconnect;
288
289   General Interface Rules & Caveats
290       The DBI does not have a concept of a "current session". Every session
291       has a handle object (i.e., a $dbh) returned from the "connect" method.
292       That handle object is used to invoke database related methods.
293
294       Most data is returned to the Perl script as strings. (Null values are
295       returned as "undef".)  This allows arbitrary precision numeric data to
296       be handled without loss of accuracy.  Beware that Perl may not preserve
297       the same accuracy when the string is used as a number.
298
299       Dates and times are returned as character strings in the current
300       default format of the corresponding database engine.  Time zone effects
301       are database/driver dependent.
302
303       Perl supports binary data in Perl strings, and the DBI will pass binary
304       data to and from the driver without change. It is up to the driver
305       implementors to decide how they wish to handle such binary data.
306
307       Perl supports two kinds of strings: Unicode (utf8 internally) and non-
308       Unicode (defaults to iso-8859-1 if forced to assume an encoding).
309       Drivers should accept both kinds of strings and, if required, convert
310       them to the character set of the database being used. Similarly, when
311       fetching from the database character data that isn't iso-8859-1 the
312       driver should convert it into utf8.
313
314       Multiple SQL statements may not be combined in a single statement
315       handle ($sth), although some databases and drivers do support this
316       (notably Sybase and SQL Server).
317
318       Non-sequential record reads are not supported in this version of the
319       DBI.  In other words, records can only be fetched in the order that the
320       database returned them, and once fetched they are forgotten.
321
322       Positioned updates and deletes are not directly supported by the DBI.
323       See the description of the "CursorName" attribute for an alternative.
324
325       Individual driver implementors are free to provide any private
326       functions and/or handle attributes that they feel are useful.  Private
327       driver functions can be invoked using the DBI "func()" method.  Private
328       driver attributes are accessed just like standard attributes.
329
330       Many methods have an optional "\%attr" parameter which can be used to
331       pass information to the driver implementing the method. Except where
332       specifically documented, the "\%attr" parameter can only be used to
333       pass driver specific hints. In general, you can ignore "\%attr"
334       parameters or pass it as "undef".
335
336   Naming Conventions and Name Space
337       The DBI package and all packages below it ("DBI::*") are reserved for
338       use by the DBI. Extensions and related modules use the "DBIx::"
339       namespace (see <http://www.perl.com/CPAN/modules/by-module/DBIx/>).
340       Package names beginning with "DBD::" are reserved for use by DBI
341       database drivers.  All environment variables used by the DBI or by
342       individual DBDs begin with ""DBI_"" or ""DBD_"".
343
344       The letter case used for attribute names is significant and plays an
345       important part in the portability of DBI scripts.  The case of the
346       attribute name is used to signify who defined the meaning of that name
347       and its values.
348
349         Case of name  Has a meaning defined by
350         ------------  ------------------------
351         UPPER_CASE    Standards, e.g.,  X/Open, ISO SQL92 etc (portable)
352         MixedCase     DBI API (portable), underscores are not used.
353         lower_case    Driver or database engine specific (non-portable)
354
355       It is of the utmost importance that Driver developers only use
356       lowercase attribute names when defining private attributes. Private
357       attribute names must be prefixed with the driver name or suitable
358       abbreviation (e.g., ""ora_"" for Oracle, ""ing_"" for Ingres, etc).
359
360   SQL - A Query Language
361       Most DBI drivers require applications to use a dialect of SQL
362       (Structured Query Language) to interact with the database engine.  The
363       "Standards Reference Information" section provides links to useful
364       information about SQL.
365
366       The DBI itself does not mandate or require any particular language to
367       be used; it is language independent. In ODBC terms, the DBI is in
368       "pass-thru" mode, although individual drivers might not be. The only
369       requirement is that queries and other statements must be expressed as a
370       single string of characters passed as the first argument to the
371       "prepare" or "do" methods.
372
373       For an interesting diversion on the real history of RDBMS and SQL, from
374       the people who made it happen, see:
375
376         http://www.mcjones.org/System_R/SQL_Reunion_95/sqlr95.html
377
378       Follow the "Full Contents" then "Intergalactic dataspeak" links for the
379       SQL history.
380
381   Placeholders and Bind Values
382       Some drivers support placeholders and bind values.  Placeholders, also
383       called parameter markers, are used to indicate values in a database
384       statement that will be supplied later, before the prepared statement is
385       executed.  For example, an application might use the following to
386       insert a row of data into the SALES table:
387
388         INSERT INTO sales (product_code, qty, price) VALUES (?, ?, ?)
389
390       or the following, to select the description for a product:
391
392         SELECT description FROM products WHERE product_code = ?
393
394       The "?" characters are the placeholders.  The association of actual
395       values with placeholders is known as binding, and the values are
396       referred to as bind values.  Note that the "?" is not enclosed in
397       quotation marks, even when the placeholder represents a string.
398
399       Some drivers also allow placeholders like ":"name and ":"N (e.g., ":1",
400       ":2", and so on) in addition to "?", but their use is not portable.
401
402       If the ":"N form of placeholder is supported by the driver you're
403       using, then you should be able to use either "bind_param" or "execute"
404       to bind values. Check your driver documentation.
405
406       With most drivers, placeholders can't be used for any element of a
407       statement that would prevent the database server from validating the
408       statement and creating a query execution plan for it. For example:
409
410         "SELECT name, age FROM ?"         # wrong (will probably fail)
411         "SELECT name, ?   FROM people"    # wrong (but may not 'fail')
412
413       Also, placeholders can only represent single scalar values.  For
414       example, the following statement won't work as expected for more than
415       one value:
416
417         "SELECT name, age FROM people WHERE name IN (?)"    # wrong
418         "SELECT name, age FROM people WHERE name IN (?,?)"  # two names
419
420       When using placeholders with the SQL "LIKE" qualifier, you must
421       remember that the placeholder substitutes for the whole string.  So you
422       should use ""... LIKE ? ..."" and include any wildcard characters in
423       the value that you bind to the placeholder.
424
425       NULL Values
426
427       Undefined values, or "undef", are used to indicate NULL values.  You
428       can insert and update columns with a NULL value as you would a non-NULL
429       value.  These examples insert and update the column "age" with a NULL
430       value:
431
432         $sth = $dbh->prepare(qq{
433           INSERT INTO people (fullname, age) VALUES (?, ?)
434         });
435         $sth->execute("Joe Bloggs", undef);
436
437         $sth = $dbh->prepare(qq{
438           UPDATE people SET age = ? WHERE fullname = ?
439         });
440         $sth->execute(undef, "Joe Bloggs");
441
442       However, care must be taken when trying to use NULL values in a "WHERE"
443       clause.  Consider:
444
445         SELECT fullname FROM people WHERE age = ?
446
447       Binding an "undef" (NULL) to the placeholder will not select rows which
448       have a NULL "age"!  At least for database engines that conform to the
449       SQL standard.  Refer to the SQL manual for your database engine or any
450       SQL book for the reasons for this.  To explicitly select NULLs you have
451       to say ""WHERE age IS NULL"".
452
453       A common issue is to have a code fragment handle a value that could be
454       either "defined" or "undef" (non-NULL or NULL) at runtime.  A simple
455       technique is to prepare the appropriate statement as needed, and
456       substitute the placeholder for non-NULL cases:
457
458         $sql_clause = defined $age? "age = ?" : "age IS NULL";
459         $sth = $dbh->prepare(qq{
460           SELECT fullname FROM people WHERE $sql_clause
461         });
462         $sth->execute(defined $age ? $age : ());
463
464       The following technique illustrates qualifying a "WHERE" clause with
465       several columns, whose associated values ("defined" or "undef") are in
466       a hash %h:
467
468         for my $col ("age", "phone", "email") {
469           if (defined $h{$col}) {
470             push @sql_qual, "$col = ?";
471             push @sql_bind, $h{$col};
472           }
473           else {
474             push @sql_qual, "$col IS NULL";
475           }
476         }
477         $sql_clause = join(" AND ", @sql_qual);
478         $sth = $dbh->prepare(qq{
479             SELECT fullname FROM people WHERE $sql_clause
480         });
481         $sth->execute(@sql_bind);
482
483       The techniques above call prepare for the SQL statement with each call
484       to execute.  Because calls to prepare() can be expensive, performance
485       can suffer when an application iterates many times over statements like
486       the above.
487
488       A better solution is a single "WHERE" clause that supports both NULL
489       and non-NULL comparisons.  Its SQL statement would need to be prepared
490       only once for all cases, thus improving performance.  Several examples
491       of "WHERE" clauses that support this are presented below.  But each
492       example lacks portability, robustness, or simplicity.  Whether an
493       example is supported on your database engine depends on what SQL
494       extensions it provides, and where it supports the "?"  placeholder in a
495       statement.
496
497         0)  age = ?
498         1)  NVL(age, xx) = NVL(?, xx)
499         2)  ISNULL(age, xx) = ISNULL(?, xx)
500         3)  DECODE(age, ?, 1, 0) = 1
501         4)  age = ? OR (age IS NULL AND ? IS NULL)
502         5)  age = ? OR (age IS NULL AND SP_ISNULL(?) = 1)
503         6)  age = ? OR (age IS NULL AND ? = 1)
504
505       Statements formed with the above "WHERE" clauses require execute
506       statements as follows.  The arguments are required, whether their
507       values are "defined" or "undef".
508
509         0,1,2,3)  $sth->execute($age);
510         4,5)      $sth->execute($age, $age);
511         6)        $sth->execute($age, defined($age) ? 0 : 1);
512
513       Example 0 should not work (as mentioned earlier), but may work on a few
514       database engines anyway (e.g. Sybase).  Example 0 is part of examples
515       4, 5, and 6, so if example 0 works, these other examples may work, even
516       if the engine does not properly support the right hand side of the "OR"
517       expression.
518
519       Examples 1 and 2 are not robust: they require that you provide a valid
520       column value xx (e.g. '~') which is not present in any row.  That means
521       you must have some notion of what data won't be stored in the column,
522       and expect clients to adhere to that.
523
524       Example 5 requires that you provide a stored procedure (SP_ISNULL in
525       this example) that acts as a function: it checks whether a value is
526       null, and returns 1 if it is, or 0 if not.
527
528       Example 6, the least simple, is probably the most portable, i.e., it
529       should work with with most, if not all, database engines.
530
531       Here is a table that indicates which examples above are known to work
532       on various database engines:
533
534                          -----Examples------
535                          0  1  2  3  4  5  6
536                          -  -  -  -  -  -  -
537         Oracle 9         N  Y  N  Y  Y  ?  Y
538         Informix IDS 9   N  N  N  Y  N  Y  Y
539         MS SQL           N  N  Y  N  Y  ?  Y
540         Sybase           Y  N  N  N  N  N  Y
541         AnyData,DBM,CSV  Y  N  N  N  Y  Y* Y
542         SQLite 3.3       N  N  N  N  Y  N  N
543         MSAccess         N  N  N  N  Y  N  Y
544
545       * Works only because Example 0 works.
546
547       DBI provides a sample perl script that will test the examples above on
548       your database engine and tell you which ones work.  It is located in
549       the ex/ subdirectory of the DBI source distribution, or here:
550       <http://svn.perl.org/modules/dbi/trunk/ex/perl_dbi_nulls_test.pl>
551       Please use the script to help us fill-in and maintain this table.
552
553       Performance
554
555       Without using placeholders, the insert statement shown previously would
556       have to contain the literal values to be inserted and would have to be
557       re-prepared and re-executed for each row. With placeholders, the insert
558       statement only needs to be prepared once. The bind values for each row
559       can be given to the "execute" method each time it's called. By avoiding
560       the need to re-prepare the statement for each row, the application
561       typically runs many times faster. Here's an example:
562
563         my $sth = $dbh->prepare(q{
564           INSERT INTO sales (product_code, qty, price) VALUES (?, ?, ?)
565         }) or die $dbh->errstr;
566         while (<>) {
567             chomp;
568             my ($product_code, $qty, $price) = split /,/;
569             $sth->execute($product_code, $qty, $price) or die $dbh->errstr;
570         }
571         $dbh->commit or die $dbh->errstr;
572
573       See "execute" and "bind_param" for more details.
574
575       The "q{...}" style quoting used in this example avoids clashing with
576       quotes that may be used in the SQL statement. Use the double-quote like
577       "qq{...}" operator if you want to interpolate variables into the
578       string.  See "Quote and Quote-like Operators" in perlop for more
579       details.
580
581       See also the "bind_columns" method, which is used to associate Perl
582       variables with the output columns of a "SELECT" statement.
583

THE DBI PACKAGE AND CLASS

585       In this section, we cover the DBI class methods, utility functions, and
586       the dynamic attributes associated with generic DBI handles.
587
588   DBI Constants
589       Constants representing the values of the SQL standard types can be
590       imported individually by name, or all together by importing the special
591       ":sql_types" tag.
592
593       The names and values of all the defined SQL standard types can be
594       produced like this:
595
596         foreach (@{ $DBI::EXPORT_TAGS{sql_types} }) {
597           printf "%s=%d\n", $_, &{"DBI::$_"};
598         }
599
600       These constants are defined by SQL/CLI, ODBC or both.  "SQL_BIGINT" is
601       (currently) omitted, because SQL/CLI and ODBC provide conflicting
602       codes.
603
604       See the "type_info", "type_info_all", and "bind_param" methods for
605       possible uses.
606
607       Note that just because the DBI defines a named constant for a given
608       data type doesn't mean that drivers will support that data type.
609
610   DBI Class Methods
611       The following methods are provided by the DBI class:
612
613       "parse_dsn"
614
615         ($scheme, $driver, $attr_string, $attr_hash, $driver_dsn) = DBI->parse_dsn($dsn)
616             or die "Can't parse DBI DSN '$dsn'";
617
618       Breaks apart a DBI Data Source Name (DSN) and returns the individual
619       parts. If $dsn doesn't contain a valid DSN then parse_dsn() returns an
620       empty list.
621
622       $scheme is the first part of the DSN and is currently always 'dbi'.
623       $driver is the driver name, possibly defaulted to $ENV{DBI_DRIVER}, and
624       may be undefined.  $attr_string is the contents of the optional
625       attribute string, which may be undefined.  If $attr_string is not empty
626       then $attr_hash is a reference to a hash containing the parsed
627       attribute names and values.  $driver_dsn is the last part of the DBI
628       DSN string. For example:
629
630         ($scheme, $driver, $attr_string, $attr_hash, $driver_dsn)
631             = DBI->parse_dsn("DBI:MyDriver(RaiseError=>1):db=test;port=42");
632         $scheme      = 'dbi';
633         $driver      = 'MyDriver';
634         $attr_string = 'RaiseError=>1';
635         $attr_hash   = { 'RaiseError' => '1' };
636         $driver_dsn  = 'db=test;port=42';
637
638       The parse_dsn() method was added in DBI 1.43.
639
640       "connect"
641
642         $dbh = DBI->connect($data_source, $username, $password)
643                   or die $DBI::errstr;
644         $dbh = DBI->connect($data_source, $username, $password, \%attr)
645                   or die $DBI::errstr;
646
647       Establishes a database connection, or session, to the requested
648       $data_source.  Returns a database handle object if the connection
649       succeeds. Use "$dbh->disconnect" to terminate the connection.
650
651       If the connect fails (see below), it returns "undef" and sets both
652       $DBI::err and $DBI::errstr. (It does not explicitly set $!.) You should
653       generally test the return status of "connect" and "print $DBI::errstr"
654       if it has failed.
655
656       Multiple simultaneous connections to multiple databases through
657       multiple drivers can be made via the DBI. Simply make one "connect"
658       call for each database and keep a copy of each returned database
659       handle.
660
661       The $data_source value must begin with ""dbi:"driver_name":"".  The
662       driver_name specifies the driver that will be used to make the
663       connection. (Letter case is significant.)
664
665       As a convenience, if the $data_source parameter is undefined or empty,
666       the DBI will substitute the value of the environment variable
667       "DBI_DSN".  If just the driver_name part is empty (i.e., the
668       $data_source prefix is ""dbi::""), the environment variable
669       "DBI_DRIVER" is used. If neither variable is set, then "connect" dies.
670
671       Examples of $data_source values are:
672
673         dbi:DriverName:database_name
674         dbi:DriverName:database_name@hostname:port
675         dbi:DriverName:database=database_name;host=hostname;port=port
676
677       There is no standard for the text following the driver name. Each
678       driver is free to use whatever syntax it wants. The only requirement
679       the DBI makes is that all the information is supplied in a single
680       string.  You must consult the documentation for the drivers you are
681       using for a description of the syntax they require.
682
683       It is recommended that drivers support the ODBC style, shown in the
684       last example above. It is also recommended that that they support the
685       three common names '"host"', '"port"', and '"database"' (plus '"db"' as
686       an alias for "database"). This simplifies automatic construction of
687       basic DSNs: "dbi:$driver:database=$db;host=$host;port=$port".  Drivers
688       should aim to 'do something reasonable' when given a DSN in this form,
689       but if any part is meaningless for that driver (such as 'port' for
690       Informix) it should generate an error if that part is not empty.
691
692       If the environment variable "DBI_AUTOPROXY" is defined (and the driver
693       in $data_source is not ""Proxy"") then the connect request will
694       automatically be changed to:
695
696         $ENV{DBI_AUTOPROXY};dsn=$data_source
697
698       "DBI_AUTOPROXY" is typically set as
699       ""dbi:Proxy:hostname=...;port=..."".  If $ENV{DBI_AUTOPROXY} doesn't
700       begin with '"dbi:"' then "dbi:Proxy:" will be prepended to it first.
701       See the DBD::Proxy documentation for more details.
702
703       If $username or $password are undefined (rather than just empty), then
704       the DBI will substitute the values of the "DBI_USER" and "DBI_PASS"
705       environment variables, respectively.  The DBI will warn if the
706       environment variables are not defined.  However, the everyday use of
707       these environment variables is not recommended for security reasons.
708       The mechanism is primarily intended to simplify testing.  See below for
709       alternative way to specify the username and password.
710
711       "DBI->connect" automatically installs the driver if it has not been
712       installed yet. Driver installation either returns a valid driver
713       handle, or it dies with an error message that includes the string
714       ""install_driver"" and the underlying problem. So "DBI->connect" will
715       die on a driver installation failure and will only return "undef" on a
716       connect failure, in which case $DBI::errstr will hold the error
717       message.  Use "eval { ... }" if you need to catch the
718       ""install_driver"" error.
719
720       The $data_source argument (with the ""dbi:...:"" prefix removed) and
721       the $username and $password arguments are then passed to the driver for
722       processing. The DBI does not define any interpretation for the contents
723       of these fields.  The driver is free to interpret the $data_source,
724       $username, and $password fields in any way, and supply whatever
725       defaults are appropriate for the engine being accessed.  (Oracle, for
726       example, uses the ORACLE_SID and TWO_TASK environment variables if no
727       $data_source is specified.)
728
729       The "AutoCommit" and "PrintError" attributes for each connection
730       default to "on". (See "AutoCommit" and "PrintError" for more
731       information.)  However, it is strongly recommended that you explicitly
732       define "AutoCommit" rather than rely on the default. The "PrintWarn"
733       attribute defaults to on if $^W is true, i.e., perl is running with
734       warnings enabled.
735
736       The "\%attr" parameter can be used to alter the default settings of
737       "PrintError", "RaiseError", "AutoCommit", and other attributes. For
738       example:
739
740         $dbh = DBI->connect($data_source, $user, $pass, {
741               PrintError => 0,
742               AutoCommit => 0
743         });
744
745       The username and password can also be specified using the attributes
746       "Username" and "Password", in which case they take precedence over the
747       $username and $password parameters.
748
749       You can also define connection attribute values within the $data_source
750       parameter. For example:
751
752         dbi:DriverName(PrintWarn=>1,PrintError=>0,Taint=>1):...
753
754       Individual attributes values specified in this way take precedence over
755       any conflicting values specified via the "\%attr" parameter to
756       "connect".
757
758       The "dbi_connect_method" attribute can be used to specify which driver
759       method should be called to establish the connection. The only useful
760       values are 'connect', 'connect_cached', or some specialized case like
761       'Apache::DBI::connect' (which is automatically the default when running
762       within Apache).
763
764       Where possible, each session ($dbh) is independent from the
765       transactions in other sessions. This is useful when you need to hold
766       cursors open across transactions--for example, if you use one session
767       for your long lifespan cursors (typically read-only) and another for
768       your short update transactions.
769
770       For compatibility with old DBI scripts, the driver can be specified by
771       passing its name as the fourth argument to "connect" (instead of
772       "\%attr"):
773
774         $dbh = DBI->connect($data_source, $user, $pass, $driver);
775
776       In this "old-style" form of "connect", the $data_source should not
777       start with ""dbi:driver_name:"". (If it does, the embedded driver_name
778       will be ignored). Also note that in this older form of "connect", the
779       "$dbh->{AutoCommit}" attribute is undefined, the "$dbh->{PrintError}"
780       attribute is off, and the old "DBI_DBNAME" environment variable is
781       checked if "DBI_DSN" is not defined. Beware that this "old-style"
782       "connect" will soon be withdrawn in a future version of DBI.
783
784       "connect_cached"
785
786         $dbh = DBI->connect_cached($data_source, $username, $password)
787                   or die $DBI::errstr;
788         $dbh = DBI->connect_cached($data_source, $username, $password, \%attr)
789                   or die $DBI::errstr;
790
791       "connect_cached" is like "connect", except that the database handle
792       returned is also stored in a hash associated with the given parameters.
793       If another call is made to "connect_cached" with the same parameter
794       values, then the corresponding cached $dbh will be returned if it is
795       still valid.  The cached database handle is replaced with a new
796       connection if it has been disconnected or if the "ping" method fails.
797
798       Note that the behaviour of this method differs in several respects from
799       the behaviour of persistent connections implemented by Apache::DBI.
800       However, if Apache::DBI is loaded then "connect_cached" will use it.
801
802       Caching connections can be useful in some applications, but it can also
803       cause problems, such as too many connections, and so should be used
804       with care. In particular, avoid changing the attributes of a database
805       handle created via connect_cached() because it will affect other code
806       that may be using the same handle. When connect_cached() returns a
807       handle the attributes will be reset to their initial values.  This can
808       cause problems, especially with the "AutoCommit" attribute.
809
810       Where multiple separate parts of a program are using connect_cached()
811       to connect to the same database with the same (initial) attributes it
812       is a good idea to add a private attribute to the connect_cached() call
813       to effectively limit the scope of the caching. For example:
814
815         DBI->connect_cached(..., { private_foo_cachekey => "Bar", ... });
816
817       Handles returned from that connect_cached() call will only be returned
818       by other connect_cached() call elsewhere in the code if those other
819       calls also pass in the same attribute values, including the private
820       one.  (I've used "private_foo_cachekey" here as an example, you can use
821       any attribute name with a "private_" prefix.)
822
823       Taking that one step further, you can limit a particular
824       connect_cached() call to return handles unique to that one place in the
825       code by setting the private attribute to a unique value for that place:
826
827         DBI->connect_cached(..., { private_foo_cachekey => __FILE__.__LINE__, ... });
828
829       By using a private attribute you still get connection caching for the
830       individual calls to connect_cached() but, by making separate database
831       connections for separate parts of the code, the database handles are
832       isolated from any attribute changes made to other handles.
833
834       The cache can be accessed (and cleared) via the "CachedKids" attribute:
835
836         my $CachedKids_hashref = $dbh->{Driver}->{CachedKids};
837         %$CachedKids_hashref = () if $CachedKids_hashref;
838
839       "available_drivers"
840
841         @ary = DBI->available_drivers;
842         @ary = DBI->available_drivers($quiet);
843
844       Returns a list of all available drivers by searching for "DBD::*"
845       modules through the directories in @INC. By default, a warning is given
846       if some drivers are hidden by others of the same name in earlier
847       directories. Passing a true value for $quiet will inhibit the warning.
848
849       "installed_drivers"
850
851         %drivers = DBI->installed_drivers();
852
853       Returns a list of driver name and driver handle pairs for all drivers
854       'installed' (loaded) into the current process.  The driver name does
855       not include the 'DBD::' prefix.
856
857       To get a list of all drivers available in your perl installation you
858       can use "available_drivers".
859
860       Added in DBI 1.49.
861
862       "installed_versions"
863
864         DBI->installed_versions;
865         @ary  = DBI->installed_versions;
866         $hash = DBI->installed_versions;
867
868       Calls available_drivers() and attempts to load each of them in turn
869       using install_driver().  For each load that succeeds the driver name
870       and version number are added to a hash. When running under
871       DBI::PurePerl drivers which appear not be pure-perl are ignored.
872
873       When called in array context the list of successfully loaded drivers is
874       returned (without the 'DBD::' prefix).
875
876       When called in scalar context an extra entry for the "DBI" is added
877       (and "DBI::PurePerl" if appropriate) and a reference to the hash is
878       returned.
879
880       When called in a void context the installed_versions() method will
881       print out a formatted list of the hash contents, one per line, along
882       with some other information about the DBI version and OS.
883
884       Due to the potentially high memory cost and unknown risks of loading in
885       an unknown number of drivers that just happen to be installed on the
886       system, this method is not recommended for general use.  Use
887       available_drivers() instead.
888
889       The installed_versions() method is primarily intended as a quick way to
890       see from the command line what's installed. For example:
891
892         perl -MDBI -e 'DBI->installed_versions'
893
894       The installed_versions() method was added in DBI 1.38.
895
896       "data_sources"
897
898         @ary = DBI->data_sources($driver);
899         @ary = DBI->data_sources($driver, \%attr);
900
901       Returns a list of data sources (databases) available via the named
902       driver.  If $driver is empty or "undef", then the value of the
903       "DBI_DRIVER" environment variable is used.
904
905       The driver will be loaded if it hasn't been already. Note that if the
906       driver loading fails then data_sources() dies with an error message
907       that includes the string ""install_driver"" and the underlying problem.
908
909       Data sources are returned in a form suitable for passing to the
910       "connect" method (that is, they will include the ""dbi:$driver:""
911       prefix).
912
913       Note that many drivers have no way of knowing what data sources might
914       be available for it. These drivers return an empty or incomplete list
915       or may require driver-specific attributes.
916
917       There is also a data_sources() method defined for database handles.
918
919       "trace"
920
921         DBI->trace($trace_setting)
922         DBI->trace($trace_setting, $trace_filename)
923         DBI->trace($trace_setting, $trace_filehandle)
924         $trace_setting = DBI->trace;
925
926       The "DBI->trace" method sets the global default trace settings and
927       returns the previous trace settings. It can also be used to change
928       where the trace output is sent.
929
930       There's a similar method, "$h->trace", which sets the trace settings
931       for the specific handle it's called on.
932
933       See the "TRACING" section for full details about the DBI's powerful
934       tracing facilities.
935
936       "visit_handles"
937
938         DBI->visit_handles( $coderef );
939         DBI->visit_handles( $coderef, $info );
940
941       Where $coderef is a reference to a subroutine and $info is an arbitrary
942       value which, if undefined, defaults to a reference to an empty hash.
943       Returns $info.
944
945       For each installed driver handle, if any, $coderef is invoked as:
946
947         $coderef->($driver_handle, $info);
948
949       If the execution of $coderef returns a true value then
950       "visit_child_handles" is called on that child handle and passed the
951       returned value as $info.
952
953       For example:
954
955         my $info = $dbh->{Driver}->visit_child_handles(sub {
956             my ($h, $info) = @_;
957             ++$info->{ $h->{Type} }; # count types of handles (dr/db/st)
958             return $info; # visit kids
959         });
960
961       See also "visit_child_handles".
962
963   DBI Utility Functions
964       In addition to the DBI methods listed in the previous section, the DBI
965       package also provides several utility functions.
966
967       These can be imported into your code by listing them in the "use"
968       statement. For example:
969
970         use DBI qw(neat data_diff);
971
972       Alternatively, all these utility functions (except hash) can be
973       imported using the ":utils" import tag. For example:
974
975         use DBI qw(:utils);
976
977       "data_string_desc"
978
979         $description = data_string_desc($string);
980
981       Returns an informal description of the string. For example:
982
983         UTF8 off, ASCII, 42 characters 42 bytes
984         UTF8 off, non-ASCII, 42 characters 42 bytes
985         UTF8 on, non-ASCII, 4 characters 6 bytes
986         UTF8 on but INVALID encoding, non-ASCII, 4 characters 6 bytes
987         UTF8 off, undef
988
989       The initial "UTF8" on/off refers to Perl's internal SvUTF8 flag.  If
990       $string has the SvUTF8 flag set but the sequence of bytes it contains
991       are not a valid UTF-8 encoding then data_string_desc() will report
992       "UTF8 on but INVALID encoding".
993
994       The "ASCII" vs "non-ASCII" portion shows "ASCII" if all the characters
995       in the string are ASCII (have code points <= 127).
996
997       The data_string_desc() function was added in DBI 1.46.
998
999       "data_string_diff"
1000
1001         $diff = data_string_diff($a, $b);
1002
1003       Returns an informal description of the first character difference
1004       between the strings. If both $a and $b contain the same sequence of
1005       characters then data_string_diff() returns an empty string.  For
1006       example:
1007
1008        Params a & b     Result
1009        ------------     ------
1010        'aaa', 'aaa'     ''
1011        'aaa', 'abc'     'Strings differ at index 2: a[2]=a, b[2]=b'
1012        'aaa', undef     'String b is undef, string a has 3 characters'
1013        'aaa', 'aa'      'String b truncated after 2 characters'
1014
1015       Unicode characters are reported in "\x{XXXX}" format. Unicode code
1016       points in the range U+0800 to U+08FF are unassigned and most likely to
1017       occur due to double-encoding. Characters in this range are reported as
1018       "\x{08XX}='C'" where "C" is the corresponding latin-1 character.
1019
1020       The data_string_diff() function only considers logical characters and
1021       not the underlying encoding. See "data_diff" for an alternative.
1022
1023       The data_string_diff() function was added in DBI 1.46.
1024
1025       "data_diff"
1026
1027         $diff = data_diff($a, $b);
1028         $diff = data_diff($a, $b, $logical);
1029
1030       Returns an informal description of the difference between two strings.
1031       It calls "data_string_desc" and "data_string_diff" and returns the
1032       combined results as a multi-line string.
1033
1034       For example, "data_diff("abc", "ab\x{263a}")" will return:
1035
1036         a: UTF8 off, ASCII, 3 characters 3 bytes
1037         b: UTF8 on, non-ASCII, 3 characters 5 bytes
1038         Strings differ at index 2: a[2]=c, b[2]=\x{263A}
1039
1040       If $a and $b are identical in both the characters they contain and
1041       their physical encoding then data_diff() returns an empty string.  If
1042       $logical is true then physical encoding differences are ignored (but
1043       are still reported if there is a difference in the characters).
1044
1045       The data_diff() function was added in DBI 1.46.
1046
1047       "neat"
1048
1049         $str = neat($value);
1050         $str = neat($value, $maxlen);
1051
1052       Return a string containing a neat (and tidy) representation of the
1053       supplied value.
1054
1055       Strings will be quoted, although internal quotes will not be escaped.
1056       Values known to be numeric will be unquoted. Undefined (NULL) values
1057       will be shown as "undef" (without quotes).
1058
1059       If the string is flagged internally as utf8 then double quotes will be
1060       used, otherwise single quotes are used and unprintable characters will
1061       be replaced by dot (.).
1062
1063       For result strings longer than $maxlen the result string will be
1064       truncated to "$maxlen-4" and ""...'"" will be appended.  If $maxlen is
1065       0 or "undef", it defaults to $DBI::neat_maxlen which, in turn, defaults
1066       to 400.
1067
1068       This function is designed to format values for human consumption.  It
1069       is used internally by the DBI for "trace" output. It should typically
1070       not be used for formatting values for database use.  (See also
1071       "quote".)
1072
1073       "neat_list"
1074
1075         $str = neat_list(\@listref, $maxlen, $field_sep);
1076
1077       Calls "neat" on each element of the list and returns a string
1078       containing the results joined with $field_sep. $field_sep defaults to
1079       ", ".
1080
1081       "looks_like_number"
1082
1083         @bool = looks_like_number(@array);
1084
1085       Returns true for each element that looks like a number.  Returns false
1086       for each element that does not look like a number.  Returns "undef" for
1087       each element that is undefined or empty.
1088
1089       "hash"
1090
1091         $hash_value = DBI::hash($buffer, $type);
1092
1093       Return a 32-bit integer 'hash' value corresponding to the contents of
1094       $buffer.  The $type parameter selects which kind of hash algorithm
1095       should be used.
1096
1097       For the technically curious, type 0 (which is the default if $type
1098       isn't specified) is based on the Perl 5.1 hash except that the value is
1099       forced to be negative (for obscure historical reasons).  Type 1 is the
1100       better "Fowler / Noll / Vo" (FNV) hash. See
1101       <http://www.isthe.com/chongo/tech/comp/fnv/> for more information.
1102       Both types are implemented in C and are very fast.
1103
1104       This function doesn't have much to do with databases, except that it
1105       can be handy to store hash values in a database.
1106
1107       "sql_type_cast"
1108
1109         $sts = DBI::sql_type_cast($sv, $sql_type, $flags);
1110
1111       sql_type_cast attempts to cast $sv to the SQL type (see DBI Constants)
1112       specified in $sql_type. At present only the SQL types "SQL_INTEGER",
1113       "SQL_DOUBLE" and "SQL_NUMERIC" are supported.
1114
1115       For "SQL_INTEGER" the effect is similar to using the value in an
1116       expression that requires an integer. It gives the perl scalar an
1117       'integer aspect'.  (Technically the value gains an IV, or possibly a UV
1118       or NV if the value is too large for an IV.)
1119
1120       For "SQL_DOUBLE" the effect is similar to using the value in an
1121       expression that requires a general numeric value. It gives the perl
1122       scalar a 'numeric aspect'.  (Technically the value gains an NV.)
1123
1124       "SQL_NUMERIC" is similar to "SQL_INTEGER" or "SQL_DOUBLE" but more
1125       general and more cautious.  It will look at the string first and if it
1126       looks like an integer (that will fit in an IV or UV) it will act like
1127       "SQL_INTEGER", if it looks like a floating point value it will act like
1128       "SQL_DOUBLE", if it looks like neither then it will do nothing - and
1129       thereby avoid the warnings that would be generated by "SQL_INTEGER" and
1130       "SQL_DOUBLE" when given non-numeric data.
1131
1132       $flags may be:
1133
1134       "DBIstcf_DISCARD_STRING"
1135           If this flag is specified then when the driver successfully casts
1136           the bound perl scalar to a non-string type then the string portion
1137           of the scalar will be discarded.
1138
1139       "DBIstcf_STRICT"
1140           If $sv cannot be cast to the requested $sql_type then by default it
1141           is left untouched and no error is generated. If you specify
1142           "DBIstcf_STRICT" and the cast fails, this will generate an error.
1143
1144       The returned $sts value is:
1145
1146         -2 sql_type is not handled
1147         -1 sv is undef so unchanged
1148          0 sv could not be cast cleanly and DBIstcf_STRICT was used
1149          1 sv could not be cast and DBIstcf_STRICT was not used
1150          2 sv was cast successfully
1151
1152       This method is exported by the :utils tag and was introduced in DBI
1153       1.611.
1154
1155   DBI Dynamic Attributes
1156       Dynamic attributes are always associated with the last handle used
1157       (that handle is represented by $h in the descriptions below).
1158
1159       Where an attribute is equivalent to a method call, then refer to the
1160       method call for all related documentation.
1161
1162       Warning: these attributes are provided as a convenience but they do
1163       have limitations. Specifically, they have a short lifespan: because
1164       they are associated with the last handle used, they should only be used
1165       immediately after calling the method that "sets" them.  If in any
1166       doubt, use the corresponding method call.
1167
1168       $DBI::err
1169
1170       Equivalent to "$h->err".
1171
1172       $DBI::errstr
1173
1174       Equivalent to "$h->errstr".
1175
1176       $DBI::state
1177
1178       Equivalent to "$h->state".
1179
1180       $DBI::rows
1181
1182       Equivalent to "$h->rows". Please refer to the documentation for the
1183       "rows" method.
1184
1185       $DBI::lasth
1186
1187       Returns the DBI object handle used for the most recent DBI method call.
1188       If the last DBI method call was a DESTROY then $DBI::lasth will return
1189       the handle of the parent of the destroyed handle, if there is one.
1190

METHODS COMMON TO ALL HANDLES

1192       The following methods can be used by all types of DBI handles.
1193
1194       "err"
1195
1196         $rv = $h->err;
1197
1198       Returns the native database engine error code from the last driver
1199       method called. The code is typically an integer but you should not
1200       assume that.
1201
1202       The DBI resets $h->err to undef before almost all DBI method calls, so
1203       the value only has a short lifespan. Also, for most drivers, the
1204       statement handles share the same error variable as the parent database
1205       handle, so calling a method on one handle may reset the error on the
1206       related handles.
1207
1208       (Methods which don't reset err before being called include err() and
1209       errstr(), obviously, state(), rows(), func(), trace(), trace_msg(),
1210       ping(), and the tied hash attribute FETCH() and STORE() methods.)
1211
1212       If you need to test for specific error conditions and have your program
1213       be portable to different database engines, then you'll need to
1214       determine what the corresponding error codes are for all those engines
1215       and test for all of them.
1216
1217       The DBI uses the value of $DBI::stderr as the "err" value for internal
1218       errors.  Drivers should also do likewise.  The default value for
1219       $DBI::stderr is 2000000000.
1220
1221       A driver may return 0 from err() to indicate a warning condition after
1222       a method call. Similarly, a driver may return an empty string to
1223       indicate a 'success with information' condition. In both these cases
1224       the value is false but not undef. The errstr() and state() methods may
1225       be used to retrieve extra information in these cases.
1226
1227       See "set_err" for more information.
1228
1229       "errstr"
1230
1231         $str = $h->errstr;
1232
1233       Returns the native database engine error message from the last DBI
1234       method called. This has the same lifespan issues as the "err" method
1235       described above.
1236
1237       The returned string may contain multiple messages separated by newline
1238       characters.
1239
1240       The errstr() method should not be used to test for errors, use err()
1241       for that, because drivers may return 'success with information' or
1242       warning messages via errstr() for methods that have not 'failed'.
1243
1244       See "set_err" for more information.
1245
1246       "state"
1247
1248         $str = $h->state;
1249
1250       Returns a state code in the standard SQLSTATE five character format.
1251       Note that the specific success code 00000 is translated to any empty
1252       string (false). If the driver does not support SQLSTATE (and most
1253       don't), then state() will return "S1000" (General Error) for all
1254       errors.
1255
1256       The driver is free to return any value via "state", e.g., warning
1257       codes, even if it has not declared an error by returning a true value
1258       via the "err" method described above.
1259
1260       The state() method should not be used to test for errors, use err() for
1261       that, because drivers may return a 'success with information' or
1262       warning state code via state() for methods that have not 'failed'.
1263
1264       "set_err"
1265
1266         $rv = $h->set_err($err, $errstr);
1267         $rv = $h->set_err($err, $errstr, $state);
1268         $rv = $h->set_err($err, $errstr, $state, $method);
1269         $rv = $h->set_err($err, $errstr, $state, $method, $rv);
1270
1271       Set the "err", "errstr", and "state" values for the handle.  This
1272       method is typically only used by DBI drivers and DBI subclasses.
1273
1274       If the "HandleSetErr" attribute holds a reference to a subroutine it is
1275       called first. The subroutine can alter the $err, $errstr, $state, and
1276       $method values. See "HandleSetErr" for full details.  If the subroutine
1277       returns a true value then the handle "err", "errstr", and "state"
1278       values are not altered and set_err() returns an empty list (it normally
1279       returns $rv which defaults to undef, see below).
1280
1281       Setting "err" to a true value indicates an error and will trigger the
1282       normal DBI error handling mechanisms, such as "RaiseError" and
1283       "HandleError", if they are enabled, when execution returns from the DBI
1284       back to the application.
1285
1286       Setting "err" to "" indicates an 'information' state, and setting it to
1287       "0" indicates a 'warning' state. Setting "err" to "undef" also sets
1288       "errstr" to undef, and "state" to "", irrespective of the values of the
1289       $errstr and $state parameters.
1290
1291       The $method parameter provides an alternate method name for the
1292       "RaiseError"/"PrintError"/"PrintWarn" error string instead of the
1293       fairly unhelpful '"set_err"'.
1294
1295       The "set_err" method normally returns undef.  The $rv parameter
1296       provides an alternate return value.
1297
1298       Some special rules apply if the "err" or "errstr" values for the handle
1299       are already set...
1300
1301       If "errstr" is true then: "" [err was %s now %s]"" is appended if $err
1302       is true and "err" is already true and the new err value differs from
1303       the original one. Similarly "" [state was %s now %s]"" is appended if
1304       $state is true and "state" is already true and the new state value
1305       differs from the original one. Finally ""\n"" and the new $errstr are
1306       appended if $errstr differs from the existing errstr value. Obviously
1307       the %s's above are replaced by the corresponding values.
1308
1309       The handle "err" value is set to $err if: $err is true; or handle "err"
1310       value is undef; or $err is defined and the length is greater than the
1311       handle "err" length. The effect is that an 'information' state only
1312       overrides undef; a 'warning' overrides undef or 'information', and an
1313       'error' state overrides anything.
1314
1315       The handle "state" value is set to $state if $state is true and the
1316       handle "err" value was set (by the rules above).
1317
1318       Support for warning and information states was added in DBI 1.41.
1319
1320       "trace"
1321
1322         $h->trace($trace_settings);
1323         $h->trace($trace_settings, $trace_filename);
1324         $trace_settings = $h->trace;
1325
1326       The trace() method is used to alter the trace settings for a handle
1327       (and any future children of that handle).  It can also be used to
1328       change where the trace output is sent.
1329
1330       There's a similar method, "DBI->trace", which sets the global default
1331       trace settings.
1332
1333       See the "TRACING" section for full details about the DBI's powerful
1334       tracing facilities.
1335
1336       "trace_msg"
1337
1338         $h->trace_msg($message_text);
1339         $h->trace_msg($message_text, $min_level);
1340
1341       Writes $message_text to the trace file if the trace level is greater
1342       than or equal to $min_level (which defaults to 1).  Can also be called
1343       as "DBI->trace_msg($msg)".
1344
1345       See "TRACING" for more details.
1346
1347       "func"
1348
1349         $h->func(@func_arguments, $func_name) or die ...;
1350
1351       The "func" method can be used to call private non-standard and non-
1352       portable methods implemented by the driver. Note that the function name
1353       is given as the last argument.
1354
1355       It's also important to note that the func() method does not clear a
1356       previous error ($DBI::err etc.) and it does not trigger automatic error
1357       detection (RaiseError etc.) so you must check the return status and/or
1358       $h->err to detect errors.
1359
1360       (This method is not directly related to calling stored procedures.
1361       Calling stored procedures is currently not defined by the DBI.  Some
1362       drivers, such as DBD::Oracle, support it in non-portable ways.  See
1363       driver documentation for more details.)
1364
1365       See also install_method() in DBI::DBD for how you can avoid needing to
1366       use func() and gain direct access to driver-private methods.
1367
1368       "can"
1369
1370         $is_implemented = $h->can($method_name);
1371
1372       Returns true if $method_name is implemented by the driver or a default
1373       method is provided by the DBI's driver base class.  It returns false
1374       where a driver hasn't implemented a method and the default method is
1375       provided by the DBI's driver base class is just an empty stub.
1376
1377       "parse_trace_flags"
1378
1379         $trace_settings_integer = $h->parse_trace_flags($trace_settings);
1380
1381       Parses a string containing trace settings and returns the corresponding
1382       integer value used internally by the DBI and drivers.
1383
1384       The $trace_settings argument is a string containing a trace level
1385       between 0 and 15 and/or trace flag names separated by vertical bar
1386       (""|"") or comma ("","") characters. For example: "SQL|3|foo".
1387
1388       It uses the parse_trace_flag() method, described below, to process the
1389       individual trace flag names.
1390
1391       The parse_trace_flags() method was added in DBI 1.42.
1392
1393       "parse_trace_flag"
1394
1395         $bit_flag = $h->parse_trace_flag($trace_flag_name);
1396
1397       Returns the bit flag corresponding to the trace flag name in
1398       $trace_flag_name.  Drivers are expected to override this method and
1399       check if $trace_flag_name is a driver specific trace flags and, if not,
1400       then call the DBI's default parse_trace_flag().
1401
1402       The parse_trace_flag() method was added in DBI 1.42.
1403
1404       "private_attribute_info"
1405
1406         $hash_ref = $h->private_attribute_info();
1407
1408       Returns a reference to a hash whose keys are the names of driver-
1409       private handle attributes available for the kind of handle (driver,
1410       database, statement) that the method was called on.
1411
1412       For example, the return value when called with a DBD::Sybase $dbh could
1413       look like this:
1414
1415         {
1416             syb_dynamic_supported => undef,
1417             syb_oc_version => undef,
1418             syb_server_version => undef,
1419             syb_server_version_string => undef,
1420         }
1421
1422       and when called with a DBD::Sybase $sth they could look like this:
1423
1424         {
1425             syb_types => undef,
1426             syb_proc_status => undef,
1427             syb_result_type => undef,
1428         }
1429
1430       The values should be undef. Meanings may be assigned to particular
1431       values in future.
1432
1433       "swap_inner_handle"
1434
1435         $rc = $h1->swap_inner_handle( $h2 );
1436         $rc = $h1->swap_inner_handle( $h2, $allow_reparent );
1437
1438       Brain transplants for handles. You don't need to know about this unless
1439       you want to become a handle surgeon.
1440
1441       A DBI handle is a reference to a tied hash. A tied hash has an inner
1442       hash that actually holds the contents.  The swap_inner_handle() method
1443       swaps the inner hashes between two handles. The $h1 and $h2 handles
1444       still point to the same tied hashes, but what those hashes are tied to
1445       has been swapped.  In effect $h1 becomes $h2 and vice-versa. This is
1446       powerful stuff, expect problems. Use with care.
1447
1448       As a small safety measure, the two handles, $h1 and $h2, have to share
1449       the same parent unless $allow_reparent is true.
1450
1451       The swap_inner_handle() method was added in DBI 1.44.
1452
1453       Here's a quick kind of 'diagram' as a worked example to help think
1454       about what's happening:
1455
1456           Original state:
1457                   dbh1o -> dbh1i
1458                   sthAo -> sthAi(dbh1i)
1459                   dbh2o -> dbh2i
1460
1461           swap_inner_handle dbh1o with dbh2o:
1462                   dbh2o -> dbh1i
1463                   sthAo -> sthAi(dbh1i)
1464                   dbh1o -> dbh2i
1465
1466           create new sth from dbh1o:
1467                   dbh2o -> dbh1i
1468                   sthAo -> sthAi(dbh1i)
1469                   dbh1o -> dbh2i
1470                   sthBo -> sthBi(dbh2i)
1471
1472           swap_inner_handle sthAo with sthBo:
1473                   dbh2o -> dbh1i
1474                   sthBo -> sthAi(dbh1i)
1475                   dbh1o -> dbh2i
1476                   sthAo -> sthBi(dbh2i)
1477
1478       "visit_child_handles"
1479
1480         $h->visit_child_handles( $coderef );
1481         $h->visit_child_handles( $coderef, $info );
1482
1483       Where $coderef is a reference to a subroutine and $info is an arbitrary
1484       value which, if undefined, defaults to a reference to an empty hash.
1485       Returns $info.
1486
1487       For each child handle of $h, if any, $coderef is invoked as:
1488
1489         $coderef->($child_handle, $info);
1490
1491       If the execution of $coderef returns a true value then
1492       "visit_child_handles" is called on that child handle and passed the
1493       returned value as $info.
1494
1495       For example:
1496
1497         # count database connections with names (DSN) matching a pattern
1498         my $connections = 0;
1499         $dbh->{Driver}->visit_child_handles(sub {
1500             my ($h, $info) = @_;
1501             ++$connections if $h->{Name} =~ /foo/;
1502             return 0; # don't visit kids
1503         })
1504
1505       See also "visit_handles".
1506

ATTRIBUTES COMMON TO ALL HANDLES

1508       These attributes are common to all types of DBI handles.
1509
1510       Some attributes are inherited by child handles. That is, the value of
1511       an inherited attribute in a newly created statement handle is the same
1512       as the value in the parent database handle. Changes to attributes in
1513       the new statement handle do not affect the parent database handle and
1514       changes to the database handle do not affect existing statement
1515       handles, only future ones.
1516
1517       Attempting to set or get the value of an unknown attribute generates a
1518       warning, except for private driver specific attributes (which all have
1519       names starting with a lowercase letter).
1520
1521       Example:
1522
1523         $h->{AttributeName} = ...;    # set/write
1524         ... = $h->{AttributeName};    # get/read
1525
1526       "Warn"
1527
1528       Type: boolean, inherited
1529
1530       The "Warn" attribute enables useful warnings for certain bad practices.
1531       It is enabled by default and should only be disabled in rare
1532       circumstances.  Since warnings are generated using the Perl "warn"
1533       function, they can be intercepted using the Perl $SIG{__WARN__} hook.
1534
1535       The "Warn" attribute is not related to the "PrintWarn" attribute.
1536
1537       "Active"
1538
1539       Type: boolean, read-only
1540
1541       The "Active" attribute is true if the handle object is "active". This
1542       is rarely used in applications. The exact meaning of active is somewhat
1543       vague at the moment. For a database handle it typically means that the
1544       handle is connected to a database ("$dbh->disconnect" sets "Active"
1545       off).  For a statement handle it typically means that the handle is a
1546       "SELECT" that may have more data to fetch. (Fetching all the data or
1547       calling "$sth->finish" sets "Active" off.)
1548
1549       "Executed"
1550
1551       Type: boolean
1552
1553       The "Executed" attribute is true if the handle object has been
1554       "executed".  Currently only the $dbh do() method and the $sth
1555       execute(), execute_array(), and execute_for_fetch() methods set the
1556       "Executed" attribute.
1557
1558       When it's set on a handle it is also set on the parent handle at the
1559       same time. So calling execute() on a $sth also sets the "Executed"
1560       attribute on the parent $dbh.
1561
1562       The "Executed" attribute for a database handle is cleared by the
1563       commit() and rollback() methods (even if they fail). The "Executed"
1564       attribute of a statement handle is not cleared by the DBI under any
1565       circumstances and so acts as a permanent record of whether the
1566       statement handle was ever used.
1567
1568       The "Executed" attribute was added in DBI 1.41.
1569
1570       "Kids"
1571
1572       Type: integer, read-only
1573
1574       For a driver handle, "Kids" is the number of currently existing
1575       database handles that were created from that driver handle.  For a
1576       database handle, "Kids" is the number of currently existing statement
1577       handles that were created from that database handle.  For a statement
1578       handle, the value is zero.
1579
1580       "ActiveKids"
1581
1582       Type: integer, read-only
1583
1584       Like "Kids", but only counting those that are "Active" (as above).
1585
1586       "CachedKids"
1587
1588       Type: hash ref
1589
1590       For a database handle, "CachedKids" returns a reference to the cache
1591       (hash) of statement handles created by the "prepare_cached" method.
1592       For a driver handle, returns a reference to the cache (hash) of
1593       database handles created by the "connect_cached" method.
1594
1595       "Type"
1596
1597       Type: scalar, read-only
1598
1599       The "Type" attribute identifies the type of a DBI handle.  Returns "dr"
1600       for driver handles, "db" for database handles and "st" for statement
1601       handles.
1602
1603       "ChildHandles"
1604
1605       Type: array ref
1606
1607       The ChildHandles attribute contains a reference to an array of all the
1608       handles created by this handle which are still accessible.  The
1609       contents of the array are weak-refs and will become undef when the
1610       handle goes out of scope.
1611
1612       "ChildHandles" returns undef if your perl version does not support weak
1613       references (check the Scalar::Util module).  The referenced array
1614       returned should be treated as read-only.
1615
1616       For example, to enumerate all driver handles, database handles and
1617       statement handles:
1618
1619           sub show_child_handles {
1620               my ($h, $level) = @_;
1621               printf "%sh %s %s\n", $h->{Type}, "\t" x $level, $h;
1622               show_child_handles($_, $level + 1)
1623                   for (grep { defined } @{$h->{ChildHandles}});
1624           }
1625
1626           my %drivers = DBI->installed_drivers();
1627           show_child_handles($_, 0) for (values %drivers);
1628
1629       "CompatMode"
1630
1631       Type: boolean, inherited
1632
1633       The "CompatMode" attribute is used by emulation layers (such as
1634       Oraperl) to enable compatible behaviour in the underlying driver (e.g.,
1635       DBD::Oracle) for this handle. Not normally set by application code.
1636
1637       It also has the effect of disabling the 'quick FETCH' of attribute
1638       values from the handles attribute cache. So all attribute values are
1639       handled by the drivers own FETCH method. This makes them slightly
1640       slower but is useful for special-purpose drivers like DBD::Multiplex.
1641
1642       "InactiveDestroy"
1643
1644       Type: boolean
1645
1646       The default value, false, means a handle will be fully destroyed as
1647       normal when the last reference to it is removed, just as you'd expect.
1648
1649       If set true then the handle will be treated by the DESTROY as if it was
1650       no longer Active, and so the database engine related effects of
1651       DESTROYing a handle will be skipped.  Think of the name as meaning
1652       'treat the handle as not-Active in the DESTROY method'.
1653
1654       For a database handle, this attribute does not disable an explicit call
1655       to the disconnect method, only the implicit call from DESTROY that
1656       happens if the handle is still marked as "Active".
1657
1658       This attribute is specifically designed for use in Unix applications
1659       that "fork" child processes.  For some drivers, when the child process
1660       exits the destruction of inherited handles cause the corresponding
1661       handles in the parent process to cease working.
1662
1663       Either the parent or the child process, but not both, should set
1664       "InactiveDestroy" true on all their shared handles. Alternatively the
1665       "AutoInactiveDestroy" can be set in the parent on connect.
1666
1667       To help tracing applications using fork the process id is shown in the
1668       trace log whenever a DBI or handle trace() method is called.  The
1669       process id also shown for every method call if the DBI trace level (not
1670       handle trace level) is set high enough to show the trace from the DBI's
1671       method dispatcher, e.g. >= 9.
1672
1673       "AutoInactiveDestroy"
1674
1675       Type: boolean, inherited
1676
1677       The "InactiveDestroy" attribute, described above, needs to be
1678       explicitly set in the child process after a fork(). This is a problem
1679       if the code that performs the fork() is not under your control, perhaps
1680       in a third-party module.  Use "AutoInactiveDestroy" to get around this
1681       situation.
1682
1683       If set true, the DESTROY method will check the process id of the handle
1684       and, if different from the current process id, it will set the
1685       InactiveDestroy attribute.
1686
1687       This is the example it's designed to deal with:
1688
1689           my $dbh = DBI->connect(...);
1690           some_code_that_forks(); # Perhaps without your knowledge
1691           # Child process dies, destroying the inherited dbh
1692           $dbh->do(...); # Breaks because parent $dbh is now broken
1693
1694       The "AutoInactiveDestroy" attribute was added in DBI 1.614.
1695
1696       "PrintWarn"
1697
1698       Type: boolean, inherited
1699
1700       The "PrintWarn" attribute controls the printing of warnings recorded by
1701       the driver.  When set to a true value the DBI will check method calls
1702       to see if a warning condition has been set. If so, the DBI will
1703       effectively do a "warn("$class $method warning: $DBI::errstr")" where
1704       $class is the driver class and $method is the name of the method which
1705       failed. E.g.,
1706
1707         DBD::Oracle::db execute warning: ... warning text here ...
1708
1709       By default, "DBI->connect" sets "PrintWarn" "on" if $^W is true, i.e.,
1710       perl is running with warnings enabled.
1711
1712       If desired, the warnings can be caught and processed using a
1713       $SIG{__WARN__} handler or modules like CGI::Carp and CGI::ErrorWrap.
1714
1715       See also "set_err" for how warnings are recorded and "HandleSetErr" for
1716       how to influence it.
1717
1718       Fetching the full details of warnings can require an extra round-trip
1719       to the database server for some drivers. In which case the driver may
1720       opt to only fetch the full details of warnings if the "PrintWarn"
1721       attribute is true. If "PrintWarn" is false then these drivers should
1722       still indicate the fact that there were warnings by setting the warning
1723       string to, for example: "3 warnings".
1724
1725       "PrintError"
1726
1727       Type: boolean, inherited
1728
1729       The "PrintError" attribute can be used to force errors to generate
1730       warnings (using "warn") in addition to returning error codes in the
1731       normal way.  When set "on", any method which results in an error
1732       occurring will cause the DBI to effectively do a "warn("$class $method
1733       failed: $DBI::errstr")" where $class is the driver class and $method is
1734       the name of the method which failed. E.g.,
1735
1736         DBD::Oracle::db prepare failed: ... error text here ...
1737
1738       By default, "DBI->connect" sets "PrintError" "on".
1739
1740       If desired, the warnings can be caught and processed using a
1741       $SIG{__WARN__} handler or modules like CGI::Carp and CGI::ErrorWrap.
1742
1743       "RaiseError"
1744
1745       Type: boolean, inherited
1746
1747       The "RaiseError" attribute can be used to force errors to raise
1748       exceptions rather than simply return error codes in the normal way. It
1749       is "off" by default.  When set "on", any method which results in an
1750       error will cause the DBI to effectively do a "die("$class $method
1751       failed: $DBI::errstr")", where $class is the driver class and $method
1752       is the name of the method that failed. E.g.,
1753
1754         DBD::Oracle::db prepare failed: ... error text here ...
1755
1756       If you turn "RaiseError" on then you'd normally turn "PrintError" off.
1757       If "PrintError" is also on, then the "PrintError" is done first
1758       (naturally).
1759
1760       Typically "RaiseError" is used in conjunction with "eval { ... }" to
1761       catch the exception that's been thrown and followed by an "if ($@) {
1762       ... }" block to handle the caught exception.  For example:
1763
1764         eval {
1765           ...
1766           $sth->execute();
1767           ...
1768         };
1769         if ($@) {
1770           # $sth->err and $DBI::err will be true if error was from DBI
1771           warn $@; # print the error
1772           ... # do whatever you need to deal with the error
1773         }
1774
1775       In that eval block the $DBI::lasth variable can be useful for diagnosis
1776       and reporting if you can't be sure which handle triggered the error.
1777       For example, $DBI::lasth->{Type} and $DBI::lasth->{Statement}.
1778
1779       See also "Transactions".
1780
1781       If you want to temporarily turn "RaiseError" off (inside a library
1782       function that is likely to fail, for example), the recommended way is
1783       like this:
1784
1785         {
1786           local $h->{RaiseError};  # localize and turn off for this block
1787           ...
1788         }
1789
1790       The original value will automatically and reliably be restored by Perl,
1791       regardless of how the block is exited.  The same logic applies to other
1792       attributes, including "PrintError".
1793
1794       "HandleError"
1795
1796       Type: code ref, inherited
1797
1798       The "HandleError" attribute can be used to provide your own alternative
1799       behaviour in case of errors. If set to a reference to a subroutine then
1800       that subroutine is called when an error is detected (at the same point
1801       that "RaiseError" and "PrintError" are handled).
1802
1803       The subroutine is called with three parameters: the error message
1804       string that "RaiseError" and "PrintError" would use, the DBI handle
1805       being used, and the first value being returned by the method that
1806       failed (typically undef).
1807
1808       If the subroutine returns a false value then the "RaiseError" and/or
1809       "PrintError" attributes are checked and acted upon as normal.
1810
1811       For example, to "die" with a full stack trace for any error:
1812
1813         use Carp;
1814         $h->{HandleError} = sub { confess(shift) };
1815
1816       Or to turn errors into exceptions:
1817
1818         use Exception; # or your own favourite exception module
1819         $h->{HandleError} = sub { Exception->new('DBI')->raise($_[0]) };
1820
1821       It is possible to 'stack' multiple HandleError handlers by using
1822       closures:
1823
1824         sub your_subroutine {
1825           my $previous_handler = $h->{HandleError};
1826           $h->{HandleError} = sub {
1827             return 1 if $previous_handler and &$previous_handler(@_);
1828             ... your code here ...
1829           };
1830         }
1831
1832       Using a "my" inside a subroutine to store the previous "HandleError"
1833       value is important.  See perlsub and perlref for more information about
1834       closures.
1835
1836       It is possible for "HandleError" to alter the error message that will
1837       be used by "RaiseError" and "PrintError" if it returns false.  It can
1838       do that by altering the value of $_[0]. This example appends a stack
1839       trace to all errors and, unlike the previous example using
1840       Carp::confess, this will work "PrintError" as well as "RaiseError":
1841
1842         $h->{HandleError} = sub { $_[0]=Carp::longmess($_[0]); 0; };
1843
1844       It is also possible for "HandleError" to hide an error, to a limited
1845       degree, by using "set_err" to reset $DBI::err and $DBI::errstr, and
1846       altering the return value of the failed method. For example:
1847
1848         $h->{HandleError} = sub {
1849           return 0 unless $_[0] =~ /^\S+ fetchrow_arrayref failed:/;
1850           return 0 unless $_[1]->err == 1234; # the error to 'hide'
1851           $h->set_err(undef,undef);   # turn off the error
1852           $_[2] = [ ... ];    # supply alternative return value
1853           return 1;
1854         };
1855
1856       This only works for methods which return a single value and is hard to
1857       make reliable (avoiding infinite loops, for example) and so isn't
1858       recommended for general use!  If you find a good use for it then please
1859       let me know.
1860
1861       "HandleSetErr"
1862
1863       Type: code ref, inherited
1864
1865       The "HandleSetErr" attribute can be used to intercept the setting of
1866       handle "err", "errstr", and "state" values.  If set to a reference to a
1867       subroutine then that subroutine is called whenever set_err() is called,
1868       typically by the driver or a subclass.
1869
1870       The subroutine is called with five arguments, the first five that were
1871       passed to set_err(): the handle, the "err", "errstr", and "state"
1872       values being set, and the method name. These can be altered by changing
1873       the values in the @_ array. The return value affects set_err()
1874       behaviour, see "set_err" for details.
1875
1876       It is possible to 'stack' multiple HandleSetErr handlers by using
1877       closures. See "HandleError" for an example.
1878
1879       The "HandleSetErr" and "HandleError" subroutines differ in subtle but
1880       significant ways. HandleError is only invoked at the point where the
1881       DBI is about to return to the application with "err" set true.  It's
1882       not invoked by the failure of a method that's been called by another
1883       DBI method.  HandleSetErr, on the other hand, is called whenever
1884       set_err() is called with a defined "err" value, even if false.  So it's
1885       not just for errors, despite the name, but also warn and info states.
1886       The set_err() method, and thus HandleSetErr, may be called multiple
1887       times within a method and is usually invoked from deep within driver
1888       code.
1889
1890       In theory a driver can use the return value from HandleSetErr via
1891       set_err() to decide whether to continue or not. If set_err() returns an
1892       empty list, indicating that the HandleSetErr code has 'handled' the
1893       'error', the driver could then continue instead of failing (if that's a
1894       reasonable thing to do).  This isn't excepted to be common and any such
1895       cases should be clearly marked in the driver documentation and
1896       discussed on the dbi-dev mailing list.
1897
1898       The "HandleSetErr" attribute was added in DBI 1.41.
1899
1900       "ErrCount"
1901
1902       Type: unsigned integer
1903
1904       The "ErrCount" attribute is incremented whenever the set_err() method
1905       records an error. It isn't incremented by warnings or information
1906       states. It is not reset by the DBI at any time.
1907
1908       The "ErrCount" attribute was added in DBI 1.41. Older drivers may not
1909       have been updated to use set_err() to record errors and so this
1910       attribute may not be incremented when using them.
1911
1912       "ShowErrorStatement"
1913
1914       Type: boolean, inherited
1915
1916       The "ShowErrorStatement" attribute can be used to cause the relevant
1917       Statement text to be appended to the error messages generated by the
1918       "RaiseError", "PrintError", and "PrintWarn" attributes.  Only applies
1919       to errors on statement handles plus the prepare(), do(), and the
1920       various "select*()" database handle methods.  (The exact format of the
1921       appended text is subject to change.)
1922
1923       If "$h->{ParamValues}" returns a hash reference of parameter
1924       (placeholder) values then those are formatted and appended to the end
1925       of the Statement text in the error message.
1926
1927       "TraceLevel"
1928
1929       Type: integer, inherited
1930
1931       The "TraceLevel" attribute can be used as an alternative to the "trace"
1932       method to set the DBI trace level and trace flags for a specific
1933       handle.  See "TRACING" for more details.
1934
1935       The "TraceLevel" attribute is especially useful combined with "local"
1936       to alter the trace settings for just a single block of code.
1937
1938       "FetchHashKeyName"
1939
1940       Type: string, inherited
1941
1942       The "FetchHashKeyName" attribute is used to specify whether the
1943       fetchrow_hashref() method should perform case conversion on the field
1944       names used for the hash keys. For historical reasons it defaults to
1945       '"NAME"' but it is recommended to set it to '"NAME_lc"' (convert to
1946       lower case) or '"NAME_uc"' (convert to upper case) according to your
1947       preference.  It can only be set for driver and database handles.  For
1948       statement handles the value is frozen when prepare() is called.
1949
1950       "ChopBlanks"
1951
1952       Type: boolean, inherited
1953
1954       The "ChopBlanks" attribute can be used to control the trimming of
1955       trailing space characters from fixed width character (CHAR) fields. No
1956       other field types are affected, even where field values have trailing
1957       spaces.
1958
1959       The default is false (although it is possible that the default may
1960       change).  Applications that need specific behaviour should set the
1961       attribute as needed.
1962
1963       Drivers are not required to support this attribute, but any driver
1964       which does not support it must arrange to return "undef" as the
1965       attribute value.
1966
1967       "LongReadLen"
1968
1969       Type: unsigned integer, inherited
1970
1971       The "LongReadLen" attribute may be used to control the maximum length
1972       of 'long' type fields (LONG, BLOB, CLOB, MEMO, etc.) which the driver
1973       will read from the database automatically when it fetches each row of
1974       data.
1975
1976       The "LongReadLen" attribute only relates to fetching and reading long
1977       values; it is not involved in inserting or updating them.
1978
1979       A value of 0 means not to automatically fetch any long data.  Drivers
1980       may return undef or an empty string for long fields when "LongReadLen"
1981       is 0.
1982
1983       The default is typically 0 (zero) or 80 bytes but may vary between
1984       drivers.  Applications fetching long fields should set this value to
1985       slightly larger than the longest long field value to be fetched.
1986
1987       Some databases return some long types encoded as pairs of hex digits.
1988       For these types, "LongReadLen" relates to the underlying data length
1989       and not the doubled-up length of the encoded string.
1990
1991       Changing the value of "LongReadLen" for a statement handle after it has
1992       been "prepare"'d will typically have no effect, so it's common to set
1993       "LongReadLen" on the $dbh before calling "prepare".
1994
1995       For most drivers the value used here has a direct effect on the memory
1996       used by the statement handle while it's active, so don't be too
1997       generous. If you can't be sure what value to use you could execute an
1998       extra select statement to determine the longest value.  For example:
1999
2000         $dbh->{LongReadLen} = $dbh->selectrow_array(qq{
2001             SELECT MAX(OCTET_LENGTH(long_column_name))
2002             FROM table WHERE ...
2003         });
2004         $sth = $dbh->prepare(qq{
2005             SELECT long_column_name, ... FROM table WHERE ...
2006         });
2007
2008       You may need to take extra care if the table can be modified between
2009       the first select and the second being executed. You may also need to
2010       use a different function if OCTET_LENGTH() does not work for long types
2011       in your database. For example, for Sybase use DATALENGTH() and for
2012       Oracle use LENGTHB().
2013
2014       See also "LongTruncOk" for information on truncation of long types.
2015
2016       "LongTruncOk"
2017
2018       Type: boolean, inherited
2019
2020       The "LongTruncOk" attribute may be used to control the effect of
2021       fetching a long field value which has been truncated (typically because
2022       it's longer than the value of the "LongReadLen" attribute).
2023
2024       By default, "LongTruncOk" is false and so fetching a long value that
2025       needs to be truncated will cause the fetch to fail.  (Applications
2026       should always be sure to check for errors after a fetch loop in case an
2027       error, such as a divide by zero or long field truncation, caused the
2028       fetch to terminate prematurely.)
2029
2030       If a fetch fails due to a long field truncation when "LongTruncOk" is
2031       false, many drivers will allow you to continue fetching further rows.
2032
2033       See also "LongReadLen".
2034
2035       "TaintIn"
2036
2037       Type: boolean, inherited
2038
2039       If the "TaintIn" attribute is set to a true value and Perl is running
2040       in taint mode (e.g., started with the "-T" option), then all the
2041       arguments to most DBI method calls are checked for being tainted. This
2042       may change.
2043
2044       The attribute defaults to off, even if Perl is in taint mode.  See
2045       perlsec for more about taint mode.  If Perl is not running in taint
2046       mode, this attribute has no effect.
2047
2048       When fetching data that you trust you can turn off the TaintIn
2049       attribute, for that statement handle, for the duration of the fetch
2050       loop.
2051
2052       The "TaintIn" attribute was added in DBI 1.31.
2053
2054       "TaintOut"
2055
2056       Type: boolean, inherited
2057
2058       If the "TaintOut" attribute is set to a true value and Perl is running
2059       in taint mode (e.g., started with the "-T" option), then most data
2060       fetched from the database is considered tainted. This may change.
2061
2062       The attribute defaults to off, even if Perl is in taint mode.  See
2063       perlsec for more about taint mode.  If Perl is not running in taint
2064       mode, this attribute has no effect.
2065
2066       When fetching data that you trust you can turn off the TaintOut
2067       attribute, for that statement handle, for the duration of the fetch
2068       loop.
2069
2070       Currently only fetched data is tainted. It is possible that the results
2071       of other DBI method calls, and the value of fetched attributes, may
2072       also be tainted in future versions. That change may well break your
2073       applications unless you take great care now. If you use DBI Taint mode,
2074       please report your experience and any suggestions for changes.
2075
2076       The "TaintOut" attribute was added in DBI 1.31.
2077
2078       "Taint"
2079
2080       Type: boolean, inherited
2081
2082       The "Taint" attribute is a shortcut for "TaintIn" and "TaintOut" (it is
2083       also present for backwards compatibility).
2084
2085       Setting this attribute sets both "TaintIn" and "TaintOut", and
2086       retrieving it returns a true value if and only if "TaintIn" and
2087       "TaintOut" are both set to true values.
2088
2089       "Profile"
2090
2091       Type: inherited
2092
2093       The "Profile" attribute enables the collection and reporting of method
2094       call timing statistics.  See the DBI::Profile module documentation for
2095       much more detail.
2096
2097       The "Profile" attribute was added in DBI 1.24.
2098
2099       "ReadOnly"
2100
2101       Type: boolean, inherited
2102
2103       An application can set the "ReadOnly" attribute of a handle to a true
2104       value to indicate that it will not be attempting to make any changes
2105       using that handle or any children of it.
2106
2107       Note that the exact definition of 'read only' is rather fuzzy.  For
2108       more details see the documentation for the driver you're using.
2109
2110       If the driver can make the handle truly read-only then it should
2111       (unless doing so would have unpleasant side effect, like changing the
2112       consistency level from per-statement to per-session).  Otherwise the
2113       attribute is simply advisory.
2114
2115       A driver can set the "ReadOnly" attribute itself to indicate that the
2116       data it is connected to cannot be changed for some reason.
2117
2118       Library modules and proxy drivers can use the attribute to influence
2119       their behavior.  For example, the DBD::Gofer driver considers the
2120       "ReadOnly" attribute when making a decision about whether to retry an
2121       operation that failed.
2122
2123       The attribute should be set to 1 or 0 (or undef). Other values are
2124       reserved.
2125
2126       "Callbacks"
2127
2128       Type: hash ref
2129
2130       The DBI callback mechanism lets you intercept, and optionally replace,
2131       any method call on a DBI handle. At the extreme, it lets you become a
2132       puppet master, deceiving the application in any way you want.
2133
2134       The "Callbacks" attribute is a hash reference where the keys are DBI
2135       method names and the values are code references. For each key naming a
2136       method, the DBI will execute the associated code reference before
2137       executing the method.
2138
2139       The arguments to the code reference will be the same as to the method,
2140       including the invocant (a database handle or statement handle). For
2141       example, say that to callback to some code on a call to "prepare()":
2142
2143         $dbh->{Callbacks} = {
2144             prepare => sub {
2145                 my ($dbh, $query, $attrs) = @_;
2146                 print "Preparing q{$query}\n"
2147             },
2148         };
2149
2150       The callback would then be executed when you called the "prepare()"
2151       method:
2152
2153         $dbh->prepare('SELECT 1');
2154
2155       And the output of course would be:
2156
2157         Preparing q{SELECT 1}
2158
2159       Because callbacks are executed before the methods they're associated
2160       with, you can modify the arguments before they're passed on to the
2161       method call. For example, to make sure that all calls to "prepare()"
2162       are immediately prepared by DBD::Pg, add a callback that makes sure
2163       that the "pg_prepare_now" attribute is always set:
2164
2165         my $dbh = DBI->connect($dsn, $username, $auth, {
2166             Callbacks => {
2167                 prepare => sub {
2168                     $_[2] ||= {};
2169                     $_[2]->{pg_prepare_now} = 1;
2170                     return; # must return nothing
2171                 },
2172             }
2173         });
2174
2175       Note that we are editing the contents of @_ directly. In this case
2176       we've created the attributes hash if it's not passed to the "prepare"
2177       call.
2178
2179       You can also prevent the associated method from ever executing. While a
2180       callback executes, $_ holds the method name. (This allows multiple
2181       callbacks to share the same code reference and still know what method
2182       was called.)  To prevent the method from executing, simply "undef $_".
2183       For example, if you wanted to disable calls to "ping()", you could do
2184       this:
2185
2186         $dbh->{Callbacks} = {
2187             ping => sub {
2188                 # tell dispatch to not call the method:
2189                 undef $_;
2190                 # return this value instead:
2191                 return "42 bells";
2192             }
2193         };
2194
2195       As with other attributes, Callbacks can be specified on a handle or via
2196       the attributes to "connect()". Callbacks can also be applied to a
2197       statement methods on a statement handle. For example:
2198
2199         $sth->{Callbacks} = {
2200             execute => sub {
2201                 print "Executing ", shift->{Statement}, "\n";
2202             }
2203         };
2204
2205       The "Callbacks" attribute of a database handle isn't copied to any
2206       statement handles it creates. So setting callbacks for a statement
2207       handle requires you to set the "Callbacks" attribute on the statement
2208       handle yourself, as in the example above, or use the special
2209       "ChildCallbacks" key described below.
2210
2211       Special Keys in Callbacks Attribute
2212
2213       In addition to DBI handle method names, the "Callbacks" hash reference
2214       supports three additional keys.
2215
2216       The first is the "ChildCallbacks" key. When a statement handle is
2217       created from a database handle the "ChildCallbacks" key of the database
2218       handle's "Callbacks" attribute, if any, becomes the new "Callbacks"
2219       attribute of the statement handle.  This allows you to define callbacks
2220       for all statement handles created from a database handle. For example,
2221       if you wanted to count how many times "execute" was called in your
2222       application, you could write:
2223
2224         my $exec_count = 0;
2225         my $dbh = DBI->connect( $dsn, $username, $auth, {
2226             Callbacks => {
2227                 ChildCallbacks => {
2228                     execute => sub { $exec_count++; return; }
2229                 }
2230             }
2231         });
2232
2233         END {
2234             print "The execute method was called $exec_count times\n";
2235         }
2236
2237       The other two special keys are "connect_cached.new" and
2238       "connect_cached.reused". These keys define callbacks that are called
2239       when "connect_cached()" is called, but allow different behaviors
2240       depending on whether a new handle is created or a handle is returned.
2241       The callback is invoked with these arguments: "$dbh, $dsn, $user,
2242       $auth, $attr".
2243
2244       For example, some applications uses "connect_cached()" to connect with
2245       "AutoCommit" enabled and then disable "AutoCommit" temporarily for
2246       transactions. If "connect_cached()" is called during a transaction,
2247       perhaps in a utility method, then it might select the same cached
2248       handle and then force "AutoCommit" on, forcing a commit of the
2249       transaction. See the "connect_cached" documentation for one way to deal
2250       with that. Here we'll describe an alternative approach using a
2251       callback.
2252
2253       Because the "connect_cached.*" callbacks are invoked before
2254       connect_cached() has applied the connect attributes you can use a
2255       callback to edit the attributes that will be applied.  To prevent a
2256       cached handle from having its transactions committed before it's
2257       returned, you can eliminate the "AutoCommit" attribute in a
2258       "connect_cached.reused" callback, like so:
2259
2260         my $cb = {
2261             'connect_cached.reused' => sub { delete $_[4]->{AutoCommit} },
2262         };
2263
2264         sub dbh {
2265             my $self = shift;
2266             DBI->connect_cached( $dsn, $username, $auth, {
2267                 PrintError => 0,
2268                 RaiseError => 1,
2269                 AutoCommit => 1,
2270                 Callbacks  => $cb,
2271             });
2272         }
2273
2274       The upshot is that new database handles are created with "AutoCommit"
2275       enabled, while cached database handles are left in whatever transaction
2276       state they happened to be in when retrieved from the cache.
2277
2278       A more common application for callbacks is setting connection state
2279       only when a new connection is made (by connect() or connect_cached()).
2280       Adding a callback to the connected method makes this easy.  This method
2281       is a no-op by default (unless you subclass the DBI and change it).  The
2282       DBI calls it to indicate that a new connection has been made and the
2283       connection attributes have all been set.  You can give it a bit of
2284       added functionality by applying a callback to it. For example, to make
2285       sure that MySQL understands your application's ANSI-compliant SQL, set
2286       it up like so:
2287
2288         my $dbh = DBI->connect($dsn, $username, $auth, {
2289             Callbacks => {
2290                 connected => sub {
2291                     shift->do(q{
2292                         SET SESSION sql_mode='ansi,strict_trans_tables,no_auto_value_on_zero';
2293                     });
2294                     return;
2295                 },
2296             }
2297         });
2298
2299       One significant limitation with callbacks is that there can only be one
2300       per method per handle. This means it's easy for one use of callbacks to
2301       interfere with, or typically simply overwrite, another use of
2302       callbacks. For this reason modules using callbacks should document the
2303       fact clearly so application authors can tell if use of callbacks by the
2304       module will clash with use of callbacks by the application.
2305
2306       You might be able to work around this issue by taking a copy of the
2307       original callback and calling it within your own. For example:
2308
2309         my $prev_cb = $h->{Callbacks}{method_name};
2310         $h->{Callbacks}{method_name} = sub {
2311           if ($prev_cb) {
2312               my @result = $prev_cb->(@_);
2313               return @result if not $_; # $prev_cb vetoed call
2314           }
2315           ... your callback logic here ...
2316         };
2317
2318       "private_your_module_name_*"
2319
2320       The DBI provides a way to store extra information in a DBI handle as
2321       "private" attributes. The DBI will allow you to store and retrieve any
2322       attribute which has a name starting with ""private_"".
2323
2324       It is strongly recommended that you use just one private attribute
2325       (e.g., use a hash ref) and give it a long and unambiguous name that
2326       includes the module or application name that the attribute relates to
2327       (e.g., ""private_YourFullModuleName_thingy"").
2328
2329       Because of the way the Perl tie mechanism works you cannot reliably use
2330       the "||=" operator directly to initialise the attribute, like this:
2331
2332         my $foo = $dbh->{private_yourmodname_foo} ||= { ... }; # WRONG
2333
2334       you should use a two step approach like this:
2335
2336         my $foo = $dbh->{private_yourmodname_foo};
2337         $foo ||= $dbh->{private_yourmodname_foo} = { ... };
2338
2339       This attribute is primarily of interest to people sub-classing DBI, or
2340       for applications to piggy-back extra information onto DBI handles.
2341

DBI DATABASE HANDLE OBJECTS

2343       This section covers the methods and attributes associated with database
2344       handles.
2345
2346   Database Handle Methods
2347       The following methods are specified for DBI database handles:
2348
2349       "clone"
2350
2351         $new_dbh = $dbh->clone(\%attr);
2352
2353       The "clone" method duplicates the $dbh connection by connecting with
2354       the same parameters ($dsn, $user, $password) as originally used.
2355
2356       The attributes for the cloned connect are the same as those used for
2357       the original connect, with any other attributes in "\%attr" merged over
2358       them.  Effectively the same as doing:
2359
2360         %attributes_used = ( %original_attributes, %attr );
2361
2362       If \%attr is not given then it defaults to a hash containing all the
2363       attributes in the attribute cache of $dbh excluding any non-code
2364       references, plus the main boolean attributes (RaiseError, PrintError,
2365       AutoCommit, etc.). This behaviour is unreliable and so use of clone
2366       without an argument is deprecated and may cause a warning in a future
2367       release.
2368
2369       The clone method can be used even if the database handle is
2370       disconnected.
2371
2372       The "clone" method was added in DBI 1.33.
2373
2374       "data_sources"
2375
2376         @ary = $dbh->data_sources();
2377         @ary = $dbh->data_sources(\%attr);
2378
2379       Returns a list of data sources (databases) available via the $dbh
2380       driver's data_sources() method, plus any extra data sources that the
2381       driver can discover via the connected $dbh. Typically the extra data
2382       sources are other databases managed by the same server process that the
2383       $dbh is connected to.
2384
2385       Data sources are returned in a form suitable for passing to the
2386       "connect" method (that is, they will include the ""dbi:$driver:""
2387       prefix).
2388
2389       The data_sources() method, for a $dbh, was added in DBI 1.38.
2390
2391       "do"
2392
2393         $rows = $dbh->do($statement)           or die $dbh->errstr;
2394         $rows = $dbh->do($statement, \%attr)   or die $dbh->errstr;
2395         $rows = $dbh->do($statement, \%attr, @bind_values) or die ...
2396
2397       Prepare and execute a single statement. Returns the number of rows
2398       affected or "undef" on error. A return value of "-1" means the number
2399       of rows is not known, not applicable, or not available.
2400
2401       This method is typically most useful for non-"SELECT" statements that
2402       either cannot be prepared in advance (due to a limitation of the
2403       driver) or do not need to be executed repeatedly. It should not be used
2404       for "SELECT" statements because it does not return a statement handle
2405       (so you can't fetch any data).
2406
2407       The default "do" method is logically similar to:
2408
2409         sub do {
2410             my($dbh, $statement, $attr, @bind_values) = @_;
2411             my $sth = $dbh->prepare($statement, $attr) or return undef;
2412             $sth->execute(@bind_values) or return undef;
2413             my $rows = $sth->rows;
2414             ($rows == 0) ? "0E0" : $rows; # always return true if no error
2415         }
2416
2417       For example:
2418
2419         my $rows_deleted = $dbh->do(q{
2420             DELETE FROM table
2421             WHERE status = ?
2422         }, undef, 'DONE') or die $dbh->errstr;
2423
2424       Using placeholders and @bind_values with the "do" method can be useful
2425       because it avoids the need to correctly quote any variables in the
2426       $statement. But if you'll be executing the statement many times then
2427       it's more efficient to "prepare" it once and call "execute" many times
2428       instead.
2429
2430       The "q{...}" style quoting used in this example avoids clashing with
2431       quotes that may be used in the SQL statement. Use the double-quote-like
2432       "qq{...}" operator if you want to interpolate variables into the
2433       string.  See "Quote and Quote-like Operators" in perlop for more
2434       details.
2435
2436       Note drivers are free to avoid the overhead of creating an DBI
2437       statement handle for do(), especially if there are no parameters. In
2438       this case error handlers, if invoked during do(), will be passed the
2439       database handle.
2440
2441       "last_insert_id"
2442
2443         $rv = $dbh->last_insert_id($catalog, $schema, $table, $field);
2444         $rv = $dbh->last_insert_id($catalog, $schema, $table, $field, \%attr);
2445
2446       Returns a value 'identifying' the row just inserted, if possible.
2447       Typically this would be a value assigned by the database server to a
2448       column with an auto_increment or serial type.  Returns undef if the
2449       driver does not support the method or can't determine the value.
2450
2451       The $catalog, $schema, $table, and $field parameters may be required
2452       for some drivers (see below).  If you don't know the parameter values
2453       and your driver does not need them, then use "undef" for each.
2454
2455       There are several caveats to be aware of with this method if you want
2456       to use it for portable applications:
2457
2458       * For some drivers the value may only available immediately after the
2459       insert statement has executed (e.g., mysql, Informix).
2460
2461       * For some drivers the $catalog, $schema, $table, and $field parameters
2462       are required, for others they are ignored (e.g., mysql).
2463
2464       * Drivers may return an indeterminate value if no insert has been
2465       performed yet.
2466
2467       * For some drivers the value may only be available if placeholders have
2468       not been used (e.g., Sybase, MS SQL). In this case the value returned
2469       would be from the last non-placeholder insert statement.
2470
2471       * Some drivers may need driver-specific hints about how to get the
2472       value. For example, being told the name of the database 'sequence'
2473       object that holds the value. Any such hints are passed as driver-
2474       specific attributes in the \%attr parameter.
2475
2476       * If the underlying database offers nothing better, then some drivers
2477       may attempt to implement this method by executing ""select max($field)
2478       from $table"". Drivers using any approach like this should issue a
2479       warning if "AutoCommit" is true because it is generally unsafe -
2480       another process may have modified the table between your insert and the
2481       select. For situations where you know it is safe, such as when you have
2482       locked the table, you can silence the warning by passing "Warn" => 0 in
2483       \%attr.
2484
2485       * If no insert has been performed yet, or the last insert failed, then
2486       the value is implementation defined.
2487
2488       Given all the caveats above, it's clear that this method must be used
2489       with care.
2490
2491       The "last_insert_id" method was added in DBI 1.38.
2492
2493       "selectrow_array"
2494
2495         @row_ary = $dbh->selectrow_array($statement);
2496         @row_ary = $dbh->selectrow_array($statement, \%attr);
2497         @row_ary = $dbh->selectrow_array($statement, \%attr, @bind_values);
2498
2499       This utility method combines "prepare", "execute" and "fetchrow_array"
2500       into a single call. If called in a list context, it returns the first
2501       row of data from the statement.  The $statement parameter can be a
2502       previously prepared statement handle, in which case the "prepare" is
2503       skipped.
2504
2505       If any method fails, and "RaiseError" is not set, "selectrow_array"
2506       will return an empty list.
2507
2508       If called in a scalar context for a statement handle that has more than
2509       one column, it is undefined whether the driver will return the value of
2510       the first column or the last. So don't do that.  Also, in a scalar
2511       context, an "undef" is returned if there are no more rows or if an
2512       error occurred. That "undef" can't be distinguished from an "undef"
2513       returned because the first field value was NULL.  For these reasons you
2514       should exercise some caution if you use "selectrow_array" in a scalar
2515       context, or just don't do that.
2516
2517       "selectrow_arrayref"
2518
2519         $ary_ref = $dbh->selectrow_arrayref($statement);
2520         $ary_ref = $dbh->selectrow_arrayref($statement, \%attr);
2521         $ary_ref = $dbh->selectrow_arrayref($statement, \%attr, @bind_values);
2522
2523       This utility method combines "prepare", "execute" and
2524       "fetchrow_arrayref" into a single call. It returns the first row of
2525       data from the statement.  The $statement parameter can be a previously
2526       prepared statement handle, in which case the "prepare" is skipped.
2527
2528       If any method fails, and "RaiseError" is not set, "selectrow_array"
2529       will return undef.
2530
2531       "selectrow_hashref"
2532
2533         $hash_ref = $dbh->selectrow_hashref($statement);
2534         $hash_ref = $dbh->selectrow_hashref($statement, \%attr);
2535         $hash_ref = $dbh->selectrow_hashref($statement, \%attr, @bind_values);
2536
2537       This utility method combines "prepare", "execute" and
2538       "fetchrow_hashref" into a single call. It returns the first row of data
2539       from the statement.  The $statement parameter can be a previously
2540       prepared statement handle, in which case the "prepare" is skipped.
2541
2542       If any method fails, and "RaiseError" is not set, "selectrow_hashref"
2543       will return undef.
2544
2545       "selectall_arrayref"
2546
2547         $ary_ref = $dbh->selectall_arrayref($statement);
2548         $ary_ref = $dbh->selectall_arrayref($statement, \%attr);
2549         $ary_ref = $dbh->selectall_arrayref($statement, \%attr, @bind_values);
2550
2551       This utility method combines "prepare", "execute" and
2552       "fetchall_arrayref" into a single call. It returns a reference to an
2553       array containing a reference to an array (or hash, see below) for each
2554       row of data fetched.
2555
2556       The $statement parameter can be a previously prepared statement handle,
2557       in which case the "prepare" is skipped. This is recommended if the
2558       statement is going to be executed many times.
2559
2560       If "RaiseError" is not set and any method except "fetchall_arrayref"
2561       fails then "selectall_arrayref" will return "undef"; if
2562       "fetchall_arrayref" fails then it will return with whatever data has
2563       been fetched thus far. You should check "$sth->err" afterwards (or use
2564       the "RaiseError" attribute) to discover if the data is complete or was
2565       truncated due to an error.
2566
2567       The "fetchall_arrayref" method called by "selectall_arrayref" supports
2568       a $max_rows parameter. You can specify a value for $max_rows by
2569       including a '"MaxRows"' attribute in \%attr. In which case finish() is
2570       called for you after fetchall_arrayref() returns.
2571
2572       The "fetchall_arrayref" method called by "selectall_arrayref" also
2573       supports a $slice parameter. You can specify a value for $slice by
2574       including a '"Slice"' or '"Columns"' attribute in \%attr. The only
2575       difference between the two is that if "Slice" is not defined and
2576       "Columns" is an array ref, then the array is assumed to contain column
2577       index values (which count from 1), rather than perl array index values.
2578       In which case the array is copied and each value decremented before
2579       passing to "/fetchall_arrayref".
2580
2581       You may often want to fetch an array of rows where each row is stored
2582       as a hash. That can be done simple using:
2583
2584         my $emps = $dbh->selectall_arrayref(
2585             "SELECT ename FROM emp ORDER BY ename",
2586             { Slice => {} }
2587         );
2588         foreach my $emp ( @$emps ) {
2589             print "Employee: $emp->{ename}\n";
2590         }
2591
2592       Or, to fetch into an array instead of an array ref:
2593
2594         @result = @{ $dbh->selectall_arrayref($sql, { Slice => {} }) };
2595
2596       See "fetchall_arrayref" method for more details.
2597
2598       "selectall_hashref"
2599
2600         $hash_ref = $dbh->selectall_hashref($statement, $key_field);
2601         $hash_ref = $dbh->selectall_hashref($statement, $key_field, \%attr);
2602         $hash_ref = $dbh->selectall_hashref($statement, $key_field, \%attr, @bind_values);
2603
2604       This utility method combines "prepare", "execute" and
2605       "fetchall_hashref" into a single call. It returns a reference to a hash
2606       containing one entry, at most, for each row, as returned by
2607       fetchall_hashref().
2608
2609       The $statement parameter can be a previously prepared statement handle,
2610       in which case the "prepare" is skipped.  This is recommended if the
2611       statement is going to be executed many times.
2612
2613       The $key_field parameter defines which column, or columns, are used as
2614       keys in the returned hash. It can either be the name of a single field,
2615       or a reference to an array containing multiple field names. Using
2616       multiple names yields a tree of nested hashes.
2617
2618       If a row has the same key as an earlier row then it replaces the
2619       earlier row.
2620
2621       If any method except "fetchrow_hashref" fails, and "RaiseError" is not
2622       set, "selectall_hashref" will return "undef".  If "fetchrow_hashref"
2623       fails and "RaiseError" is not set, then it will return with whatever
2624       data it has fetched thus far. $DBI::err should be checked to catch
2625       that.
2626
2627       See fetchall_hashref() for more details.
2628
2629       "selectcol_arrayref"
2630
2631         $ary_ref = $dbh->selectcol_arrayref($statement);
2632         $ary_ref = $dbh->selectcol_arrayref($statement, \%attr);
2633         $ary_ref = $dbh->selectcol_arrayref($statement, \%attr, @bind_values);
2634
2635       This utility method combines "prepare", "execute", and fetching one
2636       column from all the rows, into a single call. It returns a reference to
2637       an array containing the values of the first column from each row.
2638
2639       The $statement parameter can be a previously prepared statement handle,
2640       in which case the "prepare" is skipped. This is recommended if the
2641       statement is going to be executed many times.
2642
2643       If any method except "fetch" fails, and "RaiseError" is not set,
2644       "selectcol_arrayref" will return "undef".  If "fetch" fails and
2645       "RaiseError" is not set, then it will return with whatever data it has
2646       fetched thus far. $DBI::err should be checked to catch that.
2647
2648       The "selectcol_arrayref" method defaults to pushing a single column
2649       value (the first) from each row into the result array. However, it can
2650       also push another column, or even multiple columns per row, into the
2651       result array. This behaviour can be specified via a '"Columns"'
2652       attribute which must be a ref to an array containing the column number
2653       or numbers to use. For example:
2654
2655         # get array of id and name pairs:
2656         my $ary_ref = $dbh->selectcol_arrayref("select id, name from table", { Columns=>[1,2] });
2657         my %hash = @$ary_ref; # build hash from key-value pairs so $hash{$id} => name
2658
2659       You can specify a maximum number of rows to fetch by including a
2660       '"MaxRows"' attribute in \%attr.
2661
2662       "prepare"
2663
2664         $sth = $dbh->prepare($statement)          or die $dbh->errstr;
2665         $sth = $dbh->prepare($statement, \%attr)  or die $dbh->errstr;
2666
2667       Prepares a statement for later execution by the database engine and
2668       returns a reference to a statement handle object.
2669
2670       The returned statement handle can be used to get attributes of the
2671       statement and invoke the "execute" method. See "Statement Handle
2672       Methods".
2673
2674       Drivers for engines without the concept of preparing a statement will
2675       typically just store the statement in the returned handle and process
2676       it when "$sth->execute" is called. Such drivers are unlikely to give
2677       much useful information about the statement, such as
2678       "$sth->{NUM_OF_FIELDS}", until after "$sth->execute" has been called.
2679       Portable applications should take this into account.
2680
2681       In general, DBI drivers do not parse the contents of the statement
2682       (other than simply counting any "Placeholders"). The statement is
2683       passed directly to the database engine, sometimes known as pass-thru
2684       mode. This has advantages and disadvantages. On the plus side, you can
2685       access all the functionality of the engine being used. On the downside,
2686       you're limited if you're using a simple engine, and you need to take
2687       extra care if writing applications intended to be portable between
2688       engines.
2689
2690       Portable applications should not assume that a new statement can be
2691       prepared and/or executed while still fetching results from a previous
2692       statement.
2693
2694       Some command-line SQL tools use statement terminators, like a
2695       semicolon, to indicate the end of a statement. Such terminators should
2696       not normally be used with the DBI.
2697
2698       "prepare_cached"
2699
2700         $sth = $dbh->prepare_cached($statement)
2701         $sth = $dbh->prepare_cached($statement, \%attr)
2702         $sth = $dbh->prepare_cached($statement, \%attr, $if_active)
2703
2704       Like "prepare" except that the statement handle returned will be stored
2705       in a hash associated with the $dbh. If another call is made to
2706       "prepare_cached" with the same $statement and %attr parameter values,
2707       then the corresponding cached $sth will be returned without contacting
2708       the database server.
2709
2710       The $if_active parameter lets you adjust the behaviour if an already
2711       cached statement handle is still Active.  There are several
2712       alternatives:
2713
2714       0: A warning will be generated, and finish() will be called on the
2715       statement handle before it is returned.  This is the default behaviour
2716       if $if_active is not passed.
2717       1: finish() will be called on the statement handle, but the warning is
2718       suppressed.
2719       2: Disables any checking.
2720       3: The existing active statement handle will be removed from the cache
2721       and a new statement handle prepared and cached in its place. This is
2722       the safest option because it doesn't affect the state of the old
2723       handle, it just removes it from the cache. [Added in DBI 1.40]
2724
2725       Here are some examples of "prepare_cached":
2726
2727         sub insert_hash {
2728           my ($table, $field_values) = @_;
2729           # sort to keep field order, and thus sql, stable for prepare_cached
2730           my @fields = sort keys %$field_values;
2731           my @values = @{$field_values}{@fields};
2732           my $sql = sprintf "insert into %s (%s) values (%s)",
2733               $table, join(",", @fields), join(",", ("?")x@fields);
2734           my $sth = $dbh->prepare_cached($sql);
2735           return $sth->execute(@values);
2736         }
2737
2738         sub search_hash {
2739           my ($table, $field_values) = @_;
2740           # sort to keep field order, and thus sql, stable for prepare_cached
2741           my @fields = sort keys %$field_values;
2742           my @values = @{$field_values}{@fields};
2743           my $qualifier = "";
2744           $qualifier = "where ".join(" and ", map { "$_=?" } @fields) if @fields;
2745           $sth = $dbh->prepare_cached("SELECT * FROM $table $qualifier");
2746           return $dbh->selectall_arrayref($sth, {}, @values);
2747         }
2748
2749       Caveat emptor: This caching can be useful in some applications, but it
2750       can also cause problems and should be used with care. Here is a
2751       contrived case where caching would cause a significant problem:
2752
2753         my $sth = $dbh->prepare_cached('SELECT * FROM foo WHERE bar=?');
2754         $sth->execute(...);
2755         while (my $data = $sth->fetchrow_hashref) {
2756
2757           # later, in some other code called within the loop...
2758           my $sth2 = $dbh->prepare_cached('SELECT * FROM foo WHERE bar=?');
2759           $sth2->execute(...);
2760           while (my $data2 = $sth2->fetchrow_arrayref) {
2761             do_stuff(...);
2762           }
2763         }
2764
2765       In this example, since both handles are preparing the exact same
2766       statement, $sth2 will not be its own statement handle, but a duplicate
2767       of $sth returned from the cache. The results will certainly not be what
2768       you expect.  Typically the inner fetch loop will work normally,
2769       fetching all the records and terminating when there are no more, but
2770       now that $sth is the same as $sth2 the outer fetch loop will also
2771       terminate.
2772
2773       You'll know if you run into this problem because prepare_cached() will
2774       generate a warning by default (when $if_active is false).
2775
2776       The cache used by prepare_cached() is keyed by both the statement and
2777       any attributes so you can also avoid this issue by doing something
2778       like:
2779
2780         $sth = $dbh->prepare_cached("...", { dbi_dummy => __FILE__.__LINE__ });
2781
2782       which will ensure that prepare_cached only returns statements cached by
2783       that line of code in that source file.
2784
2785       If you'd like the cache to managed intelligently, you can tie the
2786       hashref returned by "CachedKids" to an appropriate caching module, such
2787       as Tie::Cache::LRU:
2788
2789         my $cache;
2790         tie %$cache, 'Tie::Cache::LRU', 500;
2791         $dbh->{CachedKids} = $cache;
2792
2793       "commit"
2794
2795         $rc  = $dbh->commit     or die $dbh->errstr;
2796
2797       Commit (make permanent) the most recent series of database changes if
2798       the database supports transactions and AutoCommit is off.
2799
2800       If "AutoCommit" is on, then calling "commit" will issue a "commit
2801       ineffective with AutoCommit" warning.
2802
2803       See also "Transactions" in the "FURTHER INFORMATION" section below.
2804
2805       "rollback"
2806
2807         $rc  = $dbh->rollback   or die $dbh->errstr;
2808
2809       Rollback (undo) the most recent series of uncommitted database changes
2810       if the database supports transactions and AutoCommit is off.
2811
2812       If "AutoCommit" is on, then calling "rollback" will issue a "rollback
2813       ineffective with AutoCommit" warning.
2814
2815       See also "Transactions" in the "FURTHER INFORMATION" section below.
2816
2817       "begin_work"
2818
2819         $rc  = $dbh->begin_work   or die $dbh->errstr;
2820
2821       Enable transactions (by turning "AutoCommit" off) until the next call
2822       to "commit" or "rollback". After the next "commit" or "rollback",
2823       "AutoCommit" will automatically be turned on again.
2824
2825       If "AutoCommit" is already off when "begin_work" is called then it does
2826       nothing except return an error. If the driver does not support
2827       transactions then when "begin_work" attempts to set "AutoCommit" off
2828       the driver will trigger a fatal error.
2829
2830       See also "Transactions" in the "FURTHER INFORMATION" section below.
2831
2832       "disconnect"
2833
2834         $rc = $dbh->disconnect  or warn $dbh->errstr;
2835
2836       Disconnects the database from the database handle. "disconnect" is
2837       typically only used before exiting the program. The handle is of little
2838       use after disconnecting.
2839
2840       The transaction behaviour of the "disconnect" method is, sadly,
2841       undefined.  Some database systems (such as Oracle and Ingres) will
2842       automatically commit any outstanding changes, but others (such as
2843       Informix) will rollback any outstanding changes.  Applications not
2844       using "AutoCommit" should explicitly call "commit" or "rollback" before
2845       calling "disconnect".
2846
2847       The database is automatically disconnected by the "DESTROY" method if
2848       still connected when there are no longer any references to the handle.
2849       The "DESTROY" method for each driver should implicitly call "rollback"
2850       to undo any uncommitted changes. This is vital behaviour to ensure that
2851       incomplete transactions don't get committed simply because Perl calls
2852       "DESTROY" on every object before exiting. Also, do not rely on the
2853       order of object destruction during "global destruction", as it is
2854       undefined.
2855
2856       Generally, if you want your changes to be committed or rolled back when
2857       you disconnect, then you should explicitly call "commit" or "rollback"
2858       before disconnecting.
2859
2860       If you disconnect from a database while you still have active statement
2861       handles (e.g., SELECT statement handles that may have more data to
2862       fetch), you will get a warning. The warning may indicate that a fetch
2863       loop terminated early, perhaps due to an uncaught error.  To avoid the
2864       warning call the "finish" method on the active handles.
2865
2866       "ping"
2867
2868         $rc = $dbh->ping;
2869
2870       Attempts to determine, in a reasonably efficient way, if the database
2871       server is still running and the connection to it is still working.
2872       Individual drivers should implement this function in the most suitable
2873       manner for their database engine.
2874
2875       The current default implementation always returns true without actually
2876       doing anything. Actually, it returns ""0 but true"" which is true but
2877       zero. That way you can tell if the return value is genuine or just the
2878       default. Drivers should override this method with one that does the
2879       right thing for their type of database.
2880
2881       Few applications would have direct use for this method. See the
2882       specialized Apache::DBI module for one example usage.
2883
2884       "get_info"
2885
2886         $value = $dbh->get_info( $info_type );
2887
2888       Returns information about the implementation, i.e. driver and data
2889       source capabilities, restrictions etc. It returns "undef" for unknown
2890       or unimplemented information types. For example:
2891
2892         $database_version  = $dbh->get_info(  18 ); # SQL_DBMS_VER
2893         $max_select_tables = $dbh->get_info( 106 ); # SQL_MAXIMUM_TABLES_IN_SELECT
2894
2895       See "Standards Reference Information" for more detailed information
2896       about the information types and their meanings and possible return
2897       values.
2898
2899       The DBI::Const::GetInfoType module exports a %GetInfoType hash that can
2900       be used to map info type names to numbers. For example:
2901
2902         $database_version = $dbh->get_info( $GetInfoType{SQL_DBMS_VER} );
2903
2904       The names are a merging of the ANSI and ODBC standards (which differ in
2905       some cases). See DBI::Const::GetInfoType for more details.
2906
2907       Because some DBI methods make use of get_info(), drivers are strongly
2908       encouraged to support at least the following very minimal set of
2909       information types to ensure the DBI itself works properly:
2910
2911        Type  Name                        Example A     Example B
2912        ----  --------------------------  ------------  ----------------
2913          17  SQL_DBMS_NAME               'ACCESS'      'Oracle'
2914          18  SQL_DBMS_VER                '03.50.0000'  '08.01.0721 ...'
2915          29  SQL_IDENTIFIER_QUOTE_CHAR   '`'           '"'
2916          41  SQL_CATALOG_NAME_SEPARATOR  '.'           '@'
2917         114  SQL_CATALOG_LOCATION        1             2
2918
2919       "table_info"
2920
2921         $sth = $dbh->table_info( $catalog, $schema, $table, $type );
2922         $sth = $dbh->table_info( $catalog, $schema, $table, $type, \%attr );
2923
2924         # then $sth->fetchall_arrayref or $sth->fetchall_hashref etc
2925
2926       Returns an active statement handle that can be used to fetch
2927       information about tables and views that exist in the database.
2928
2929       The arguments $catalog, $schema and $table may accept search patterns
2930       according to the database/driver, for example: $table = '%FOO%';
2931       Remember that the underscore character ('"_"') is a search pattern that
2932       means match any character, so 'FOO_%' is the same as 'FOO%' and
2933       'FOO_BAR%' will match names like 'FOO1BAR'.
2934
2935       The value of $type is a comma-separated list of one or more types of
2936       tables to be returned in the result set. Each value may optionally be
2937       quoted, e.g.:
2938
2939         $type = "TABLE";
2940         $type = "'TABLE','VIEW'";
2941
2942       In addition the following special cases may also be supported by some
2943       drivers:
2944
2945       ·   If the value of $catalog is '%' and $schema and $table name are
2946           empty strings, the result set contains a list of catalog names.
2947           For example:
2948
2949             $sth = $dbh->table_info('%', '', '');
2950
2951       ·   If the value of $schema is '%' and $catalog and $table are empty
2952           strings, the result set contains a list of schema names.
2953
2954       ·   If the value of $type is '%' and $catalog, $schema, and $table are
2955           all empty strings, the result set contains a list of table types.
2956
2957       If your driver doesn't support one or more of the selection filter
2958       parameters then you may get back more than you asked for and can do the
2959       filtering yourself.
2960
2961       This method can be expensive, and can return a large amount of data.
2962       (For example, small Oracle installation returns over 2000 rows.)  So
2963       it's a good idea to use the filters to limit the data as much as
2964       possible.
2965
2966       The statement handle returned has at least the following fields in the
2967       order show below. Other fields, after these, may also be present.
2968
2969       TABLE_CAT: Table catalog identifier. This field is NULL ("undef") if
2970       not applicable to the data source, which is usually the case. This
2971       field is empty if not applicable to the table.
2972
2973       TABLE_SCHEM: The name of the schema containing the TABLE_NAME value.
2974       This field is NULL ("undef") if not applicable to data source, and
2975       empty if not applicable to the table.
2976
2977       TABLE_NAME: Name of the table (or view, synonym, etc).
2978
2979       TABLE_TYPE: One of the following: "TABLE", "VIEW", "SYSTEM TABLE",
2980       "GLOBAL TEMPORARY", "LOCAL TEMPORARY", "ALIAS", "SYNONYM" or a type
2981       identifier that is specific to the data source.
2982
2983       REMARKS: A description of the table. May be NULL ("undef").
2984
2985       Note that "table_info" might not return records for all tables.
2986       Applications can use any valid table regardless of whether it's
2987       returned by "table_info".
2988
2989       See also "tables", "Catalog Methods" and "Standards Reference
2990       Information".
2991
2992       "column_info"
2993
2994         $sth = $dbh->column_info( $catalog, $schema, $table, $column );
2995
2996         # then $sth->fetchall_arrayref or $sth->fetchall_hashref etc
2997
2998       Returns an active statement handle that can be used to fetch
2999       information about columns in specified tables.
3000
3001       The arguments $schema, $table and $column may accept search patterns
3002       according to the database/driver, for example: $table = '%FOO%';
3003
3004       Note: The support for the selection criteria is driver specific. If the
3005       driver doesn't support one or more of them then you may get back more
3006       than you asked for and can do the filtering yourself.
3007
3008       Note: If your driver does not support column_info an undef is returned.
3009       This is distinct from asking for something which does not exist in a
3010       driver which supports column_info as a valid statement handle to an
3011       empty result-set will be returned in this case.
3012
3013       If the arguments don't match any tables then you'll still get a
3014       statement handle, it'll just return no rows.
3015
3016       The statement handle returned has at least the following fields in the
3017       order shown below. Other fields, after these, may also be present.
3018
3019       TABLE_CAT: The catalog identifier.  This field is NULL ("undef") if not
3020       applicable to the data source, which is often the case.  This field is
3021       empty if not applicable to the table.
3022
3023       TABLE_SCHEM: The schema identifier.  This field is NULL ("undef") if
3024       not applicable to the data source, and empty if not applicable to the
3025       table.
3026
3027       TABLE_NAME: The table identifier.  Note: A driver may provide column
3028       metadata not only for base tables, but also for derived objects like
3029       SYNONYMS etc.
3030
3031       COLUMN_NAME: The column identifier.
3032
3033       DATA_TYPE: The concise data type code.
3034
3035       TYPE_NAME: A data source dependent data type name.
3036
3037       COLUMN_SIZE: The column size.  This is the maximum length in characters
3038       for character data types, the number of digits or bits for numeric data
3039       types or the length in the representation of temporal types.  See the
3040       relevant specifications for detailed information.
3041
3042       BUFFER_LENGTH: The length in bytes of transferred data.
3043
3044       DECIMAL_DIGITS: The total number of significant digits to the right of
3045       the decimal point.
3046
3047       NUM_PREC_RADIX: The radix for numeric precision.  The value is 10 or 2
3048       for numeric data types and NULL ("undef") if not applicable.
3049
3050       NULLABLE: Indicates if a column can accept NULLs.  The following values
3051       are defined:
3052
3053         SQL_NO_NULLS          0
3054         SQL_NULLABLE          1
3055         SQL_NULLABLE_UNKNOWN  2
3056
3057       REMARKS: A description of the column.
3058
3059       COLUMN_DEF: The default value of the column, in a format that can be
3060       used directly in an SQL statement.
3061
3062       Note that this may be an expression and not simply the text used for
3063       the default value in the original CREATE TABLE statement. For example,
3064       given:
3065
3066           col1 char(30) default current_user    -- a 'function'
3067           col2 char(30) default 'string'        -- a string literal
3068
3069       where "current_user" is the name of a function, the corresponding
3070       "COLUMN_DEF" values would be:
3071
3072           Database        col1                     col2
3073           --------        ----                     ----
3074           Oracle:         current_user             'string'
3075           Postgres:       "current_user"()         'string'::text
3076           MS SQL:         (user_name())            ('string')
3077
3078       SQL_DATA_TYPE: The SQL data type.
3079
3080       SQL_DATETIME_SUB: The subtype code for datetime and interval data
3081       types.
3082
3083       CHAR_OCTET_LENGTH: The maximum length in bytes of a character or binary
3084       data type column.
3085
3086       ORDINAL_POSITION: The column sequence number (starting with 1).
3087
3088       IS_NULLABLE: Indicates if the column can accept NULLs.  Possible values
3089       are: 'NO', 'YES' and ''.
3090
3091       SQL/CLI defines the following additional columns:
3092
3093         CHAR_SET_CAT
3094         CHAR_SET_SCHEM
3095         CHAR_SET_NAME
3096         COLLATION_CAT
3097         COLLATION_SCHEM
3098         COLLATION_NAME
3099         UDT_CAT
3100         UDT_SCHEM
3101         UDT_NAME
3102         DOMAIN_CAT
3103         DOMAIN_SCHEM
3104         DOMAIN_NAME
3105         SCOPE_CAT
3106         SCOPE_SCHEM
3107         SCOPE_NAME
3108         MAX_CARDINALITY
3109         DTD_IDENTIFIER
3110         IS_SELF_REF
3111
3112       Drivers capable of supplying any of those values should do so in the
3113       corresponding column and supply undef values for the others.
3114
3115       Drivers wishing to provide extra database/driver specific information
3116       should do so in extra columns beyond all those listed above, and use
3117       lowercase field names with the driver-specific prefix (i.e.,
3118       'ora_...'). Applications accessing such fields should do so by name and
3119       not by column number.
3120
3121       The result set is ordered by TABLE_CAT, TABLE_SCHEM, TABLE_NAME and
3122       ORDINAL_POSITION.
3123
3124       Note: There is some overlap with statement handle attributes (in perl)
3125       and SQLDescribeCol (in ODBC). However, SQLColumns provides more
3126       metadata.
3127
3128       See also "Catalog Methods" and "Standards Reference Information".
3129
3130       "primary_key_info"
3131
3132         $sth = $dbh->primary_key_info( $catalog, $schema, $table );
3133
3134         # then $sth->fetchall_arrayref or $sth->fetchall_hashref etc
3135
3136       Returns an active statement handle that can be used to fetch
3137       information about columns that make up the primary key for a table.
3138       The arguments don't accept search patterns (unlike table_info()).
3139
3140       The statement handle will return one row per column, ordered by
3141       TABLE_CAT, TABLE_SCHEM, TABLE_NAME, and KEY_SEQ.  If there is no
3142       primary key then the statement handle will fetch no rows.
3143
3144       Note: The support for the selection criteria, such as $catalog, is
3145       driver specific.  If the driver doesn't support catalogs and/or
3146       schemas, it may ignore these criteria.
3147
3148       The statement handle returned has at least the following fields in the
3149       order shown below. Other fields, after these, may also be present.
3150
3151       TABLE_CAT: The catalog identifier.  This field is NULL ("undef") if not
3152       applicable to the data source, which is often the case.  This field is
3153       empty if not applicable to the table.
3154
3155       TABLE_SCHEM: The schema identifier.  This field is NULL ("undef") if
3156       not applicable to the data source, and empty if not applicable to the
3157       table.
3158
3159       TABLE_NAME: The table identifier.
3160
3161       COLUMN_NAME: The column identifier.
3162
3163       KEY_SEQ: The column sequence number (starting with 1).  Note: This
3164       field is named ORDINAL_POSITION in SQL/CLI.
3165
3166       PK_NAME: The primary key constraint identifier.  This field is NULL
3167       ("undef") if not applicable to the data source.
3168
3169       See also "Catalog Methods" and "Standards Reference Information".
3170
3171       "primary_key"
3172
3173         @key_column_names = $dbh->primary_key( $catalog, $schema, $table );
3174
3175       Simple interface to the primary_key_info() method. Returns a list of
3176       the column names that comprise the primary key of the specified table.
3177       The list is in primary key column sequence order.  If there is no
3178       primary key then an empty list is returned.
3179
3180       "foreign_key_info"
3181
3182         $sth = $dbh->foreign_key_info( $pk_catalog, $pk_schema, $pk_table
3183                                      , $fk_catalog, $fk_schema, $fk_table );
3184
3185         $sth = $dbh->foreign_key_info( $pk_catalog, $pk_schema, $pk_table
3186                                      , $fk_catalog, $fk_schema, $fk_table
3187                                      , \%attr );
3188
3189         # then $sth->fetchall_arrayref or $sth->fetchall_hashref etc
3190
3191       Returns an active statement handle that can be used to fetch
3192       information about foreign keys in and/or referencing the specified
3193       table(s).  The arguments don't accept search patterns (unlike
3194       table_info()).
3195
3196       $pk_catalog, $pk_schema, $pk_table identify the primary (unique) key
3197       table (PKT).
3198
3199       $fk_catalog, $fk_schema, $fk_table identify the foreign key table
3200       (FKT).
3201
3202       If both PKT and FKT are given, the function returns the foreign key, if
3203       any, in table FKT that refers to the primary (unique) key of table PKT.
3204       (Note: In SQL/CLI, the result is implementation-defined.)
3205
3206       If only PKT is given, then the result set contains the primary key of
3207       that table and all foreign keys that refer to it.
3208
3209       If only FKT is given, then the result set contains all foreign keys in
3210       that table and the primary keys to which they refer.  (Note: In
3211       SQL/CLI, the result includes unique keys too.)
3212
3213       For example:
3214
3215         $sth = $dbh->foreign_key_info( undef, $user, 'master');
3216         $sth = $dbh->foreign_key_info( undef, undef,   undef , undef, $user, 'detail');
3217         $sth = $dbh->foreign_key_info( undef, $user, 'master', undef, $user, 'detail');
3218
3219         # then $sth->fetchall_arrayref or $sth->fetchall_hashref etc
3220
3221       Note: The support for the selection criteria, such as $catalog, is
3222       driver specific.  If the driver doesn't support catalogs and/or
3223       schemas, it may ignore these criteria.
3224
3225       The statement handle returned has the following fields in the order
3226       shown below.  Because ODBC never includes unique keys, they define
3227       different columns in the result set than SQL/CLI. SQL/CLI column names
3228       are shown in parentheses.
3229
3230       PKTABLE_CAT    ( UK_TABLE_CAT      ): The primary (unique) key table
3231       catalog identifier.  This field is NULL ("undef") if not applicable to
3232       the data source, which is often the case.  This field is empty if not
3233       applicable to the table.
3234
3235       PKTABLE_SCHEM  ( UK_TABLE_SCHEM    ): The primary (unique) key table
3236       schema identifier.  This field is NULL ("undef") if not applicable to
3237       the data source, and empty if not applicable to the table.
3238
3239       PKTABLE_NAME   ( UK_TABLE_NAME     ): The primary (unique) key table
3240       identifier.
3241
3242       PKCOLUMN_NAME  (UK_COLUMN_NAME    ): The primary (unique) key column
3243       identifier.
3244
3245       FKTABLE_CAT    ( FK_TABLE_CAT      ): The foreign key table catalog
3246       identifier.  This field is NULL ("undef") if not applicable to the data
3247       source, which is often the case.  This field is empty if not applicable
3248       to the table.
3249
3250       FKTABLE_SCHEM  ( FK_TABLE_SCHEM    ): The foreign key table schema
3251       identifier.  This field is NULL ("undef") if not applicable to the data
3252       source, and empty if not applicable to the table.
3253
3254       FKTABLE_NAME   ( FK_TABLE_NAME     ): The foreign key table identifier.
3255
3256       FKCOLUMN_NAME  ( FK_COLUMN_NAME    ): The foreign key column
3257       identifier.
3258
3259       KEY_SEQ        ( ORDINAL_POSITION  ): The column sequence number
3260       (starting with 1).
3261
3262       UPDATE_RULE    ( UPDATE_RULE       ): The referential action for the
3263       UPDATE rule.  The following codes are defined:
3264
3265         CASCADE              0
3266         RESTRICT             1
3267         SET NULL             2
3268         NO ACTION            3
3269         SET DEFAULT          4
3270
3271       DELETE_RULE    ( DELETE_RULE       ): The referential action for the
3272       DELETE rule.  The codes are the same as for UPDATE_RULE.
3273
3274       FK_NAME        ( FK_NAME           ): The foreign key name.
3275
3276       PK_NAME        ( UK_NAME           ): The primary (unique) key name.
3277
3278       DEFERRABILITY  ( DEFERABILITY      ): The deferrability of the foreign
3279       key constraint.  The following codes are defined:
3280
3281         INITIALLY DEFERRED   5
3282         INITIALLY IMMEDIATE  6
3283         NOT DEFERRABLE       7
3284
3285                      ( UNIQUE_OR_PRIMARY ): This column is necessary if a
3286       driver includes all candidate (i.e. primary and alternate) keys in the
3287       result set (as specified by SQL/CLI).  The value of this column is
3288       UNIQUE if the foreign key references an alternate key and PRIMARY if
3289       the foreign key references a primary key, or it may be undefined if the
3290       driver doesn't have access to the information.
3291
3292       See also "Catalog Methods" and "Standards Reference Information".
3293
3294       "statistics_info"
3295
3296       Warning: This method is experimental and may change.
3297
3298         $sth = $dbh->statistics_info( $catalog, $schema, $table, $unique_only, $quick );
3299
3300         # then $sth->fetchall_arrayref or $sth->fetchall_hashref etc
3301
3302       Returns an active statement handle that can be used to fetch
3303       statistical information about a table and its indexes.
3304
3305       The arguments don't accept search patterns (unlike "table_info").
3306
3307       If the boolean argument $unique_only is true, only UNIQUE indexes will
3308       be returned in the result set, otherwise all indexes will be returned.
3309
3310       If the boolean argument $quick is set, the actual statistical
3311       information columns (CARDINALITY and PAGES) will only be returned if
3312       they are readily available from the server, and might not be current.
3313       Some databases may return stale statistics or no statistics at all with
3314       this flag set.
3315
3316       The statement handle will return at most one row per column name per
3317       index, plus at most one row for the entire table itself, ordered by
3318       NON_UNIQUE, TYPE, INDEX_QUALIFIER, INDEX_NAME, and ORDINAL_POSITION.
3319
3320       Note: The support for the selection criteria, such as $catalog, is
3321       driver specific.  If the driver doesn't support catalogs and/or
3322       schemas, it may ignore these criteria.
3323
3324       The statement handle returned has at least the following fields in the
3325       order shown below. Other fields, after these, may also be present.
3326
3327       TABLE_CAT: The catalog identifier.  This field is NULL ("undef") if not
3328       applicable to the data source, which is often the case.  This field is
3329       empty if not applicable to the table.
3330
3331       TABLE_SCHEM: The schema identifier.  This field is NULL ("undef") if
3332       not applicable to the data source, and empty if not applicable to the
3333       table.
3334
3335       TABLE_NAME: The table identifier.
3336
3337       NON_UNIQUE: Unique index indicator.  Returns 0 for unique indexes, 1
3338       for non-unique indexes
3339
3340       INDEX_QUALIFIER: Index qualifier identifier.  The identifier that is
3341       used to qualify the index name when doing a "DROP INDEX"; NULL
3342       ("undef") is returned if an index qualifier is not supported by the
3343       data source.  If a non-NULL (defined) value is returned in this column,
3344       it must be used to qualify the index name on a "DROP INDEX" statement;
3345       otherwise, the TABLE_SCHEM should be used to qualify the index name.
3346
3347       INDEX_NAME: The index identifier.
3348
3349       TYPE: The type of information being returned.  Can be any of the
3350       following values: 'table', 'btree', 'clustered', 'content', 'hashed',
3351       or 'other'.
3352
3353       In the case that this field is 'table', all fields other than
3354       TABLE_CAT, TABLE_SCHEM, TABLE_NAME, TYPE, CARDINALITY, and PAGES will
3355       be NULL ("undef").
3356
3357       ORDINAL_POSITION: Column sequence number (starting with 1).
3358
3359       COLUMN_NAME: The column identifier.
3360
3361       ASC_OR_DESC: Column sort sequence.  "A" for Ascending, "D" for
3362       Descending, or NULL ("undef") if not supported for this index.
3363
3364       CARDINALITY: Cardinality of the table or index.  For indexes, this is
3365       the number of unique values in the index.  For tables, this is the
3366       number of rows in the table.  If not supported, the value will be NULL
3367       ("undef").
3368
3369       PAGES: Number of storage pages used by this table or index.  If not
3370       supported, the value will be NULL ("undef").
3371
3372       FILTER_CONDITION: The index filter condition as a string.  If the index
3373       is not a filtered index, or it cannot be determined whether the index
3374       is a filtered index, this value is NULL ("undef").  If the index is a
3375       filtered index, but the filter condition cannot be determined, this
3376       value is the empty string ''.  Otherwise it will be the literal filter
3377       condition as a string, such as "SALARY <= 4500".
3378
3379       See also "Catalog Methods" and "Standards Reference Information".
3380
3381       "tables"
3382
3383         @names = $dbh->tables( $catalog, $schema, $table, $type );
3384         @names = $dbh->tables;        # deprecated
3385
3386       Simple interface to table_info(). Returns a list of matching table
3387       names, possibly including a catalog/schema prefix.
3388
3389       See "table_info" for a description of the parameters.
3390
3391       If "$dbh->get_info(29)" returns true (29 is SQL_IDENTIFIER_QUOTE_CHAR)
3392       then the table names are constructed and quoted by "quote_identifier"
3393       to ensure they are usable even if they contain whitespace or reserved
3394       words etc. This means that the table names returned will include quote
3395       characters.
3396
3397       "type_info_all"
3398
3399         $type_info_all = $dbh->type_info_all;
3400
3401       Returns a reference to an array which holds information about each data
3402       type variant supported by the database and driver. The array and its
3403       contents should be treated as read-only.
3404
3405       The first item is a reference to an 'index' hash of "Name ="> "Index"
3406       pairs.  The items following that are references to arrays, one per
3407       supported data type variant. The leading index hash defines the names
3408       and order of the fields within the arrays that follow it.  For example:
3409
3410         $type_info_all = [
3411           {   TYPE_NAME         => 0,
3412               DATA_TYPE         => 1,
3413               COLUMN_SIZE       => 2,     # was PRECISION originally
3414               LITERAL_PREFIX    => 3,
3415               LITERAL_SUFFIX    => 4,
3416               CREATE_PARAMS     => 5,
3417               NULLABLE          => 6,
3418               CASE_SENSITIVE    => 7,
3419               SEARCHABLE        => 8,
3420               UNSIGNED_ATTRIBUTE=> 9,
3421               FIXED_PREC_SCALE  => 10,    # was MONEY originally
3422               AUTO_UNIQUE_VALUE => 11,    # was AUTO_INCREMENT originally
3423               LOCAL_TYPE_NAME   => 12,
3424               MINIMUM_SCALE     => 13,
3425               MAXIMUM_SCALE     => 14,
3426               SQL_DATA_TYPE     => 15,
3427               SQL_DATETIME_SUB  => 16,
3428               NUM_PREC_RADIX    => 17,
3429               INTERVAL_PRECISION=> 18,
3430           },
3431           [ 'VARCHAR', SQL_VARCHAR,
3432               undef, "'","'", undef,0, 1,1,0,0,0,undef,1,255, undef
3433           ],
3434           [ 'INTEGER', SQL_INTEGER,
3435               undef,  "", "", undef,0, 0,1,0,0,0,undef,0,  0, 10
3436           ],
3437         ];
3438
3439       More than one row may have the same value in the "DATA_TYPE" field if
3440       there are different ways to spell the type name and/or there are
3441       variants of the type with different attributes (e.g., with and without
3442       "AUTO_UNIQUE_VALUE" set, with and without "UNSIGNED_ATTRIBUTE", etc).
3443
3444       The rows are ordered by "DATA_TYPE" first and then by how closely each
3445       type maps to the corresponding ODBC SQL data type, closest first.
3446
3447       The meaning of the fields is described in the documentation for the
3448       "type_info" method.
3449
3450       An 'index' hash is provided so you don't need to rely on index values
3451       defined above.  However, using DBD::ODBC with some old ODBC drivers may
3452       return older names, shown as comments in the example above.  Another
3453       issue with the index hash is that the lettercase of the keys is not
3454       defined. It is usually uppercase, as show here, but drivers may return
3455       names with any lettercase.
3456
3457       Drivers are also free to return extra driver-specific columns of
3458       information - though it's recommended that they start at column index
3459       50 to leave room for expansion of the DBI/ODBC specification.
3460
3461       The type_info_all() method is not normally used directly.  The
3462       "type_info" method provides a more usable and useful interface to the
3463       data.
3464
3465       "type_info"
3466
3467         @type_info = $dbh->type_info($data_type);
3468
3469       Returns a list of hash references holding information about one or more
3470       variants of $data_type. The list is ordered by "DATA_TYPE" first and
3471       then by how closely each type maps to the corresponding ODBC SQL data
3472       type, closest first.  If called in a scalar context then only the first
3473       (best) element is returned.
3474
3475       If $data_type is undefined or "SQL_ALL_TYPES", then the list will
3476       contain hashes for all data type variants supported by the database and
3477       driver.
3478
3479       If $data_type is an array reference then "type_info" returns the
3480       information for the first type in the array that has any matches.
3481
3482       The keys of the hash follow the same letter case conventions as the
3483       rest of the DBI (see "Naming Conventions and Name Space"). The
3484       following uppercase items should always exist, though may be undef:
3485
3486       TYPE_NAME (string)
3487           Data type name for use in CREATE TABLE statements etc.
3488
3489       DATA_TYPE (integer)
3490           SQL data type number.
3491
3492       COLUMN_SIZE (integer)
3493           For numeric types, this is either the total number of digits (if
3494           the NUM_PREC_RADIX value is 10) or the total number of bits allowed
3495           in the column (if NUM_PREC_RADIX is 2).
3496
3497           For string types, this is the maximum size of the string in
3498           characters.
3499
3500           For date and interval types, this is the maximum number of
3501           characters needed to display the value.
3502
3503       LITERAL_PREFIX (string)
3504           Characters used to prefix a literal. A typical prefix is ""'"" for
3505           characters, or possibly ""0x"" for binary values passed as
3506           hexadecimal.  NULL ("undef") is returned for data types for which
3507           this is not applicable.
3508
3509       LITERAL_SUFFIX (string)
3510           Characters used to suffix a literal. Typically ""'"" for
3511           characters.  NULL ("undef") is returned for data types where this
3512           is not applicable.
3513
3514       CREATE_PARAMS (string)
3515           Parameter names for data type definition. For example,
3516           "CREATE_PARAMS" for a "DECIMAL" would be ""precision,scale"" if the
3517           DECIMAL type should be declared as "DECIMAL("precision,scale")"
3518           where precision and scale are integer values.  For a "VARCHAR" it
3519           would be ""max length"".  NULL ("undef") is returned for data types
3520           for which this is not applicable.
3521
3522       NULLABLE (integer)
3523           Indicates whether the data type accepts a NULL value: 0 or an empty
3524           string = no, 1 = yes, 2 = unknown.
3525
3526       CASE_SENSITIVE (boolean)
3527           Indicates whether the data type is case sensitive in collations and
3528           comparisons.
3529
3530       SEARCHABLE (integer)
3531           Indicates how the data type can be used in a WHERE clause, as
3532           follows:
3533
3534             0 - Cannot be used in a WHERE clause
3535             1 - Only with a LIKE predicate
3536             2 - All comparison operators except LIKE
3537             3 - Can be used in a WHERE clause with any comparison operator
3538
3539       UNSIGNED_ATTRIBUTE (boolean)
3540           Indicates whether the data type is unsigned.  NULL ("undef") is
3541           returned for data types for which this is not applicable.
3542
3543       FIXED_PREC_SCALE (boolean)
3544           Indicates whether the data type always has the same precision and
3545           scale (such as a money type).  NULL ("undef") is returned for data
3546           types for which this is not applicable.
3547
3548       AUTO_UNIQUE_VALUE (boolean)
3549           Indicates whether a column of this data type is automatically set
3550           to a unique value whenever a new row is inserted.  NULL ("undef")
3551           is returned for data types for which this is not applicable.
3552
3553       LOCAL_TYPE_NAME (string)
3554           Localized version of the "TYPE_NAME" for use in dialog with users.
3555           NULL ("undef") is returned if a localized name is not available (in
3556           which case "TYPE_NAME" should be used).
3557
3558       MINIMUM_SCALE (integer)
3559           The minimum scale of the data type. If a data type has a fixed
3560           scale, then "MAXIMUM_SCALE" holds the same value.  NULL ("undef")
3561           is returned for data types for which this is not applicable.
3562
3563       MAXIMUM_SCALE (integer)
3564           The maximum scale of the data type. If a data type has a fixed
3565           scale, then "MINIMUM_SCALE" holds the same value.  NULL ("undef")
3566           is returned for data types for which this is not applicable.
3567
3568       SQL_DATA_TYPE (integer)
3569           This column is the same as the "DATA_TYPE" column, except for
3570           interval and datetime data types.  For interval and datetime data
3571           types, the "SQL_DATA_TYPE" field will return "SQL_INTERVAL" or
3572           "SQL_DATETIME", and the "SQL_DATETIME_SUB" field below will return
3573           the subcode for the specific interval or datetime data type. If
3574           this field is NULL, then the driver does not support or report on
3575           interval or datetime subtypes.
3576
3577       SQL_DATETIME_SUB (integer)
3578           For interval or datetime data types, where the "SQL_DATA_TYPE"
3579           field above is "SQL_INTERVAL" or "SQL_DATETIME", this field will
3580           hold the subcode for the specific interval or datetime data type.
3581           Otherwise it will be NULL ("undef").
3582
3583           Although not mentioned explicitly in the standards, it seems there
3584           is a simple relationship between these values:
3585
3586             DATA_TYPE == (10 * SQL_DATA_TYPE) + SQL_DATETIME_SUB
3587
3588       NUM_PREC_RADIX (integer)
3589           The radix value of the data type. For approximate numeric types,
3590           "NUM_PREC_RADIX" contains the value 2 and "COLUMN_SIZE" holds the
3591           number of bits. For exact numeric types, "NUM_PREC_RADIX" contains
3592           the value 10 and "COLUMN_SIZE" holds the number of decimal digits.
3593           NULL ("undef") is returned either for data types for which this is
3594           not applicable or if the driver cannot report this information.
3595
3596       INTERVAL_PRECISION (integer)
3597           The interval leading precision for interval types. NULL is returned
3598           either for data types for which this is not applicable or if the
3599           driver cannot report this information.
3600
3601       For example, to find the type name for the fields in a select statement
3602       you can do:
3603
3604         @names = map { scalar $dbh->type_info($_)->{TYPE_NAME} } @{ $sth->{TYPE} }
3605
3606       Since DBI and ODBC drivers vary in how they map their types into the
3607       ISO standard types you may need to search for more than one type.
3608       Here's an example looking for a usable type to store a date:
3609
3610         $my_date_type = $dbh->type_info( [ SQL_DATE, SQL_TIMESTAMP ] );
3611
3612       Similarly, to more reliably find a type to store small integers, you
3613       could use a list starting with "SQL_SMALLINT", "SQL_INTEGER",
3614       "SQL_DECIMAL", etc.
3615
3616       See also "Standards Reference Information".
3617
3618       "quote"
3619
3620         $sql = $dbh->quote($value);
3621         $sql = $dbh->quote($value, $data_type);
3622
3623       Quote a string literal for use as a literal value in an SQL statement,
3624       by escaping any special characters (such as quotation marks) contained
3625       within the string and adding the required type of outer quotation
3626       marks.
3627
3628         $sql = sprintf "SELECT foo FROM bar WHERE baz = %s",
3629                       $dbh->quote("Don't");
3630
3631       For most database types, at least those that conform to SQL standards,
3632       quote would return 'Don''t' (including the outer quotation marks). For
3633       others it may return something like 'Don\'t'
3634
3635       An undefined $value value will be returned as the string "NULL"
3636       (without single quotation marks) to match how NULLs are represented in
3637       SQL.
3638
3639       If $data_type is supplied, it is used to try to determine the required
3640       quoting behaviour by using the information returned by "type_info".  As
3641       a special case, the standard numeric types are optimized to return
3642       $value without calling "type_info".
3643
3644       Quote will probably not be able to deal with all possible input (such
3645       as binary data or data containing newlines), and is not related in any
3646       way with escaping or quoting shell meta-characters.
3647
3648       It is valid for the quote() method to return an SQL expression that
3649       evaluates to the desired string. For example:
3650
3651         $quoted = $dbh->quote("one\ntwo\0three")
3652
3653       may return something like:
3654
3655         CONCAT('one', CHAR(12), 'two', CHAR(0), 'three')
3656
3657       The quote() method should not be used with "Placeholders and Bind
3658       Values".
3659
3660       "quote_identifier"
3661
3662         $sql = $dbh->quote_identifier( $name );
3663         $sql = $dbh->quote_identifier( $catalog, $schema, $table, \%attr );
3664
3665       Quote an identifier (table name etc.) for use in an SQL statement, by
3666       escaping any special characters (such as double quotation marks) it
3667       contains and adding the required type of outer quotation marks.
3668
3669       Undefined names are ignored and the remainder are quoted and then
3670       joined together, typically with a dot (".") character. For example:
3671
3672         $id = $dbh->quote_identifier( undef, 'Her schema', 'My table' );
3673
3674       would, for most database types, return "Her schema"."My table"
3675       (including all the double quotation marks).
3676
3677       If three names are supplied then the first is assumed to be a catalog
3678       name and special rules may be applied based on what "get_info" returns
3679       for SQL_CATALOG_NAME_SEPARATOR (41) and SQL_CATALOG_LOCATION (114).
3680       For example, for Oracle:
3681
3682         $id = $dbh->quote_identifier( 'link', 'schema', 'table' );
3683
3684       would return "schema"."table"@"link".
3685
3686       "take_imp_data"
3687
3688         $imp_data = $dbh->take_imp_data;
3689
3690       Leaves the $dbh in an almost dead, zombie-like, state and returns a
3691       binary string of raw implementation data from the driver which
3692       describes the current database connection. Effectively it detaches the
3693       underlying database API connection data from the DBI handle.  After
3694       calling take_imp_data(), all other methods except "DESTROY" will
3695       generate a warning and return undef.
3696
3697       Why would you want to do this? You don't, forget I even mentioned it.
3698       Unless, that is, you're implementing something advanced like a multi-
3699       threaded connection pool. See DBI::Pool.
3700
3701       The returned $imp_data can be passed as a "dbi_imp_data" attribute to a
3702       later connect() call, even in a separate thread in the same process,
3703       where the driver can use it to 'adopt' the existing connection that the
3704       implementation data was taken from.
3705
3706       Some things to keep in mind...
3707
3708       * the $imp_data holds the only reference to the underlying database API
3709       connection data. That connection is still 'live' and won't be cleaned
3710       up properly unless the $imp_data is used to create a new $dbh which is
3711       then allowed to disconnect() normally.
3712
3713       * using the same $imp_data to create more than one other new $dbh at a
3714       time may well lead to unpleasant problems. Don't do that.
3715
3716       Any child statement handles are effectively destroyed when
3717       take_imp_data() is called.
3718
3719       The "take_imp_data" method was added in DBI 1.36 but wasn't useful till
3720       1.49.
3721
3722   Database Handle Attributes
3723       This section describes attributes specific to database handles.
3724
3725       Changes to these database handle attributes do not affect any other
3726       existing or future database handles.
3727
3728       Attempting to set or get the value of an unknown attribute generates a
3729       warning, except for private driver-specific attributes (which all have
3730       names starting with a lowercase letter).
3731
3732       Example:
3733
3734         $h->{AutoCommit} = ...;       # set/write
3735         ... = $h->{AutoCommit};       # get/read
3736
3737       "AutoCommit"
3738
3739       Type: boolean
3740
3741       If true, then database changes cannot be rolled-back (undone).  If
3742       false, then database changes automatically occur within a
3743       "transaction", which must either be committed or rolled back using the
3744       "commit" or "rollback" methods.
3745
3746       Drivers should always default to "AutoCommit" mode (an unfortunate
3747       choice largely forced on the DBI by ODBC and JDBC conventions.)
3748
3749       Attempting to set "AutoCommit" to an unsupported value is a fatal
3750       error.  This is an important feature of the DBI. Applications that need
3751       full transaction behaviour can set "$dbh->{AutoCommit} = 0" (or set
3752       "AutoCommit" to 0 via "connect") without having to check that the value
3753       was assigned successfully.
3754
3755       For the purposes of this description, we can divide databases into
3756       three categories:
3757
3758         Databases which don't support transactions at all.
3759         Databases in which a transaction is always active.
3760         Databases in which a transaction must be explicitly started (C<'BEGIN WORK'>).
3761
3762       * Databases which don't support transactions at all
3763
3764       For these databases, attempting to turn "AutoCommit" off is a fatal
3765       error.  "commit" and "rollback" both issue warnings about being
3766       ineffective while "AutoCommit" is in effect.
3767
3768       * Databases in which a transaction is always active
3769
3770       These are typically mainstream commercial relational databases with
3771       "ANSI standard" transaction behaviour.  If "AutoCommit" is off, then
3772       changes to the database won't have any lasting effect unless "commit"
3773       is called (but see also "disconnect"). If "rollback" is called then any
3774       changes since the last commit are undone.
3775
3776       If "AutoCommit" is on, then the effect is the same as if the DBI called
3777       "commit" automatically after every successful database operation. So
3778       calling "commit" or "rollback" explicitly while "AutoCommit" is on
3779       would be ineffective because the changes would have already been
3780       committed.
3781
3782       Changing "AutoCommit" from off to on will trigger a "commit".
3783
3784       For databases which don't support a specific auto-commit mode, the
3785       driver has to commit each statement automatically using an explicit
3786       "COMMIT" after it completes successfully (and roll it back using an
3787       explicit "ROLLBACK" if it fails).  The error information reported to
3788       the application will correspond to the statement which was executed,
3789       unless it succeeded and the commit or rollback failed.
3790
3791       * Databases in which a transaction must be explicitly started
3792
3793       For these databases, the intention is to have them act like databases
3794       in which a transaction is always active (as described above).
3795
3796       To do this, the driver will automatically begin an explicit transaction
3797       when "AutoCommit" is turned off, or after a "commit" or "rollback" (or
3798       when the application issues the next database operation after one of
3799       those events).
3800
3801       In this way, the application does not have to treat these databases as
3802       a special case.
3803
3804       See "commit", "disconnect" and "Transactions" for other important notes
3805       about transactions.
3806
3807       "Driver"
3808
3809       Type: handle
3810
3811       Holds the handle of the parent driver. The only recommended use for
3812       this is to find the name of the driver using:
3813
3814         $dbh->{Driver}->{Name}
3815
3816       "Name"
3817
3818       Type: string
3819
3820       Holds the "name" of the database. Usually (and recommended to be) the
3821       same as the ""dbi:DriverName:..."" string used to connect to the
3822       database, but with the leading ""dbi:DriverName:"" removed.
3823
3824       "Statement"
3825
3826       Type: string, read-only
3827
3828       Returns the statement string passed to the most recent "prepare" or
3829       "do" method called in this database handle, even if that method failed.
3830       This is especially useful where "RaiseError" is enabled and the
3831       exception handler checks $@ and sees that a 'prepare' method call
3832       failed.
3833
3834       "RowCacheSize"
3835
3836       Type: integer
3837
3838       A hint to the driver indicating the size of the local row cache that
3839       the application would like the driver to use for future "SELECT"
3840       statements.  If a row cache is not implemented, then setting
3841       "RowCacheSize" is ignored and getting the value returns "undef".
3842
3843       Some "RowCacheSize" values have special meaning, as follows:
3844
3845         0 - Automatically determine a reasonable cache size for each C<SELECT>
3846         1 - Disable the local row cache
3847        >1 - Cache this many rows
3848        <0 - Cache as many rows that will fit into this much memory for each C<SELECT>.
3849
3850       Note that large cache sizes may require a very large amount of memory
3851       (cached rows * maximum size of row). Also, a large cache will cause a
3852       longer delay not only for the first fetch, but also whenever the cache
3853       needs refilling.
3854
3855       See also the "RowsInCache" statement handle attribute.
3856
3857       "Username"
3858
3859       Type: string
3860
3861       Returns the username used to connect to the database.
3862

DBI STATEMENT HANDLE OBJECTS

3864       This section lists the methods and attributes associated with DBI
3865       statement handles.
3866
3867   Statement Handle Methods
3868       The DBI defines the following methods for use on DBI statement handles:
3869
3870       "bind_param"
3871
3872         $sth->bind_param($p_num, $bind_value)
3873         $sth->bind_param($p_num, $bind_value, \%attr)
3874         $sth->bind_param($p_num, $bind_value, $bind_type)
3875
3876       The "bind_param" method takes a copy of $bind_value and associates it
3877       (binds it) with a placeholder, identified by $p_num, embedded in the
3878       prepared statement. Placeholders are indicated with question mark
3879       character ("?"). For example:
3880
3881         $dbh->{RaiseError} = 1;        # save having to check each method call
3882         $sth = $dbh->prepare("SELECT name, age FROM people WHERE name LIKE ?");
3883         $sth->bind_param(1, "John%");  # placeholders are numbered from 1
3884         $sth->execute;
3885         DBI::dump_results($sth);
3886
3887       See "Placeholders and Bind Values" for more information.
3888
3889       Data Types for Placeholders
3890
3891       The "\%attr" parameter can be used to hint at the data type the
3892       placeholder should have. This is rarely needed. Typically, the driver
3893       is only interested in knowing if the placeholder should be bound as a
3894       number or a string.
3895
3896         $sth->bind_param(1, $value, { TYPE => SQL_INTEGER });
3897
3898       As a short-cut for the common case, the data type can be passed
3899       directly, in place of the "\%attr" hash reference. This example is
3900       equivalent to the one above:
3901
3902         $sth->bind_param(1, $value, SQL_INTEGER);
3903
3904       The "TYPE" value indicates the standard (non-driver-specific) type for
3905       this parameter. To specify the driver-specific type, the driver may
3906       support a driver-specific attribute, such as "{ ora_type => 97 }".
3907
3908       The SQL_INTEGER and other related constants can be imported using
3909
3910         use DBI qw(:sql_types);
3911
3912       See "DBI Constants" for more information.
3913
3914       The data type is 'sticky' in that bind values passed to execute() are
3915       bound with the data type specified by earlier bind_param() calls, if
3916       any.  Portable applications should not rely on being able to change the
3917       data type after the first "bind_param" call.
3918
3919       Perl only has string and number scalar data types. All database types
3920       that aren't numbers are bound as strings and must be in a format the
3921       database will understand except where the bind_param() TYPE attribute
3922       specifies a type that implies a particular format. For example, given:
3923
3924         $sth->bind_param(1, $value, SQL_DATETIME);
3925
3926       the driver should expect $value to be in the ODBC standard SQL_DATETIME
3927       format, which is 'YYYY-MM-DD HH:MM:SS'. Similarly for SQL_DATE,
3928       SQL_TIME etc.
3929
3930       As an alternative to specifying the data type in the "bind_param" call,
3931       you can let the driver pass the value as the default type ("VARCHAR").
3932       You can then use an SQL function to convert the type within the
3933       statement.  For example:
3934
3935         INSERT INTO price(code, price) VALUES (?, CONVERT(MONEY,?))
3936
3937       The "CONVERT" function used here is just an example. The actual
3938       function and syntax will vary between different databases and is non-
3939       portable.
3940
3941       See also "Placeholders and Bind Values" for more information.
3942
3943       "bind_param_inout"
3944
3945         $rc = $sth->bind_param_inout($p_num, \$bind_value, $max_len)  or die $sth->errstr;
3946         $rv = $sth->bind_param_inout($p_num, \$bind_value, $max_len, \%attr)     or ...
3947         $rv = $sth->bind_param_inout($p_num, \$bind_value, $max_len, $bind_type) or ...
3948
3949       This method acts like "bind_param", but also enables values to be
3950       updated by the statement. The statement is typically a call to a stored
3951       procedure. The $bind_value must be passed as a reference to the actual
3952       value to be used.
3953
3954       Note that unlike "bind_param", the $bind_value variable is not copied
3955       when "bind_param_inout" is called. Instead, the value in the variable
3956       is read at the time "execute" is called.
3957
3958       The additional $max_len parameter specifies the minimum amount of
3959       memory to allocate to $bind_value for the new value. If the value
3960       returned from the database is too big to fit, then the execution should
3961       fail. If unsure what value to use, pick a generous length, i.e., a
3962       length larger than the longest value that would ever be returned.  The
3963       only cost of using a larger value than needed is wasted memory.
3964
3965       Undefined values or "undef" are used to indicate null values.  See also
3966       "Placeholders and Bind Values" for more information.
3967
3968       "bind_param_array"
3969
3970         $rc = $sth->bind_param_array($p_num, $array_ref_or_value)
3971         $rc = $sth->bind_param_array($p_num, $array_ref_or_value, \%attr)
3972         $rc = $sth->bind_param_array($p_num, $array_ref_or_value, $bind_type)
3973
3974       The "bind_param_array" method is used to bind an array of values to a
3975       placeholder embedded in the prepared statement which is to be executed
3976       with "execute_array". For example:
3977
3978         $dbh->{RaiseError} = 1;        # save having to check each method call
3979         $sth = $dbh->prepare("INSERT INTO staff (first_name, last_name, dept) VALUES(?, ?, ?)");
3980         $sth->bind_param_array(1, [ 'John', 'Mary', 'Tim' ]);
3981         $sth->bind_param_array(2, [ 'Booth', 'Todd', 'Robinson' ]);
3982         $sth->bind_param_array(3, "SALES"); # scalar will be reused for each row
3983         $sth->execute_array( { ArrayTupleStatus => \my @tuple_status } );
3984
3985       The %attr ($bind_type) argument is the same as defined for
3986       "bind_param".  Refer to "bind_param" for general details on using
3987       placeholders.
3988
3989       (Note that bind_param_array() can not be used to expand a placeholder
3990       into a list of values for a statement like "SELECT foo WHERE bar IN
3991       (?)".  A placeholder can only ever represent one value per execution.)
3992
3993       Scalar values, including "undef", may also be bound by
3994       "bind_param_array". In which case the same value will be used for each
3995       "execute" call. Driver-specific implementations may behave differently,
3996       e.g., when binding to a stored procedure call, some databases may
3997       permit mixing scalars and arrays as arguments.
3998
3999       The default implementation provided by DBI (for drivers that have not
4000       implemented array binding) is to iteratively call "execute" for each
4001       parameter tuple provided in the bound arrays.  Drivers may provide more
4002       optimized implementations using whatever bulk operation support the
4003       database API provides. The default driver behaviour should match the
4004       default DBI behaviour, but always consult your driver documentation as
4005       there may be driver specific issues to consider.
4006
4007       Note that the default implementation currently only supports non-data
4008       returning statements (INSERT, UPDATE, but not SELECT). Also,
4009       "bind_param_array" and "bind_param" cannot be mixed in the same
4010       statement execution, and "bind_param_array" must be used with
4011       "execute_array"; using "bind_param_array" will have no effect for
4012       "execute".
4013
4014       The "bind_param_array" method was added in DBI 1.22.
4015
4016       "execute"
4017
4018         $rv = $sth->execute                or die $sth->errstr;
4019         $rv = $sth->execute(@bind_values)  or die $sth->errstr;
4020
4021       Perform whatever processing is necessary to execute the prepared
4022       statement.  An "undef" is returned if an error occurs.  A successful
4023       "execute" always returns true regardless of the number of rows
4024       affected, even if it's zero (see below). It is always important to
4025       check the return status of "execute" (and most other DBI methods) for
4026       errors if you're not using "RaiseError".
4027
4028       For a non-"SELECT" statement, "execute" returns the number of rows
4029       affected, if known. If no rows were affected, then "execute" returns
4030       "0E0", which Perl will treat as 0 but will regard as true. Note that it
4031       is not an error for no rows to be affected by a statement. If the
4032       number of rows affected is not known, then "execute" returns -1.
4033
4034       For "SELECT" statements, execute simply "starts" the query within the
4035       database engine. Use one of the fetch methods to retrieve the data
4036       after calling "execute".  The "execute" method does not return the
4037       number of rows that will be returned by the query (because most
4038       databases can't tell in advance), it simply returns a true value.
4039
4040       You can tell if the statement was a "SELECT" statement by checking if
4041       "$sth->{NUM_OF_FIELDS}" is greater than zero after calling "execute".
4042
4043       If any arguments are given, then "execute" will effectively call
4044       "bind_param" for each value before executing the statement.  Values
4045       bound in this way are usually treated as "SQL_VARCHAR" types unless the
4046       driver can determine the correct type (which is rare), or unless
4047       "bind_param" (or "bind_param_inout") has already been used to specify
4048       the type.
4049
4050       Note that passing "execute" an empty array is the same as passing no
4051       arguments at all, which will execute the statement with previously
4052       bound values.  That's probably not what you want.
4053
4054       If execute() is called on a statement handle that's still active
4055       ($sth->{Active} is true) then it should effectively call finish() to
4056       tidy up the previous execution results before starting this new
4057       execution.
4058
4059       "execute_array"
4060
4061         $tuples = $sth->execute_array(\%attr) or die $sth->errstr;
4062         $tuples = $sth->execute_array(\%attr, @bind_values) or die $sth->errstr;
4063
4064         ($tuples, $rows) = $sth->execute_array(\%attr) or die $sth->errstr;
4065         ($tuples, $rows) = $sth->execute_array(\%attr, @bind_values) or die $sth->errstr;
4066
4067       Execute the prepared statement once for each parameter tuple (group of
4068       values) provided either in the @bind_values, or by prior calls to
4069       "bind_param_array", or via a reference passed in \%attr.
4070
4071       When called in scalar context the execute_array() method returns the
4072       number of tuples executed, or "undef" if an error occurred.  Like
4073       execute(), a successful execute_array() always returns true regardless
4074       of the number of tuples executed, even if it's zero. If there were any
4075       errors the ArrayTupleStatus array can be used to discover which tuples
4076       failed and with what errors.
4077
4078       When called in list context the execute_array() method returns two
4079       scalars; $tuples is the same as calling execute_array() in scalar
4080       context and $rows is the number of rows affected for each tuple, if
4081       available or -1 if the driver cannot determine this. NOTE, some drivers
4082       cannot determine the number of rows affected per tuple but can provide
4083       the number of rows affected for the batch.  If you are doing an update
4084       operation the returned rows affected may not be what you expect if, for
4085       instance, one or more of the tuples affected the same row multiple
4086       times.  Some drivers may not yet support list context, in which case
4087       $rows will be undef, or may not be able to provide the number of rows
4088       affected when performing this batch operation, in which case $rows will
4089       be -1.
4090
4091       Bind values for the tuples to be executed may be supplied row-wise by
4092       an "ArrayTupleFetch" attribute, or else column-wise in the @bind_values
4093       argument, or else column-wise by prior calls to "bind_param_array".
4094
4095       Where column-wise binding is used (via the @bind_values argument or
4096       calls to bind_param_array()) the maximum number of elements in any one
4097       of the bound value arrays determines the number of tuples executed.
4098       Placeholders with fewer values in their parameter arrays are treated as
4099       if padded with undef (NULL) values.
4100
4101       If a scalar value is bound, instead of an array reference, it is
4102       treated as a variable length array with all elements having the same
4103       value. It does not influence the number of tuples executed, so if all
4104       bound arrays have zero elements then zero tuples will be executed. If
4105       all bound values are scalars then one tuple will be executed, making
4106       execute_array() act just like execute().
4107
4108       The "ArrayTupleFetch" attribute can be used to specify a reference to a
4109       subroutine that will be called to provide the bind values for each
4110       tuple execution. The subroutine should return an reference to an array
4111       which contains the appropriate number of bind values, or return an
4112       undef if there is no more data to execute.
4113
4114       As a convenience, the "ArrayTupleFetch" attribute can also be used to
4115       specify a statement handle. In which case the fetchrow_arrayref()
4116       method will be called on the given statement handle in order to provide
4117       the bind values for each tuple execution.
4118
4119       The values specified via bind_param_array() or the @bind_values
4120       parameter may be either scalars, or arrayrefs.  If any @bind_values are
4121       given, then "execute_array" will effectively call "bind_param_array"
4122       for each value before executing the statement.  Values bound in this
4123       way are usually treated as "SQL_VARCHAR" types unless the driver can
4124       determine the correct type (which is rare), or unless "bind_param",
4125       "bind_param_inout", "bind_param_array", or "bind_param_inout_array" has
4126       already been used to specify the type.  See "bind_param_array" for
4127       details.
4128
4129       The "ArrayTupleStatus" attribute can be used to specify a reference to
4130       an array which will receive the execute status of each executed
4131       parameter tuple. Note the "ArrayTupleStatus" attribute was mandatory
4132       until DBI 1.38.
4133
4134       For tuples which are successfully executed, the element at the same
4135       ordinal position in the status array is the resulting rowcount (or -1
4136       if unknown).  If the execution of a tuple causes an error, then the
4137       corresponding status array element will be set to a reference to an
4138       array containing "err", "errstr" and "state" set by the failed
4139       execution.
4140
4141       If any tuple execution returns an error, "execute_array" will return
4142       "undef". In that case, the application should inspect the status array
4143       to determine which parameter tuples failed.  Some databases may not
4144       continue executing tuples beyond the first failure. In this case the
4145       status array will either hold fewer elements, or the elements beyond
4146       the failure will be undef.
4147
4148       If all parameter tuples are successfully executed, "execute_array"
4149       returns the number tuples executed.  If no tuples were executed, then
4150       execute_array() returns "0E0", just like execute() does, which Perl
4151       will treat as 0 but will regard as true.
4152
4153       For example:
4154
4155         $sth = $dbh->prepare("INSERT INTO staff (first_name, last_name) VALUES (?, ?)");
4156         my $tuples = $sth->execute_array(
4157             { ArrayTupleStatus => \my @tuple_status },
4158             \@first_names,
4159             \@last_names,
4160         );
4161         if ($tuples) {
4162             print "Successfully inserted $tuples records\n";
4163         }
4164         else {
4165             for my $tuple (0..@last_names-1) {
4166                 my $status = $tuple_status[$tuple];
4167                 $status = [0, "Skipped"] unless defined $status;
4168                 next unless ref $status;
4169                 printf "Failed to insert (%s, %s): %s\n",
4170                     $first_names[$tuple], $last_names[$tuple], $status->[1];
4171             }
4172         }
4173
4174       Support for data returning statements such as SELECT is driver-specific
4175       and subject to change. At present, the default implementation provided
4176       by DBI only supports non-data returning statements.
4177
4178       Transaction semantics when using array binding are driver and database
4179       specific.  If "AutoCommit" is on, the default DBI implementation will
4180       cause each parameter tuple to be individually committed (or rolled back
4181       in the event of an error). If "AutoCommit" is off, the application is
4182       responsible for explicitly committing the entire set of bound parameter
4183       tuples.  Note that different drivers and databases may have different
4184       behaviours when some parameter tuples cause failures. In some cases,
4185       the driver or database may automatically rollback the effect of all
4186       prior parameter tuples that succeeded in the transaction; other drivers
4187       or databases may retain the effect of prior successfully executed
4188       parameter tuples. Be sure to check your driver and database for its
4189       specific behaviour.
4190
4191       Note that, in general, performance will usually be better with
4192       "AutoCommit" turned off, and using explicit "commit" after each
4193       "execute_array" call.
4194
4195       The "execute_array" method was added in DBI 1.22, and ArrayTupleFetch
4196       was added in 1.36.
4197
4198       "execute_for_fetch"
4199
4200         $tuples = $sth->execute_for_fetch($fetch_tuple_sub);
4201         $tuples = $sth->execute_for_fetch($fetch_tuple_sub, \@tuple_status);
4202
4203         ($tuples, $rows) = $sth->execute_for_fetch($fetch_tuple_sub);
4204         ($tuples, $rows) = $sth->execute_for_fetch($fetch_tuple_sub, \@tuple_status);
4205
4206       The execute_for_fetch() method is used to perform bulk operations and
4207       although it is most often used via the execute_array() method you can
4208       use it directly. The main difference between execute_array and
4209       execute_for_fetch is the former does column or row-wise binding and the
4210       latter uses row-wise binding.
4211
4212       The fetch subroutine, referenced by $fetch_tuple_sub, is expected to
4213       return a reference to an array (known as a 'tuple') or undef.
4214
4215       The execute_for_fetch() method calls $fetch_tuple_sub, without any
4216       parameters, until it returns a false value. Each tuple returned is used
4217       to provide bind values for an $sth->execute(@$tuple) call.
4218
4219       In scalar context execute_for_fetch() returns "undef" if there were any
4220       errors and the number of tuples executed otherwise. Like execute() and
4221       execute_array() a zero is returned as "0E0" so execute_for_fetch() is
4222       only false on error.  If there were any errors the @tuple_status array
4223       can be used to discover which tuples failed and with what errors.
4224
4225       When called in list context execute_for_fetch() returns two scalars;
4226       $tuples is the same as calling execute_for_fetch() in scalar context
4227       and $rows is the sum of the number of rows affected for each tuple, if
4228       available or -1 if the driver cannot determine this.  If you are doing
4229       an update operation the returned rows affected may not be what you
4230       expect if, for instance, one or more of the tuples affected the same
4231       row multiple times.  Some drivers may not yet support list context, in
4232       which case $rows will be undef, or may not be able to provide the
4233       number of rows affected when performing this batch operation, in which
4234       case $rows will be -1.
4235
4236       If \@tuple_status is passed then the execute_for_fetch method uses it
4237       to return status information. The tuple_status array holds one element
4238       per tuple. If the corresponding execute() did not fail then the element
4239       holds the return value from execute(), which is typically a row count.
4240       If the execute() did fail then the element holds a reference to an
4241       array containing ($sth->err, $sth->errstr, $sth->state).
4242
4243       If the driver detects an error that it knows means no further tuples
4244       can be executed then it may return, with an error status, even though
4245       $fetch_tuple_sub may still have more tuples to be executed.
4246
4247       Although each tuple returned by $fetch_tuple_sub is effectively used to
4248       call $sth->execute(@$tuple_array_ref) the exact timing may vary.
4249       Drivers are free to accumulate sets of tuples to pass to the database
4250       server in bulk group operations for more efficient execution.  However,
4251       the $fetch_tuple_sub is specifically allowed to return the same array
4252       reference each time (which is what fetchrow_arrayref() usually does).
4253
4254       For example:
4255
4256         my $sel = $dbh1->prepare("select foo, bar from table1");
4257         $sel->execute;
4258
4259         my $ins = $dbh2->prepare("insert into table2 (foo, bar) values (?,?)");
4260         my $fetch_tuple_sub = sub { $sel->fetchrow_arrayref };
4261
4262         my @tuple_status;
4263         $rc = $ins->execute_for_fetch($fetch_tuple_sub, \@tuple_status);
4264         my @errors = grep { ref $_ } @tuple_status;
4265
4266       Similarly, if you already have an array containing the data rows to be
4267       processed you'd use a subroutine to shift off and return each array ref
4268       in turn:
4269
4270         $ins->execute_for_fetch( sub { shift @array_of_arrays }, \@tuple_status);
4271
4272       The "execute_for_fetch" method was added in DBI 1.38.
4273
4274       "fetchrow_arrayref"
4275
4276         $ary_ref = $sth->fetchrow_arrayref;
4277         $ary_ref = $sth->fetch;    # alias
4278
4279       Fetches the next row of data and returns a reference to an array
4280       holding the field values.  Null fields are returned as "undef" values
4281       in the array.  This is the fastest way to fetch data, particularly if
4282       used with "$sth->bind_columns".
4283
4284       If there are no more rows or if an error occurs, then
4285       "fetchrow_arrayref" returns an "undef". You should check "$sth->err"
4286       afterwards (or use the "RaiseError" attribute) to discover if the
4287       "undef" returned was due to an error.
4288
4289       Note that the same array reference is returned for each fetch, so don't
4290       store the reference and then use it after a later fetch.  Also, the
4291       elements of the array are also reused for each row, so take care if you
4292       want to take a reference to an element. See also "bind_columns".
4293
4294       "fetchrow_array"
4295
4296        @ary = $sth->fetchrow_array;
4297
4298       An alternative to "fetchrow_arrayref". Fetches the next row of data and
4299       returns it as a list containing the field values.  Null fields are
4300       returned as "undef" values in the list.
4301
4302       If there are no more rows or if an error occurs, then "fetchrow_array"
4303       returns an empty list. You should check "$sth->err" afterwards (or use
4304       the "RaiseError" attribute) to discover if the empty list returned was
4305       due to an error.
4306
4307       If called in a scalar context for a statement handle that has more than
4308       one column, it is undefined whether the driver will return the value of
4309       the first column or the last. So don't do that.  Also, in a scalar
4310       context, an "undef" is returned if there are no more rows or if an
4311       error occurred. That "undef" can't be distinguished from an "undef"
4312       returned because the first field value was NULL.  For these reasons you
4313       should exercise some caution if you use "fetchrow_array" in a scalar
4314       context.
4315
4316       "fetchrow_hashref"
4317
4318        $hash_ref = $sth->fetchrow_hashref;
4319        $hash_ref = $sth->fetchrow_hashref($name);
4320
4321       An alternative to "fetchrow_arrayref". Fetches the next row of data and
4322       returns it as a reference to a hash containing field name and field
4323       value pairs.  Null fields are returned as "undef" values in the hash.
4324
4325       If there are no more rows or if an error occurs, then
4326       "fetchrow_hashref" returns an "undef". You should check "$sth->err"
4327       afterwards (or use the "RaiseError" attribute) to discover if the
4328       "undef" returned was due to an error.
4329
4330       The optional $name parameter specifies the name of the statement handle
4331       attribute. For historical reasons it defaults to ""NAME"", however
4332       using either ""NAME_lc"" or ""NAME_uc"" is recommended for portability.
4333
4334       The keys of the hash are the same names returned by "$sth->{$name}". If
4335       more than one field has the same name, there will only be one entry in
4336       the returned hash for those fields, so statements like ""select foo,
4337       foo from bar"" will return only a single key from "fetchrow_hashref".
4338       In these cases use column aliases or "fetchrow_arrayref".  Note that it
4339       is the database server (and not the DBD implementation) which provides
4340       the name for fields containing functions like "count(*)" or
4341       ""max(c_foo)"" and they may clash with existing column names (most
4342       databases don't care about duplicate column names in a result-set). If
4343       you want these to return as unique names that are the same across
4344       databases, use aliases, as in ""select count(*) as cnt"" or ""select
4345       max(c_foo) mx_foo, ..."" depending on the syntax your database
4346       supports.
4347
4348       Because of the extra work "fetchrow_hashref" and Perl have to perform,
4349       it is not as efficient as "fetchrow_arrayref" or "fetchrow_array".
4350
4351       By default a reference to a new hash is returned for each row.  It is
4352       likely that a future version of the DBI will support an attribute which
4353       will enable the same hash to be reused for each row. This will give a
4354       significant performance boost, but it won't be enabled by default
4355       because of the risk of breaking old code.
4356
4357       "fetchall_arrayref"
4358
4359         $tbl_ary_ref = $sth->fetchall_arrayref;
4360         $tbl_ary_ref = $sth->fetchall_arrayref( $slice );
4361         $tbl_ary_ref = $sth->fetchall_arrayref( $slice, $max_rows  );
4362
4363       The "fetchall_arrayref" method can be used to fetch all the data to be
4364       returned from a prepared and executed statement handle. It returns a
4365       reference to an array that contains one reference per row.
4366
4367       If called on an inactive statement handle, "fetchall_arrayref" returns
4368       undef.
4369
4370       If there are no rows left to return from an active statement handle,
4371       "fetchall_arrayref" returns a reference to an empty array. If an error
4372       occurs, "fetchall_arrayref" returns the data fetched thus far, which
4373       may be none.  You should check "$sth->err" afterwards (or use the
4374       "RaiseError" attribute) to discover if the data is complete or was
4375       truncated due to an error.
4376
4377       If $slice is an array reference, "fetchall_arrayref" uses
4378       "fetchrow_arrayref" to fetch each row as an array ref. If the $slice
4379       array is not empty then it is used as a slice to select individual
4380       columns by perl array index number (starting at 0, unlike column and
4381       parameter numbers which start at 1).
4382
4383       With no parameters, or if $slice is undefined, "fetchall_arrayref" acts
4384       as if passed an empty array ref.
4385
4386       For example, to fetch just the first column of every row:
4387
4388         $tbl_ary_ref = $sth->fetchall_arrayref([0]);
4389
4390       To fetch the second to last and last column of every row:
4391
4392         $tbl_ary_ref = $sth->fetchall_arrayref([-2,-1]);
4393
4394       Those two examples both return a reference to an array of array refs.
4395
4396       If $slice is a hash reference, "fetchall_arrayref" fetches each row as
4397       a hash reference. If the $slice hash is empty then the keys in the
4398       hashes have whatever name lettercase is returned by default. (See
4399       "FetchHashKeyName" attribute.) If the $slice hash is not empty, then it
4400       is used as a slice to select individual columns by name. The values of
4401       the hash should be set to 1.  The key names of the returned hashes
4402       match the letter case of the names in the parameter hash, regardless of
4403       the "FetchHashKeyName" attribute.
4404
4405       For example, to fetch all fields of every row as a hash ref:
4406
4407         $tbl_ary_ref = $sth->fetchall_arrayref({});
4408
4409       To fetch only the fields called "foo" and "bar" of every row as a hash
4410       ref (with keys named "foo" and "BAR", regardless of the original
4411       capitalization):
4412
4413         $tbl_ary_ref = $sth->fetchall_arrayref({ foo=>1, BAR=>1 });
4414
4415       Those two examples both return a reference to an array of hash refs.
4416
4417       If $slice is a reference to a hash reference, that hash is used to
4418       select and rename columns. The keys are 0-based column index numbers
4419       and the values are the corresponding keys for the returned row hashes.
4420
4421       For example, to fetch only the first and second columns of every row as
4422       a hash ref (with keys named "k" and "v" regardless of their original
4423       names):
4424
4425         $tbl_ary_ref = $sth->fetchall_arrayref( \{ 0 => 'k', 1 => 'v' } );
4426
4427       If $max_rows is defined and greater than or equal to zero then it is
4428       used to limit the number of rows fetched before returning.
4429       fetchall_arrayref() can then be called again to fetch more rows.  This
4430       is especially useful when you need the better performance of
4431       fetchall_arrayref() but don't have enough memory to fetch and return
4432       all the rows in one go.
4433
4434       Here's an example (assumes RaiseError is enabled):
4435
4436         my $rows = []; # cache for batches of rows
4437         while( my $row = ( shift(@$rows) || # get row from cache, or reload cache:
4438                            shift(@{$rows=$sth->fetchall_arrayref(undef,10_000)||[]}) )
4439         ) {
4440           ...
4441         }
4442
4443       That might be the fastest way to fetch and process lots of rows using
4444       the DBI, but it depends on the relative cost of method calls vs memory
4445       allocation.
4446
4447       A standard "while" loop with column binding is often faster because the
4448       cost of allocating memory for the batch of rows is greater than the
4449       saving by reducing method calls. It's possible that the DBI may provide
4450       a way to reuse the memory of a previous batch in future, which would
4451       then shift the balance back towards fetchall_arrayref().
4452
4453       "fetchall_hashref"
4454
4455         $hash_ref = $sth->fetchall_hashref($key_field);
4456
4457       The "fetchall_hashref" method can be used to fetch all the data to be
4458       returned from a prepared and executed statement handle. It returns a
4459       reference to a hash containing a key for each distinct value of the
4460       $key_field column that was fetched. For each key the corresponding
4461       value is a reference to a hash containing all the selected columns and
4462       their values, as returned by "fetchrow_hashref()".
4463
4464       If there are no rows to return, "fetchall_hashref" returns a reference
4465       to an empty hash. If an error occurs, "fetchall_hashref" returns the
4466       data fetched thus far, which may be none.  You should check "$sth->err"
4467       afterwards (or use the "RaiseError" attribute) to discover if the data
4468       is complete or was truncated due to an error.
4469
4470       The $key_field parameter provides the name of the field that holds the
4471       value to be used for the key for the returned hash.  For example:
4472
4473         $dbh->{FetchHashKeyName} = 'NAME_lc';
4474         $sth = $dbh->prepare("SELECT FOO, BAR, ID, NAME, BAZ FROM TABLE");
4475         $sth->execute;
4476         $hash_ref = $sth->fetchall_hashref('id');
4477         print "Name for id 42 is $hash_ref->{42}->{name}\n";
4478
4479       The $key_field parameter can also be specified as an integer column
4480       number (counting from 1).  If $key_field doesn't match any column in
4481       the statement, as a name first then as a number, then an error is
4482       returned.
4483
4484       For queries returning more than one 'key' column, you can specify
4485       multiple column names by passing $key_field as a reference to an array
4486       containing one or more key column names (or index numbers).  For
4487       example:
4488
4489         $sth = $dbh->prepare("SELECT foo, bar, baz FROM table");
4490         $sth->execute;
4491         $hash_ref = $sth->fetchall_hashref( [ qw(foo bar) ] );
4492         print "For foo 42 and bar 38, baz is $hash_ref->{42}->{38}->{baz}\n";
4493
4494       The fetchall_hashref() method is normally used only where the key
4495       fields values for each row are unique.  If multiple rows are returned
4496       with the same values for the key fields then later rows overwrite
4497       earlier ones.
4498
4499       "finish"
4500
4501         $rc  = $sth->finish;
4502
4503       Indicate that no more data will be fetched from this statement handle
4504       before it is either executed again or destroyed.  You almost certainly
4505       do not need to call this method.
4506
4507       Adding calls to "finish" after loop that fetches all rows is a common
4508       mistake, don't do it, it can mask genuine problems like uncaught fetch
4509       errors.
4510
4511       When all the data has been fetched from a "SELECT" statement, the
4512       driver will automatically call "finish" for you. So you should not call
4513       it explicitly except when you know that you've not fetched all the data
4514       from a statement handle and the handle won't be destroyed soon.
4515
4516       The most common example is when you only want to fetch just one row,
4517       but in that case the "selectrow_*" methods are usually better anyway.
4518
4519       Consider a query like:
4520
4521         SELECT foo FROM table WHERE bar=? ORDER BY baz
4522
4523       on a very large table. When executed, the database server will have to
4524       use temporary buffer space to store the sorted rows. If, after
4525       executing the handle and selecting just a few rows, the handle won't be
4526       re-executed for some time and won't be destroyed, the "finish" method
4527       can be used to tell the server that the buffer space can be freed.
4528
4529       Calling "finish" resets the "Active" attribute for the statement.  It
4530       may also make some statement handle attributes (such as "NAME" and
4531       "TYPE") unavailable if they have not already been accessed (and thus
4532       cached).
4533
4534       The "finish" method does not affect the transaction status of the
4535       database connection.  It has nothing to do with transactions. It's
4536       mostly an internal "housekeeping" method that is rarely needed.  See
4537       also "disconnect" and the "Active" attribute.
4538
4539       The "finish" method should have been called "discard_pending_rows".
4540
4541       "rows"
4542
4543         $rv = $sth->rows;
4544
4545       Returns the number of rows affected by the last row affecting command,
4546       or -1 if the number of rows is not known or not available.
4547
4548       Generally, you can only rely on a row count after a non-"SELECT"
4549       "execute" (for some specific operations like "UPDATE" and "DELETE"), or
4550       after fetching all the rows of a "SELECT" statement.
4551
4552       For "SELECT" statements, it is generally not possible to know how many
4553       rows will be returned except by fetching them all.  Some drivers will
4554       return the number of rows the application has fetched so far, but
4555       others may return -1 until all rows have been fetched.  So use of the
4556       "rows" method or $DBI::rows with "SELECT" statements is not
4557       recommended.
4558
4559       One alternative method to get a row count for a "SELECT" is to execute
4560       a "SELECT COUNT(*) FROM ..." SQL statement with the same "..." as your
4561       query and then fetch the row count from that.
4562
4563       "bind_col"
4564
4565         $rc = $sth->bind_col($column_number, \$var_to_bind);
4566         $rc = $sth->bind_col($column_number, \$var_to_bind, \%attr );
4567         $rc = $sth->bind_col($column_number, \$var_to_bind, $bind_type );
4568
4569       Binds a Perl variable and/or some attributes to an output column
4570       (field) of a "SELECT" statement.  Column numbers count up from 1.  You
4571       do not need to bind output columns in order to fetch data.  For maximum
4572       portability between drivers, bind_col() should be called after
4573       execute() and not before.  See also "bind_columns" for an example.
4574
4575       The binding is performed at a low level using Perl aliasing.  Whenever
4576       a row is fetched from the database $var_to_bind appears to be
4577       automatically updated simply because it now refers to the same memory
4578       location as the corresponding column value.  This makes using bound
4579       variables very efficient.  Binding a tied variable doesn't work,
4580       currently.
4581
4582       The "bind_param" method performs a similar, but opposite, function for
4583       input variables.
4584
4585       Data Types for Column Binding
4586
4587       The "\%attr" parameter can be used to hint at the data type formatting
4588       the column should have. For example, you can use:
4589
4590         $sth->bind_col(1, undef, { TYPE => SQL_DATETIME });
4591
4592       to specify that you'd like the column (which presumably is some kind of
4593       datetime type) to be returned in the standard format for SQL_DATETIME,
4594       which is 'YYYY-MM-DD HH:MM:SS', rather than the native formatting the
4595       database would normally use.
4596
4597       There's no $var_to_bind in that example to emphasize the point that
4598       bind_col() works on the underlying column and not just a particular
4599       bound variable.
4600
4601       As a short-cut for the common case, the data type can be passed
4602       directly, in place of the "\%attr" hash reference. This example is
4603       equivalent to the one above:
4604
4605         $sth->bind_col(1, undef, SQL_DATETIME);
4606
4607       The "TYPE" value indicates the standard (non-driver-specific) type for
4608       this parameter. To specify the driver-specific type, the driver may
4609       support a driver-specific attribute, such as "{ ora_type => 97 }".
4610
4611       The SQL_DATETIME and other related constants can be imported using
4612
4613         use DBI qw(:sql_types);
4614
4615       See "DBI Constants" for more information.
4616
4617       Few drivers support specifying a data type via a "bind_col" call (most
4618       will simply ignore the data type). Fewer still allow the data type to
4619       be altered once set.
4620
4621       The TYPE attribute for bind_col() was first specified in DBI 1.41.
4622
4623       From DBI 1.611, drivers can use the "TYPE" attribute to attempt to cast
4624       the bound scalar to a perl type which more closely matches "TYPE". At
4625       present DBI supports "SQL_INTEGER", "SQL_DOUBLE" and "SQL_NUMERIC". See
4626       "sql_type_cast" for details of how types are cast.
4627
4628       Other attributes for Column Binding
4629
4630       The "\%attr" parameter may also contain the following attributes:
4631
4632       "StrictlyTyped"
4633           If a "TYPE" attribute is passed to bind_col, then the driver will
4634           attempt to change the bound perl scalar to match the type more
4635           closely. If the bound value cannot be cast to the requested "TYPE"
4636           then by default it is left untouched and no error is generated. If
4637           you specify "StrictlyTyped" as 1 and the cast fails, this will
4638           generate an error.
4639
4640           This attribute was first added in DBI 1.611. When 1.611 was
4641           released few drivers actually supported this attribute but
4642           DBD::Oracle and DBD::ODBC should from versions 1.24.
4643
4644       "DiscardString"
4645           When the "TYPE" attribute is passed to "bind_col" and the driver
4646           successfully casts the bound perl scalar to a non-string type then
4647           if "DiscardString" is set to 1, the string portion of the scalar
4648           will be discarded. By default, "DiscardString" is not set.
4649
4650           This attribute was first added in DBI 1.611. When 1.611 was
4651           released few drivers actually supported this attribute but
4652           DBD::Oracle and DBD::ODBC should from versions 1.24.
4653
4654       "bind_columns"
4655
4656         $rc = $sth->bind_columns(@list_of_refs_to_vars_to_bind);
4657
4658       Calls "bind_col" for each column of the "SELECT" statement.
4659
4660       The list of references should have the same number of elements as the
4661       number of columns in the "SELECT" statement. If it doesn't then
4662       "bind_columns" will bind the elements given, up to the number of
4663       columns, and then return an error.
4664
4665       For maximum portability between drivers, bind_columns() should be
4666       called after execute() and not before.
4667
4668       For example:
4669
4670         $dbh->{RaiseError} = 1; # do this, or check every call for errors
4671         $sth = $dbh->prepare(q{ SELECT region, sales FROM sales_by_region });
4672         $sth->execute;
4673         my ($region, $sales);
4674
4675         # Bind Perl variables to columns:
4676         $rv = $sth->bind_columns(\$region, \$sales);
4677
4678         # you can also use Perl's \(...) syntax (see perlref docs):
4679         #     $sth->bind_columns(\($region, $sales));
4680
4681         # Column binding is the most efficient way to fetch data
4682         while ($sth->fetch) {
4683             print "$region: $sales\n";
4684         }
4685
4686       For compatibility with old scripts, the first parameter will be ignored
4687       if it is "undef" or a hash reference.
4688
4689       Here's a more fancy example that binds columns to the values inside a
4690       hash (thanks to H.Merijn Brand):
4691
4692         $sth->execute;
4693         my %row;
4694         $sth->bind_columns( \( @row{ @{$sth->{NAME_lc} } } ));
4695         while ($sth->fetch) {
4696             print "$row{region}: $row{sales}\n";
4697         }
4698
4699       "dump_results"
4700
4701         $rows = $sth->dump_results($maxlen, $lsep, $fsep, $fh);
4702
4703       Fetches all the rows from $sth, calls "DBI::neat_list" for each row,
4704       and prints the results to $fh (defaults to "STDOUT") separated by $lsep
4705       (default "\n"). $fsep defaults to ", " and $maxlen defaults to 35.
4706
4707       This method is designed as a handy utility for prototyping and testing
4708       queries. Since it uses "neat_list" to format and edit the string for
4709       reading by humans, it is not recommended for data transfer
4710       applications.
4711
4712   Statement Handle Attributes
4713       This section describes attributes specific to statement handles. Most
4714       of these attributes are read-only.
4715
4716       Changes to these statement handle attributes do not affect any other
4717       existing or future statement handles.
4718
4719       Attempting to set or get the value of an unknown attribute generates a
4720       warning, except for private driver specific attributes (which all have
4721       names starting with a lowercase letter).
4722
4723       Example:
4724
4725         ... = $h->{NUM_OF_FIELDS};    # get/read
4726
4727       Some drivers cannot provide valid values for some or all of these
4728       attributes until after "$sth->execute" has been successfully called.
4729       Typically the attribute will be "undef" in these situations.
4730
4731       Some attributes, like NAME, are not appropriate to some types of
4732       statement, like SELECT. Typically the attribute will be "undef" in
4733       these situations.
4734
4735       For drivers which support stored procedures and multiple result sets
4736       (see "more_results") these attributes relate to the current result set.
4737
4738       See also "finish" to learn more about the effect it may have on some
4739       attributes.
4740
4741       "NUM_OF_FIELDS"
4742
4743       Type: integer, read-only
4744
4745       Number of fields (columns) in the data the prepared statement may
4746       return.  Statements that don't return rows of data, like "DELETE" and
4747       "CREATE" set "NUM_OF_FIELDS" to 0 (though it may be undef in some
4748       drivers).
4749
4750       "NUM_OF_PARAMS"
4751
4752       Type: integer, read-only
4753
4754       The number of parameters (placeholders) in the prepared statement.  See
4755       SUBSTITUTION VARIABLES below for more details.
4756
4757       "NAME"
4758
4759       Type: array-ref, read-only
4760
4761       Returns a reference to an array of field names for each column. The
4762       names may contain spaces but should not be truncated or have any
4763       trailing space. Note that the names have the letter case (upper, lower
4764       or mixed) as returned by the driver being used. Portable applications
4765       should use "NAME_lc" or "NAME_uc".
4766
4767         print "First column name: $sth->{NAME}->[0]\n";
4768
4769       Also note that the name returned for (aggregate) functions like
4770       count(*) or "max(c_foo)" is determined by the database server and not
4771       by "DBI" or the "DBD" backend.
4772
4773       "NAME_lc"
4774
4775       Type: array-ref, read-only
4776
4777       Like "/NAME" but always returns lowercase names.
4778
4779       "NAME_uc"
4780
4781       Type: array-ref, read-only
4782
4783       Like "/NAME" but always returns uppercase names.
4784
4785       "NAME_hash"
4786
4787       Type: hash-ref, read-only
4788
4789       "NAME_lc_hash"
4790
4791       Type: hash-ref, read-only
4792
4793       "NAME_uc_hash"
4794
4795       Type: hash-ref, read-only
4796
4797       The "NAME_hash", "NAME_lc_hash", and "NAME_uc_hash" attributes return
4798       column name information as a reference to a hash.
4799
4800       The keys of the hash are the names of the columns.  The letter case of
4801       the keys corresponds to the letter case returned by the "NAME",
4802       "NAME_lc", and "NAME_uc" attributes respectively (as described above).
4803
4804       The value of each hash entry is the perl index number of the
4805       corresponding column (counting from 0). For example:
4806
4807         $sth = $dbh->prepare("select Id, Name from table");
4808         $sth->execute;
4809         @row = $sth->fetchrow_array;
4810         print "Name $row[ $sth->{NAME_lc_hash}{name} ]\n";
4811
4812       "TYPE"
4813
4814       Type: array-ref, read-only
4815
4816       Returns a reference to an array of integer values for each column. The
4817       value indicates the data type of the corresponding column.
4818
4819       The values correspond to the international standards (ANSI X3.135 and
4820       ISO/IEC 9075) which, in general terms, means ODBC. Driver-specific
4821       types that don't exactly match standard types should generally return
4822       the same values as an ODBC driver supplied by the makers of the
4823       database. That might include private type numbers in ranges the vendor
4824       has officially registered with the ISO working group:
4825
4826         ftp://sqlstandards.org/SC32/SQL_Registry/
4827
4828       Where there's no vendor-supplied ODBC driver to be compatible with, the
4829       DBI driver can use type numbers in the range that is now officially
4830       reserved for use by the DBI: -9999 to -9000.
4831
4832       All possible values for "TYPE" should have at least one entry in the
4833       output of the "type_info_all" method (see "type_info_all").
4834
4835       "PRECISION"
4836
4837       Type: array-ref, read-only
4838
4839       Returns a reference to an array of integer values for each column.
4840
4841       For numeric columns, the value is the maximum number of digits (without
4842       considering a sign character or decimal point). Note that the "display
4843       size" for floating point types (REAL, FLOAT, DOUBLE) can be up to 7
4844       characters greater than the precision (for the sign + decimal point +
4845       the letter E + a sign + 2 or 3 digits).
4846
4847       For any character type column the value is the OCTET_LENGTH, in other
4848       words the number of bytes, not characters.
4849
4850       (More recent standards refer to this as COLUMN_SIZE but we stick with
4851       PRECISION for backwards compatibility.)
4852
4853       "SCALE"
4854
4855       Type: array-ref, read-only
4856
4857       Returns a reference to an array of integer values for each column.
4858       NULL ("undef") values indicate columns where scale is not applicable.
4859
4860       "NULLABLE"
4861
4862       Type: array-ref, read-only
4863
4864       Returns a reference to an array indicating the possibility of each
4865       column returning a null.  Possible values are 0 (or an empty string) =
4866       no, 1 = yes, 2 = unknown.
4867
4868         print "First column may return NULL\n" if $sth->{NULLABLE}->[0];
4869
4870       "CursorName"
4871
4872       Type: string, read-only
4873
4874       Returns the name of the cursor associated with the statement handle, if
4875       available. If not available or if the database driver does not support
4876       the "where current of ..." SQL syntax, then it returns "undef".
4877
4878       "Database"
4879
4880       Type: dbh, read-only
4881
4882       Returns the parent $dbh of the statement handle.
4883
4884       "Statement"
4885
4886       Type: string, read-only
4887
4888       Returns the statement string passed to the "prepare" method.
4889
4890       "ParamValues"
4891
4892       Type: hash ref, read-only
4893
4894       Returns a reference to a hash containing the values currently bound to
4895       placeholders.  The keys of the hash are the 'names' of the
4896       placeholders, typically integers starting at 1.  Returns undef if not
4897       supported by the driver.
4898
4899       See "ShowErrorStatement" for an example of how this is used.
4900
4901       * Keys:
4902
4903       If the driver supports "ParamValues" but no values have been bound yet
4904       then the driver should return a hash with placeholders names in the
4905       keys but all the values undef, but some drivers may return a ref to an
4906       empty hash because they can't pre-determine the names.
4907
4908       It is possible that the keys in the hash returned by "ParamValues" are
4909       not exactly the same as those implied by the prepared statement.  For
4910       example, DBD::Oracle translates '"?"' placeholders into '":pN"' where N
4911       is a sequence number starting at 1.
4912
4913       * Values:
4914
4915       It is possible that the values in the hash returned by "ParamValues"
4916       are not exactly the same as those passed to bind_param() or execute().
4917       The driver may have slightly modified values in some way based on the
4918       TYPE the value was bound with. For example a floating point value bound
4919       as an SQL_INTEGER type may be returned as an integer.  The values
4920       returned by "ParamValues" can be passed to another bind_param() method
4921       with the same TYPE and will be seen by the database as the same value.
4922       See also "ParamTypes" below.
4923
4924       The "ParamValues" attribute was added in DBI 1.28.
4925
4926       "ParamTypes"
4927
4928       Type: hash ref, read-only
4929
4930       Returns a reference to a hash containing the type information currently
4931       bound to placeholders.  Returns undef if not supported by the driver.
4932
4933       * Keys:
4934
4935       See "ParamValues" above.
4936
4937       * Values:
4938
4939       The hash values are hashrefs of type information in the same form as
4940       that passed to the various bind_param() methods (See "bind_param" for
4941       the format and values).
4942
4943       It is possible that the values in the hash returned by "ParamTypes" are
4944       not exactly the same as those passed to bind_param() or execute().
4945       Param attributes specified using the abbreviated form, like this:
4946
4947           $sth->bind_param(1, SQL_INTEGER);
4948
4949       are returned in the expanded form, as if called like this:
4950
4951           $sth->bind_param(1, { TYPE => SQL_INTEGER });
4952
4953       The driver may have modified the type information in some way based on
4954       the bound values, other hints provided by the prepare()'d SQL
4955       statement, or alternate type mappings required by the driver or target
4956       database system. The driver may also add private keys (with names
4957       beginning with the drivers reserved prefix, e.g., odbc_xxx).
4958
4959       * Example:
4960
4961       The keys and values in the returned hash can be passed to the various
4962       bind_param() methods to effectively reproduce a previous param binding.
4963       For example:
4964
4965         # assuming $sth1 is a previously prepared statement handle
4966         my $sth2 = $dbh->prepare( $sth1->{Statement} );
4967         my $ParamValues = $sth1->{ParamValues} || {};
4968         my $ParamTypes  = $sth1->{ParamTypes}  || {};
4969         $sth2->bind_param($_, $ParamValues->{$_} $ParamTypes->{$_})
4970           for keys %{ {%$ParamValues, %$ParamTypes} };
4971         $sth2->execute();
4972
4973       The "ParamTypes" attribute was added in DBI 1.49. Implementation is the
4974       responsibility of individual drivers; the DBI layer default
4975       implementation simply returns undef.
4976
4977       "ParamArrays"
4978
4979       Type: hash ref, read-only
4980
4981       Returns a reference to a hash containing the values currently bound to
4982       placeholders with "execute_array" or "bind_param_array".  The keys of
4983       the hash are the 'names' of the placeholders, typically integers
4984       starting at 1.  Returns undef if not supported by the driver or no
4985       arrays of parameters are bound.
4986
4987       Each key value is an array reference containing a list of the bound
4988       parameters for that column.
4989
4990       For example:
4991
4992         $sth = $dbh->prepare("INSERT INTO staff (id, name) values (?,?)");
4993         $sth->execute_array({},[1,2], ['fred','dave']);
4994         if ($sth->{ParamArrays}) {
4995             foreach $param (keys %{$sth->{ParamArrays}}) {
4996                 printf "Parameters for %s : %s\n", $param,
4997                 join(",", @{$sth->{ParamArrays}->{$param}});
4998             }
4999         }
5000
5001       It is possible that the values in the hash returned by "ParamArrays"
5002       are not exactly the same as those passed to "bind_param_array" or
5003       "execute_array".  The driver may have slightly modified values in some
5004       way based on the TYPE the value was bound with. For example a floating
5005       point value bound as an SQL_INTEGER type may be returned as an integer.
5006
5007       It is also possible that the keys in the hash returned by "ParamArrays"
5008       are not exactly the same as those implied by the prepared statement.
5009       For example, DBD::Oracle translates '"?"'  placeholders into '":pN"'
5010       where N is a sequence number starting at 1.
5011
5012       "RowsInCache"
5013
5014       Type: integer, read-only
5015
5016       If the driver supports a local row cache for "SELECT" statements, then
5017       this attribute holds the number of un-fetched rows in the cache. If the
5018       driver doesn't, then it returns "undef". Note that some drivers pre-
5019       fetch rows on execute, whereas others wait till the first fetch.
5020
5021       See also the "RowCacheSize" database handle attribute.
5022

FURTHER INFORMATION

5024   Catalog Methods
5025       An application can retrieve metadata information from the DBMS by
5026       issuing appropriate queries on the views of the Information Schema.
5027       Unfortunately, "INFORMATION_SCHEMA" views are seldom supported by the
5028       DBMS.  Special methods (catalog methods) are available to return result
5029       sets for a small but important portion of that metadata:
5030
5031         column_info
5032         foreign_key_info
5033         primary_key_info
5034         table_info
5035         statistics_info
5036
5037       All catalog methods accept arguments in order to restrict the result
5038       sets.  Passing "undef" to an optional argument does not constrain the
5039       search for that argument.  However, an empty string ('') is treated as
5040       a regular search criteria and will only match an empty value.
5041
5042       Note: SQL/CLI and ODBC differ in the handling of empty strings. An
5043       empty string will not restrict the result set in SQL/CLI.
5044
5045       Most arguments in the catalog methods accept only ordinary values, e.g.
5046       the arguments of "primary_key_info()".  Such arguments are treated as a
5047       literal string, i.e. the case is significant and quote characters are
5048       taken literally.
5049
5050       Some arguments in the catalog methods accept search patterns (strings
5051       containing '_' and/or '%'), e.g. the $table argument of
5052       "column_info()".  Passing '%' is equivalent to leaving the argument
5053       "undef".
5054
5055       Caveat: The underscore ('_') is valid and often used in SQL
5056       identifiers.  Passing such a value to a search pattern argument may
5057       return more rows than expected!  To include pattern characters as
5058       literals, they must be preceded by an escape character which can be
5059       achieved with
5060
5061         $esc = $dbh->get_info( 14 );  # SQL_SEARCH_PATTERN_ESCAPE
5062         $search_pattern =~ s/([_%])/$esc$1/g;
5063
5064       The ODBC and SQL/CLI specifications define a way to change the default
5065       behaviour described above: All arguments (except list value arguments)
5066       are treated as identifier if the "SQL_ATTR_METADATA_ID" attribute is
5067       set to "SQL_TRUE".  Quoted identifiers are very similar to ordinary
5068       values, i.e. their body (the string within the quotes) is interpreted
5069       literally.  Unquoted identifiers are compared in UPPERCASE.
5070
5071       The DBI (currently) does not support the "SQL_ATTR_METADATA_ID"
5072       attribute, i.e. it behaves like an ODBC driver where
5073       "SQL_ATTR_METADATA_ID" is set to "SQL_FALSE".
5074
5075   Transactions
5076       Transactions are a fundamental part of any robust database system. They
5077       protect against errors and database corruption by ensuring that sets of
5078       related changes to the database take place in atomic (indivisible, all-
5079       or-nothing) units.
5080
5081       This section applies to databases that support transactions and where
5082       "AutoCommit" is off.  See "AutoCommit" for details of using
5083       "AutoCommit" with various types of databases.
5084
5085       The recommended way to implement robust transactions in Perl
5086       applications is to use "RaiseError" and "eval { ... }" (which is very
5087       fast, unlike "eval "...""). For example:
5088
5089         $dbh->{AutoCommit} = 0;  # enable transactions, if possible
5090         $dbh->{RaiseError} = 1;
5091         eval {
5092             foo(...)        # do lots of work here
5093             bar(...)        # including inserts
5094             baz(...)        # and updates
5095             $dbh->commit;   # commit the changes if we get this far
5096         };
5097         if ($@) {
5098             warn "Transaction aborted because $@";
5099             # now rollback to undo the incomplete changes
5100             # but do it in an eval{} as it may also fail
5101             eval { $dbh->rollback };
5102             # add other application on-error-clean-up code here
5103         }
5104
5105       If the "RaiseError" attribute is not set, then DBI calls would need to
5106       be manually checked for errors, typically like this:
5107
5108         $h->method(@args) or die $h->errstr;
5109
5110       With "RaiseError" set, the DBI will automatically "die" if any DBI
5111       method call on that handle (or a child handle) fails, so you don't have
5112       to test the return value of each method call. See "RaiseError" for more
5113       details.
5114
5115       A major advantage of the "eval" approach is that the transaction will
5116       be properly rolled back if any code (not just DBI calls) in the inner
5117       application dies for any reason. The major advantage of using the
5118       "$h->{RaiseError}" attribute is that all DBI calls will be checked
5119       automatically. Both techniques are strongly recommended.
5120
5121       After calling "commit" or "rollback" many drivers will not let you
5122       fetch from a previously active "SELECT" statement handle that's a child
5123       of the same database handle. A typical way round this is to connect the
5124       the database twice and use one connection for "SELECT" statements.
5125
5126       See "AutoCommit" and "disconnect" for other important information about
5127       transactions.
5128
5129   Handling BLOB / LONG / Memo Fields
5130       Many databases support "blob" (binary large objects), "long", or
5131       similar datatypes for holding very long strings or large amounts of
5132       binary data in a single field. Some databases support variable length
5133       long values over 2,000,000,000 bytes in length.
5134
5135       Since values of that size can't usually be held in memory, and because
5136       databases can't usually know in advance the length of the longest long
5137       that will be returned from a "SELECT" statement (unlike other data
5138       types), some special handling is required.
5139
5140       In this situation, the value of the "$h->{LongReadLen}" attribute is
5141       used to determine how much buffer space to allocate when fetching such
5142       fields.  The "$h->{LongTruncOk}" attribute is used to determine how to
5143       behave if a fetched value can't fit into the buffer.
5144
5145       See the description of "LongReadLen" for more information.
5146
5147       When trying to insert long or binary values, placeholders should be
5148       used since there are often limits on the maximum size of an "INSERT"
5149       statement and the "quote" method generally can't cope with binary data.
5150       See "Placeholders and Bind Values".
5151
5152   Simple Examples
5153       Here's a complete example program to select and fetch some data:
5154
5155         my $data_source = "dbi::DriverName:db_name";
5156         my $dbh = DBI->connect($data_source, $user, $password)
5157             or die "Can't connect to $data_source: $DBI::errstr";
5158
5159         my $sth = $dbh->prepare( q{
5160                 SELECT name, phone
5161                 FROM mytelbook
5162         }) or die "Can't prepare statement: $DBI::errstr";
5163
5164         my $rc = $sth->execute
5165             or die "Can't execute statement: $DBI::errstr";
5166
5167         print "Query will return $sth->{NUM_OF_FIELDS} fields.\n\n";
5168         print "Field names: @{ $sth->{NAME} }\n";
5169
5170         while (($name, $phone) = $sth->fetchrow_array) {
5171             print "$name: $phone\n";
5172         }
5173         # check for problems which may have terminated the fetch early
5174         die $sth->errstr if $sth->err;
5175
5176         $dbh->disconnect;
5177
5178       Here's a complete example program to insert some data from a file.
5179       (This example uses "RaiseError" to avoid needing to check each call).
5180
5181         my $dbh = DBI->connect("dbi:DriverName:db_name", $user, $password, {
5182             RaiseError => 1, AutoCommit => 0
5183         });
5184
5185         my $sth = $dbh->prepare( q{
5186             INSERT INTO table (name, phone) VALUES (?, ?)
5187         });
5188
5189         open FH, "<phone.csv" or die "Unable to open phone.csv: $!";
5190         while (<FH>) {
5191             chomp;
5192             my ($name, $phone) = split /,/;
5193             $sth->execute($name, $phone);
5194         }
5195         close FH;
5196
5197         $dbh->commit;
5198         $dbh->disconnect;
5199
5200       Here's how to convert fetched NULLs (undefined values) into empty
5201       strings:
5202
5203         while($row = $sth->fetchrow_arrayref) {
5204           # this is a fast and simple way to deal with nulls:
5205           foreach (@$row) { $_ = '' unless defined }
5206           print "@$row\n";
5207         }
5208
5209       The "q{...}" style quoting used in these examples avoids clashing with
5210       quotes that may be used in the SQL statement. Use the double-quote like
5211       "qq{...}" operator if you want to interpolate variables into the
5212       string.  See "Quote and Quote-like Operators" in perlop for more
5213       details.
5214
5215   Threads and Thread Safety
5216       Perl 5.7 and later support a new threading model called iThreads.  (The
5217       old "5.005 style" threads are not supported by the DBI.)
5218
5219       In the iThreads model each thread has its own copy of the perl
5220       interpreter.  When a new thread is created the original perl
5221       interpreter is 'cloned' to create a new copy for the new thread.
5222
5223       If the DBI and drivers are loaded and handles created before the thread
5224       is created then it will get a cloned copy of the DBI, the drivers and
5225       the handles.
5226
5227       However, the internal pointer data within the handles will refer to the
5228       DBI and drivers in the original interpreter. Using those handles in the
5229       new interpreter thread is not safe, so the DBI detects this and croaks
5230       on any method call using handles that don't belong to the current
5231       thread (except for DESTROY).
5232
5233       Because of this (possibly temporary) restriction, newly created threads
5234       must make their own connections to the database. Handles can't be
5235       shared across threads.
5236
5237       But BEWARE, some underlying database APIs (the code the DBD driver uses
5238       to talk to the database, often supplied by the database vendor) are not
5239       thread safe. If it's not thread safe, then allowing more than one
5240       thread to enter the code at the same time may cause subtle/serious
5241       problems. In some cases allowing more than one thread to enter the
5242       code, even if not at the same time, can cause problems. You have been
5243       warned.
5244
5245       Using DBI with perl threads is not yet recommended for production
5246       environments. For more information see
5247       <http://www.perlmonks.org/index.pl?node_id=288022>
5248
5249       Note: There is a bug in perl 5.8.2 when configured with threads and
5250       debugging enabled (bug #24463) which causes a DBI test to fail.
5251
5252   Signal Handling and Canceling Operations
5253       [The following only applies to systems with unix-like signal handling.
5254       I'd welcome additions for other systems, especially Windows.]
5255
5256       The first thing to say is that signal handling in Perl versions less
5257       than 5.8 is not safe. There is always a small risk of Perl crashing
5258       and/or core dumping when, or after, handling a signal because the
5259       signal could arrive and be handled while internal data structures are
5260       being changed. If the signal handling code used those same internal
5261       data structures it could cause all manner of subtle and not-so-subtle
5262       problems.  The risk was reduced with 5.4.4 but was still present in all
5263       perls up through 5.8.0.
5264
5265       Beginning in perl 5.8.0 perl implements 'safe' signal handling if your
5266       system has the POSIX sigaction() routine. Now when a signal is
5267       delivered perl just makes a note of it but does not run the %SIG
5268       handler. The handling is 'deferred' until a 'safe' moment.
5269
5270       Although this change made signal handling safe, it also lead to a
5271       problem with signals being deferred for longer than you'd like.  If a
5272       signal arrived while executing a system call, such as waiting for data
5273       on a network connection, the signal is noted and then the system call
5274       that was executing returns with an EINTR error code to indicate that it
5275       was interrupted. All fine so far.
5276
5277       The problem comes when the code that made the system call sees the
5278       EINTR code and decides it's going to call it again. Perl doesn't do
5279       that, but database code sometimes does. If that happens then the signal
5280       handler doesn't get called until later. Maybe much later.
5281
5282       Fortunately there are ways around this which we'll discuss below.
5283       Unfortunately they make signals unsafe again.
5284
5285       The two most common uses of signals in relation to the DBI are for
5286       canceling operations when the user types Ctrl-C (interrupt), and for
5287       implementing a timeout using "alarm()" and $SIG{ALRM}.
5288
5289       Cancel
5290           The DBI provides a "cancel" method for statement handles. The
5291           "cancel" method should abort the current operation and is designed
5292           to be called from a signal handler.  For example:
5293
5294             $SIG{INT} = sub { $sth->cancel };
5295
5296           However, few drivers implement this (the DBI provides a default
5297           method that just returns "undef") and, even if implemented, there
5298           is still a possibility that the statement handle, and even the
5299           parent database handle, will not be usable afterwards.
5300
5301           If "cancel" returns true, then it has successfully invoked the
5302           database engine's own cancel function.  If it returns false, then
5303           "cancel" failed. If it returns "undef", then the database driver
5304           does not have cancel implemented - very few do.
5305
5306       Timeout
5307           The traditional way to implement a timeout is to set $SIG{ALRM} to
5308           refer to some code that will be executed when an ALRM signal
5309           arrives and then to call alarm($seconds) to schedule an ALRM signal
5310           to be delivered $seconds in the future. For example:
5311
5312             eval {
5313               local $SIG{ALRM} = sub { die "TIMEOUT\n" }; # N.B. \n required
5314               eval {
5315                 alarm($seconds);
5316                 ... code to execute with timeout here (which may die) ...
5317               };
5318               # outer eval catches alarm that might fire JUST before this alarm(0)
5319               alarm(0);  # cancel alarm (if code ran fast)
5320               die "$@" if $@;
5321             };
5322             if ( $@ eq "TIMEOUT\n" ) { ... }
5323             elsif ($@) { ... } # some other error
5324
5325           The first (outer) eval is used to avoid the unlikely but possible
5326           chance that the "code to execute" dies and the alarm fires before
5327           it is cancelled. Without the outer eval, if this happened your
5328           program will die if you have no ALRM handler or a non-local alarm
5329           handler will be called.
5330
5331           Unfortunately, as described above, this won't always work as
5332           expected, depending on your perl version and the underlying
5333           database code.
5334
5335           With Oracle for instance (DBD::Oracle), if the system which hosts
5336           the database is down the DBI->connect() call will hang for several
5337           minutes before returning an error.
5338
5339       The solution on these systems is to use the "POSIX::sigaction()"
5340       routine to gain low level access to how the signal handler is
5341       installed.
5342
5343       The code would look something like this (for the DBD-Oracle connect()):
5344
5345          use POSIX qw(:signal_h);
5346
5347          my $mask = POSIX::SigSet->new( SIGALRM ); # signals to mask in the handler
5348          my $action = POSIX::SigAction->new(
5349              sub { die "connect timeout\n" },        # the handler code ref
5350              $mask,
5351              # not using (perl 5.8.2 and later) 'safe' switch or sa_flags
5352          );
5353          my $oldaction = POSIX::SigAction->new();
5354          sigaction( SIGALRM, $action, $oldaction );
5355          my $dbh;
5356          eval {
5357             eval {
5358               alarm(5); # seconds before time out
5359               $dbh = DBI->connect("dbi:Oracle:$dsn" ... );
5360             };
5361             alarm(0); # cancel alarm (if connect worked fast)
5362             die "$@\n" if $@; # connect died
5363          };
5364          sigaction( SIGALRM, $oldaction );  # restore original signal handler
5365          if ( $@ ) {
5366            if ($@ eq "connect timeout\n") {...}
5367            else { # connect died }
5368          }
5369
5370       See previous example for the reasoning around the double eval.
5371
5372       Similar techniques can be used for canceling statement execution.
5373
5374       Unfortunately, this solution is somewhat messy, and it does not work
5375       with perl versions less than perl 5.8 where "POSIX::sigaction()"
5376       appears to be broken.
5377
5378       For a cleaner implementation that works across perl versions, see
5379       Lincoln Baxter's Sys::SigAction module at
5380       <http://search.cpan.org/~lbaxter/Sys-SigAction/>.  The documentation
5381       for Sys::SigAction includes an longer discussion of this problem, and a
5382       DBD::Oracle test script.
5383
5384       Be sure to read all the signal handling sections of the perlipc manual.
5385
5386       And finally, two more points to keep firmly in mind. Firstly, remember
5387       that what we've done here is essentially revert to old style unsafe
5388       handling of these signals. So do as little as possible in the handler.
5389       Ideally just die(). Secondly, the handles in use at the time the signal
5390       is handled may not be safe to use afterwards.
5391
5392   Subclassing the DBI
5393       DBI can be subclassed and extended just like any other object oriented
5394       module.  Before we talk about how to do that, it's important to be
5395       clear about the various DBI classes and how they work together.
5396
5397       By default "$dbh = DBI->connect(...)" returns a $dbh blessed into the
5398       "DBI::db" class.  And the "$dbh->prepare" method returns an $sth
5399       blessed into the "DBI::st" class (actually it simply changes the last
5400       four characters of the calling handle class to be "::st").
5401
5402       The leading '"DBI"' is known as the 'root class' and the extra '"::db"'
5403       or '"::st"' are the 'handle type suffixes'. If you want to subclass the
5404       DBI you'll need to put your overriding methods into the appropriate
5405       classes.  For example, if you want to use a root class of "MySubDBI"
5406       and override the do(), prepare() and execute() methods, then your do()
5407       and prepare() methods should be in the "MySubDBI::db" class and the
5408       execute() method should be in the "MySubDBI::st" class.
5409
5410       To setup the inheritance hierarchy the @ISA variable in "MySubDBI::db"
5411       should include "DBI::db" and the @ISA variable in "MySubDBI::st" should
5412       include "DBI::st".  The "MySubDBI" root class itself isn't currently
5413       used for anything visible and so, apart from setting @ISA to include
5414       "DBI", it can be left empty.
5415
5416       So, having put your overriding methods into the right classes, and
5417       setup the inheritance hierarchy, how do you get the DBI to use them?
5418       You have two choices, either a static method call using the name of
5419       your subclass:
5420
5421         $dbh = MySubDBI->connect(...);
5422
5423       or specifying a "RootClass" attribute:
5424
5425         $dbh = DBI->connect(..., { RootClass => 'MySubDBI' });
5426
5427       If both forms are used then the attribute takes precedence.
5428
5429       The only differences between the two are that using an explicit
5430       RootClass attribute will a) make the DBI automatically attempt to load
5431       a module by that name if the class doesn't exist, and b) won't call
5432       your MySubDBI::connect() method, if you have one.
5433
5434       When subclassing is being used then, after a successful new connect,
5435       the DBI->connect method automatically calls:
5436
5437         $dbh->connected($dsn, $user, $pass, \%attr);
5438
5439       The default method does nothing. The call is made just to simplify any
5440       post-connection setup that your subclass may want to perform.  The
5441       parameters are the same as passed to DBI->connect.  If your subclass
5442       supplies a connected method, it should be part of the MySubDBI::db
5443       package.
5444
5445       One more thing to note: you must let the DBI do the handle creation.
5446       If you want to override the connect() method in your *::dr class then
5447       it must still call SUPER::connect to get a $dbh to work with.
5448       Similarly, an overridden prepare() method in *::db must still call
5449       SUPER::prepare to get a $sth.  If you try to create your own handles
5450       using bless() then you'll find the DBI will reject them with an "is not
5451       a DBI handle (has no magic)" error.
5452
5453       Here's a brief example of a DBI subclass.  A more thorough example can
5454       be found in t/subclass.t in the DBI distribution.
5455
5456         package MySubDBI;
5457
5458         use strict;
5459
5460         use DBI;
5461         use vars qw(@ISA);
5462         @ISA = qw(DBI);
5463
5464         package MySubDBI::db;
5465         use vars qw(@ISA);
5466         @ISA = qw(DBI::db);
5467
5468         sub prepare {
5469           my ($dbh, @args) = @_;
5470           my $sth = $dbh->SUPER::prepare(@args)
5471               or return;
5472           $sth->{private_mysubdbi_info} = { foo => 'bar' };
5473           return $sth;
5474         }
5475
5476         package MySubDBI::st;
5477         use vars qw(@ISA);
5478         @ISA = qw(DBI::st);
5479
5480         sub fetch {
5481           my ($sth, @args) = @_;
5482           my $row = $sth->SUPER::fetch(@args)
5483               or return;
5484           do_something_magical_with_row_data($row)
5485               or return $sth->set_err(1234, "The magic failed", undef, "fetch");
5486           return $row;
5487         }
5488
5489       When calling a SUPER::method that returns a handle, be careful to check
5490       the return value before trying to do other things with it in your
5491       overridden method. This is especially important if you want to set a
5492       hash attribute on the handle, as Perl's autovivification will bite you
5493       by (in)conveniently creating an unblessed hashref, which your method
5494       will then return with usually baffling results later on like the error
5495       "dbih_getcom handle HASH(0xa4451a8) is not a DBI handle (has no magic".
5496       It's best to check right after the call and return undef immediately on
5497       error, just like DBI would and just like the example above.
5498
5499       If your method needs to record an error it should call the set_err()
5500       method with the error code and error string, as shown in the example
5501       above. The error code and error string will be recorded in the handle
5502       and available via "$h->err" and $DBI::errstr etc.  The set_err() method
5503       always returns an undef or empty list as appropriate. Since your method
5504       should nearly always return an undef or empty list as soon as an error
5505       is detected it's handy to simply return what set_err() returns, as
5506       shown in the example above.
5507
5508       If the handle has "RaiseError", "PrintError", or "HandleError" etc. set
5509       then the set_err() method will honour them. This means that if
5510       "RaiseError" is set then set_err() won't return in the normal way but
5511       will 'throw an exception' that can be caught with an "eval" block.
5512
5513       You can stash private data into DBI handles via "$h->{private_..._*}".
5514       See the entry under "ATTRIBUTES COMMON TO ALL HANDLES" for info and
5515       important caveats.
5516

TRACING

5518       The DBI has a powerful tracing mechanism built in. It enables you to
5519       see what's going on 'behind the scenes', both within the DBI and the
5520       drivers you're using.
5521
5522   Trace Settings
5523       Which details are written to the trace output is controlled by a
5524       combination of a trace level, an integer from 0 to 15, and a set of
5525       trace flags that are either on or off. Together these are known as the
5526       trace settings and are stored together in a single integer.  For normal
5527       use you only need to set the trace level, and generally only to a value
5528       between 1 and 4.
5529
5530       Each handle has its own trace settings, and so does the DBI.  When you
5531       call a method the DBI merges the handles settings into its own for the
5532       duration of the call: the trace flags of the handle are OR'd into the
5533       trace flags of the DBI, and if the handle has a higher trace level then
5534       the DBI trace level is raised to match it.  The previous DBI trace
5535       settings are restored when the called method returns.
5536
5537   Trace Levels
5538       Trace levels are as follows:
5539
5540         0 - Trace disabled.
5541         1 - Trace top-level DBI method calls returning with results or errors.
5542         2 - As above, adding tracing of top-level method entry with parameters.
5543         3 - As above, adding some high-level information from the driver
5544             and some internal information from the DBI.
5545         4 - As above, adding more detailed information from the driver.
5546             This is the first level to trace all the rows being fetched.
5547         5 to 15 - As above but with more and more internal information.
5548
5549       Trace level 1 is best for a simple overview of what's happening.  Trace
5550       levels 2 thru 4 a good choice for general purpose tracing.  Levels 5
5551       and above are best reserved for investigating a specific problem, when
5552       you need to see "inside" the driver and DBI.
5553
5554       The trace output is detailed and typically very useful. Much of the
5555       trace output is formatted using the "neat" function, so strings in the
5556       trace output may be edited and truncated by that function.
5557
5558   Trace Flags
5559       Trace flags are used to enable tracing of specific activities within
5560       the DBI and drivers. The DBI defines some trace flags and drivers can
5561       define others. DBI trace flag names begin with a capital letter and
5562       driver specific names begin with a lowercase letter, as usual.
5563
5564       Currently the DBI defines these trace flags:
5565
5566         ALL - turn on all DBI and driver flags (not recommended)
5567         SQL - trace SQL statements executed
5568               (not yet implemented in DBI but implemented in some DBDs)
5569         CON - trace connection process
5570         ENC - trace encoding (unicode translations etc)
5571               (not yet implemented in DBI but implemented in some DBDs)
5572         DBD - trace only DBD messages
5573               (not implemented by all DBDs yet)
5574         TXN - trace transactions
5575               (not implemented in all DBDs yet)
5576
5577       The "parse_trace_flags" and "parse_trace_flag" methods are used to
5578       convert trace flag names into the corresponding integer bit flags.
5579
5580   Enabling Trace
5581       The "$h->trace" method sets the trace settings for a handle and
5582       "DBI->trace" does the same for the DBI.
5583
5584       In addition to the "trace" method, you can enable the same trace
5585       information, and direct the output to a file, by setting the
5586       "DBI_TRACE" environment variable before starting Perl.  See "DBI_TRACE"
5587       for more information.
5588
5589       Finally, you can set, or get, the trace settings for a handle using the
5590       "TraceLevel" attribute.
5591
5592       All of those methods use parse_trace_flags() and so allow you set both
5593       the trace level and multiple trace flags by using a string containing
5594       the trace level and/or flag names separated by vertical bar (""|"") or
5595       comma ("","") characters. For example:
5596
5597         local $h->{TraceLevel} = "3|SQL|foo";
5598
5599   Trace Output
5600       Initially trace output is written to "STDERR".  Both the "$h->trace"
5601       and "DBI->trace" methods take an optional $trace_file parameter, which
5602       may be either the name of a file to be opened by DBI in append mode, or
5603       a reference to an existing writable (possibly layered) filehandle. If
5604       $trace_file is a filename, and can be opened in append mode, or
5605       $trace_file is a writable filehandle, then all trace output (currently
5606       including that from other handles) is redirected to that file. A
5607       warning is generated if $trace_file can't be opened or is not writable.
5608
5609       Further calls to trace() without $trace_file do not alter where the
5610       trace output is sent. If $trace_file is undefined, then trace output is
5611       sent to "STDERR" and, if the prior trace was opened with $trace_file as
5612       a filename, the previous trace file is closed; if $trace_file was a
5613       filehandle, the filehandle is not closed.
5614
5615       NOTE: If $trace_file is specified as a filehandle, the filehandle
5616       should not be closed until all DBI operations are completed, or the
5617       application has reset the trace file via another call to "trace()" that
5618       changes the trace file.
5619
5620   Tracing to Layered Filehandles
5621       NOTE:
5622
5623       ·   Tied filehandles are not currently supported, as tie operations are
5624           not available to the PerlIO methods used by the DBI.
5625
5626       ·   PerlIO layer support requires Perl version 5.8 or higher.
5627
5628       As of version 5.8, Perl provides the ability to layer various
5629       "disciplines" on an open filehandle via the PerlIO module.
5630
5631       A simple example of using PerlIO layers is to use a scalar as the
5632       output:
5633
5634           my $scalar = '';
5635           open( my $fh, "+>:scalar", \$scalar );
5636           $dbh->trace( 2, $fh );
5637
5638       Now all trace output is simply appended to $scalar.
5639
5640       A more complex application of tracing to a layered filehandle is the
5641       use of a custom layer (Refer to Perlio::via for details on creating
5642       custom PerlIO layers.). Consider an application with the following
5643       logger module:
5644
5645           package MyFancyLogger;
5646
5647           sub new
5648           {
5649               my $self = {};
5650               my $fh;
5651               open $fh, '>', 'fancylog.log';
5652               $self->{_fh} = $fh;
5653               $self->{_buf} = '';
5654               return bless $self, shift;
5655           }
5656
5657           sub log
5658           {
5659               my $self = shift;
5660               return unless exists $self->{_fh};
5661               my $fh = $self->{_fh};
5662               $self->{_buf} .= shift;
5663           #
5664           # DBI feeds us pieces at a time, so accumulate a complete line
5665           # before outputing
5666           #
5667               print $fh "At ", scalar localtime(), ':', $self->{_buf}, "\n" and
5668               $self->{_buf} = ''
5669                   if $self->{_buf}=~tr/\n//;
5670           }
5671
5672           sub close {
5673               my $self = shift;
5674               return unless exists $self->{_fh};
5675               my $fh = $self->{_fh};
5676               print $fh "At ", scalar localtime(), ':', $self->{_buf}, "\n" and
5677               $self->{_buf} = ''
5678                   if $self->{_buf};
5679               close $fh;
5680               delete $self->{_fh};
5681           }
5682
5683           1;
5684
5685       To redirect DBI traces to this logger requires creating a package for
5686       the layer:
5687
5688           package PerlIO::via::MyFancyLogLayer;
5689
5690           sub PUSHED
5691           {
5692               my ($class,$mode,$fh) = @_;
5693               my $logger;
5694               return bless \$logger,$class;
5695           }
5696
5697           sub OPEN {
5698               my ($self, $path, $mode, $fh) = @_;
5699               #
5700               # $path is actually our logger object
5701               #
5702               $$self = $path;
5703               return 1;
5704           }
5705
5706           sub WRITE
5707           {
5708               my ($self, $buf, $fh) = @_;
5709               $$self->log($buf);
5710               return length($buf);
5711           }
5712
5713           sub CLOSE {
5714               my $self = shift;
5715               $$self->close();
5716               return 0;
5717           }
5718
5719           1;
5720
5721       The application can then cause DBI traces to be routed to the logger
5722       using
5723
5724           use PerlIO::via::MyFancyLogLayer;
5725
5726           open my $fh, '>:via(MyFancyLogLayer)', MyFancyLogger->new();
5727
5728           $dbh->trace('SQL', $fh);
5729
5730       Now all trace output will be processed by MyFancyLogger's log() method.
5731
5732   Trace Content
5733       Many of the values embedded in trace output are formatted using the
5734       neat() utility function. This means they may be quoted, sanitized, and
5735       possibly truncated if longer than $DBI::neat_maxlen. See "neat" for
5736       more details.
5737
5738   Tracing Tips
5739       You can add tracing to your own application code using the "trace_msg"
5740       method.
5741
5742       It can sometimes be handy to compare trace files from two different
5743       runs of the same script. However using a tool like "diff" on the
5744       original log output doesn't work well because the trace file is full of
5745       object addresses that may differ on each run.
5746
5747       The DBI includes a handy utility called dbilogstrip that can be used to
5748       'normalize' the log content. It can be used as a filter like this:
5749
5750           DBI_TRACE=2 perl yourscript.pl ...args1... 2>&1 | dbilogstrip > dbitrace1.log
5751           DBI_TRACE=2 perl yourscript.pl ...args2... 2>&1 | dbilogstrip > dbitrace2.log
5752           diff -u dbitrace1.log dbitrace2.log
5753
5754       See dbilogstrip for more information.
5755

DBI ENVIRONMENT VARIABLES

5757       The DBI module recognizes a number of environment variables, but most
5758       of them should not be used most of the time.  It is better to be
5759       explicit about what you are doing to avoid the need for environment
5760       variables, especially in a web serving system where web servers are
5761       stingy about which environment variables are available.
5762
5763   DBI_DSN
5764       The DBI_DSN environment variable is used by DBI->connect if you do not
5765       specify a data source when you issue the connect.  It should have a
5766       format such as "dbi:Driver:databasename".
5767
5768   DBI_DRIVER
5769       The DBI_DRIVER environment variable is used to fill in the database
5770       driver name in DBI->connect if the data source string starts "dbi::"
5771       (thereby omitting the driver).  If DBI_DSN omits the driver name,
5772       DBI_DRIVER can fill the gap.
5773
5774   DBI_AUTOPROXY
5775       The DBI_AUTOPROXY environment variable takes a string value that starts
5776       "dbi:Proxy:" and is typically followed by "hostname=...;port=...".  It
5777       is used to alter the behaviour of DBI->connect.  For full details, see
5778       DBI::Proxy documentation.
5779
5780   DBI_USER
5781       The DBI_USER environment variable takes a string value that is used as
5782       the user name if the DBI->connect call is given undef (as distinct from
5783       an empty string) as the username argument.  Be wary of the security
5784       implications of using this.
5785
5786   DBI_PASS
5787       The DBI_PASS environment variable takes a string value that is used as
5788       the password if the DBI->connect call is given undef (as distinct from
5789       an empty string) as the password argument.  Be extra wary of the
5790       security implications of using this.
5791
5792   DBI_DBNAME (obsolete)
5793       The DBI_DBNAME environment variable takes a string value that is used
5794       only when the obsolescent style of DBI->connect (with driver name as
5795       fourth parameter) is used, and when no value is provided for the first
5796       (database name) argument.
5797
5798   DBI_TRACE
5799       The DBI_TRACE environment variable specifies the global default trace
5800       settings for the DBI at startup. Can also be used to direct trace
5801       output to a file. When the DBI is loaded it does:
5802
5803         DBI->trace(split /=/, $ENV{DBI_TRACE}, 2) if $ENV{DBI_TRACE};
5804
5805       So if "DBI_TRACE" contains an ""="" character then what follows it is
5806       used as the name of the file to append the trace to.
5807
5808       output appended to that file. If the name begins with a number followed
5809       by an equal sign ("="), then the number and the equal sign are stripped
5810       off from the name, and the number is used to set the trace level. For
5811       example:
5812
5813         DBI_TRACE=1=dbitrace.log perl your_test_script.pl
5814
5815       On Unix-like systems using a Bourne-like shell, you can do this easily
5816       on the command line:
5817
5818         DBI_TRACE=2 perl your_test_script.pl
5819
5820       See "TRACING" for more information.
5821
5822   PERL_DBI_DEBUG (obsolete)
5823       An old variable that should no longer be used; equivalent to DBI_TRACE.
5824
5825   DBI_PROFILE
5826       The DBI_PROFILE environment variable can be used to enable profiling of
5827       DBI method calls. See DBI::Profile for more information.
5828
5829   DBI_PUREPERL
5830       The DBI_PUREPERL environment variable can be used to enable the use of
5831       DBI::PurePerl.  See DBI::PurePerl for more information.
5832

WARNING AND ERROR MESSAGES

5834   Fatal Errors
5835       Can't call method "prepare" without a package or object reference
5836           The $dbh handle you're using to call "prepare" is probably
5837           undefined because the preceding "connect" failed. You should always
5838           check the return status of DBI methods, or use the "RaiseError"
5839           attribute.
5840
5841       Can't call method "execute" without a package or object reference
5842           The $sth handle you're using to call "execute" is probably
5843           undefined because the preceding "prepare" failed. You should always
5844           check the return status of DBI methods, or use the "RaiseError"
5845           attribute.
5846
5847       DBI/DBD internal version mismatch
5848           The DBD driver module was built with a different version of DBI
5849           than the one currently being used.  You should rebuild the DBD
5850           module under the current version of DBI.
5851
5852           (Some rare platforms require "static linking". On those platforms,
5853           there may be an old DBI or DBD driver version actually embedded in
5854           the Perl executable being used.)
5855
5856       DBD driver has not implemented the AutoCommit attribute
5857           The DBD driver implementation is incomplete. Consult the author.
5858
5859       Can't [sg]et %s->{%s}: unrecognised attribute
5860           You attempted to set or get an unknown attribute of a handle.  Make
5861           sure you have spelled the attribute name correctly; case is
5862           significant (e.g., "Autocommit" is not the same as "AutoCommit").
5863

Pure-Perl DBI

5865       A pure-perl emulation of the DBI is included in the distribution for
5866       people using pure-perl drivers who, for whatever reason, can't install
5867       the compiled DBI. See DBI::PurePerl.
5868

SEE ALSO

5870   Driver and Database Documentation
5871       Refer to the documentation for the DBD driver that you are using.
5872
5873       Refer to the SQL Language Reference Manual for the database engine that
5874       you are using.
5875
5876   ODBC and SQL/CLI Standards Reference Information
5877       More detailed information about the semantics of certain DBI methods
5878       that are based on ODBC and SQL/CLI standards is available on-line via
5879       microsoft.com, for ODBC, and www.jtc1sc32.org for the SQL/CLI standard:
5880
5881        DBI method        ODBC function     SQL/CLI Working Draft
5882        ----------        -------------     ---------------------
5883        column_info       SQLColumns        Page 124
5884        foreign_key_info  SQLForeignKeys    Page 163
5885        get_info          SQLGetInfo        Page 214
5886        primary_key_info  SQLPrimaryKeys    Page 254
5887        table_info        SQLTables         Page 294
5888        type_info         SQLGetTypeInfo    Page 239
5889        statistics_info   SQLStatistics
5890
5891       To find documentation on the ODBC function you can use the MSDN search
5892       facility at:
5893
5894           http://msdn.microsoft.com/Search
5895
5896       and search for something like "SQLColumns returns".
5897
5898       And for SQL/CLI standard information on SQLColumns you'd read page 124
5899       of the (very large) SQL/CLI Working Draft available from:
5900
5901         http://jtc1sc32.org/doc/N0701-0750/32N0744T.pdf
5902
5903   Standards Reference Information
5904       A hyperlinked, browsable version of the BNF syntax for SQL92 (plus
5905       Oracle 7 SQL and PL/SQL) is available here:
5906
5907         http://cui.unige.ch/db-research/Enseignement/analyseinfo/SQL92/BNFindex.html
5908
5909       A BNF syntax for SQL3 is available here:
5910
5911         http://www.sqlstandards.org/SC32/WG3/Progression_Documents/Informal_working_drafts/iso-9075-2-1999.bnf
5912
5913       The following links provide further useful information about SQL.  Some
5914       of these are rather dated now but may still be useful.
5915
5916         http://www.jcc.com/SQLPages/jccs_sql.htm
5917         http://www.contrib.andrew.cmu.edu/~shadow/sql.html
5918         http://www.altavista.com/query?q=sql+tutorial
5919
5920   Books and Articles
5921       Programming the Perl DBI, by Alligator Descartes and Tim Bunce.
5922       <http://books.perl.org/book/154>
5923
5924       Programming Perl 3rd Ed. by Larry Wall, Tom Christiansen & Jon Orwant.
5925       <http://books.perl.org/book/134>
5926
5927       Learning Perl by Randal Schwartz.  <http://books.perl.org/book/101>
5928
5929       Details of many other books related to perl can be found at
5930       <http://books.perl.org>
5931
5932   Perl Modules
5933       Index of DBI related modules available from CPAN:
5934
5935        http://search.cpan.org/search?mode=module&query=DBIx%3A%3A
5936        http://search.cpan.org/search?mode=doc&query=DBI
5937
5938       For a good comparison of RDBMS-OO mappers and some OO-RDBMS mappers
5939       (including Class::DBI, Alzabo, and DBIx::RecordSet in the former
5940       category and Tangram and SPOPS in the latter) see the Perl Object-
5941       Oriented Persistence project pages at:
5942
5943        http://poop.sourceforge.net
5944
5945       A similar page for Java toolkits can be found at:
5946
5947        http://c2.com/cgi-bin/wiki?ObjectRelationalToolComparison
5948
5949   Mailing List
5950       The dbi-users mailing list is the primary means of communication among
5951       users of the DBI and its related modules. For details send email to:
5952
5953        dbi-users-help@perl.org
5954
5955       There are typically between 700 and 900 messages per month.  You have
5956       to subscribe in order to be able to post. However you can opt for a
5957       'post-only' subscription.
5958
5959       Mailing list archives (of variable quality) are held at:
5960
5961        http://groups.google.com/groups?group=perl.dbi.users
5962        http://www.xray.mpe.mpg.de/mailing-lists/dbi/
5963        http://www.mail-archive.com/dbi-users%40perl.org/
5964
5965   Assorted Related WWW Links
5966       The DBI "Home Page":
5967
5968        http://dbi.perl.org/
5969
5970       Other DBI related links:
5971
5972        http://tegan.deltanet.com/~phlip/DBUIdoc.html
5973        http://dc.pm.org/perl_db.html
5974        http://wdvl.com/Authoring/DB/Intro/toc.html
5975        http://www.hotwired.com/webmonkey/backend/tutorials/tutorial1.html
5976        http://bumppo.net/lists/macperl/1999/06/msg00197.html
5977        http://www.perlmonks.org/?node=DBI%20recipes
5978        http://www.perlmonks.org/?node=Speeding%20up%20the%20DBI
5979
5980       Other database related links:
5981
5982        http://www.jcc.com/sql_stnd.html
5983        http://cuiwww.unige.ch/OSG/info/FreeDB/FreeDB.home.html
5984        http://www.connectionstrings.com/
5985
5986       Security, especially the "SQL Injection" attack:
5987
5988        http://www.ngssoftware.com/research/papers.html
5989        http://www.ngssoftware.com/papers/advanced_sql_injection.pdf
5990        http://www.ngssoftware.com/papers/more_advanced_sql_injection.pdf
5991        http://www.esecurityplanet.com/trends/article.php/2243461
5992        http://www.spidynamics.com/papers/SQLInjectionWhitePaper.pdf
5993        http://www.imperva.com/application_defense_center/white_papers/blind_sql_server_injection.html
5994        http://online.securityfocus.com/infocus/1644
5995
5996       Commercial and Data Warehouse Links
5997
5998        http://www.dwinfocenter.org
5999        http://www.datawarehouse.com
6000        http://www.datamining.org
6001        http://www.olapcouncil.org
6002        http://www.idwa.org
6003        http://www.knowledgecenters.org/dwcenter.asp
6004
6005       Recommended Perl Programming Links
6006
6007        http://language.perl.com/style/
6008
6009   FAQ
6010       See <http://faq.dbi-support.com/>
6011

AUTHORS

6013       DBI by Tim Bunce, <http://www.tim.bunce.name>
6014
6015       This pod text by Tim Bunce, J. Douglas Dunlop, Jonathan Leffler and
6016       others.  Perl by Larry Wall and the "perl5-porters".
6017
6019       The DBI module is Copyright (c) 1994-2012 Tim Bunce. Ireland.  All
6020       rights reserved.
6021
6022       You may distribute under the terms of either the GNU General Public
6023       License or the Artistic License, as specified in the Perl 5.10.0 README
6024       file.
6025

SUPPORT / WARRANTY

6027       The DBI is free Open Source software. IT COMES WITHOUT WARRANTY OF ANY
6028       KIND.
6029
6030   Support
6031       My consulting company, Data Plan Services, offers annual and multi-
6032       annual support contracts for the DBI. These provide sustained support
6033       for DBI development, and sustained value for you in return.  Contact me
6034       for details.
6035
6036   Sponsor Enhancements
6037       The DBI Roadmap is available at
6038       <http://search.cpan.org/~timb/DBI/Roadmap.pod>
6039
6040       If your company would benefit from a specific new DBI feature, please
6041       consider sponsoring its development.  Work is performed rapidly, and
6042       usually on a fixed-price payment-on-delivery basis.  Contact me for
6043       details.
6044
6045       Using such targeted financing allows you to contribute to DBI
6046       development, and rapidly get something specific and valuable in return.
6047

ACKNOWLEDGEMENTS

6049       I would like to acknowledge the valuable contributions of the many
6050       people I have worked with on the DBI project, especially in the early
6051       years (1992-1994). In no particular order: Kevin Stock, Buzz Moschetti,
6052       Kurt Andersen, Ted Lemon, William Hails, Garth Kennedy, Michael
6053       Peppler, Neil S. Briscoe, Jeff Urlwin, David J. Hughes, Jeff Stander,
6054       Forrest D Whitcher, Larry Wall, Jeff Fried, Roy Johnson, Paul Hudson,
6055       Georg Rehfeld, Steve Sizemore, Ron Pool, Jon Meek, Tom Christiansen,
6056       Steve Baumgarten, Randal Schwartz, and a whole lot more.
6057
6058       Then, of course, there are the poor souls who have struggled through
6059       untold and undocumented obstacles to actually implement DBI drivers.
6060       Among their ranks are Jochen Wiedmann, Alligator Descartes, Jonathan
6061       Leffler, Jeff Urlwin, Michael Peppler, Henrik Tougaard, Edwin Pratomo,
6062       Davide Migliavacca, Jan Pazdziora, Peter Haworth, Edmund Mergl, Steve
6063       Williams, Thomas Lowery, and Phlip Plumlee. Without them, the DBI would
6064       not be the practical reality it is today.  I'm also especially grateful
6065       to Alligator Descartes for starting work on the first edition of the
6066       "Programming the Perl DBI" book and letting me jump on board.
6067
6068       The DBI and DBD::Oracle were originally developed while I was Technical
6069       Director (CTO) of Ingeneering in the UK (<http://www.ig.co.uk>)
6070       (formerly known as the Paul Ingram Group).  So I'd especially like to
6071       thank Paul for his generosity and vision in supporting this work for
6072       many years.
6073
6074       A couple of specific DBI features have been sponsored by enlightened
6075       companies:
6076
6077       The development of the swap_inner_handle() method was sponsored by
6078       BizRate.com (<http://BizRate.com>)
6079
6080       The development of DBD::Gofer and related modules was sponsored by
6081       Shopzilla.com (<http://Shopzilla.com>), where I currently work.
6082

CONTRIBUTING

6084       As you can see above, many people have contributed to the DBI and
6085       drivers in many ways over many years.
6086
6087       If you'd like to help then see <http://dbi.perl.org/contributing> and
6088       <http://search.cpan.org/~timb/DBI/Roadmap.pod>
6089
6090       If you'd like the DBI to do something new or different then a good way
6091       to make that happen is to do it yourself and send me a patch to the
6092       source code that shows the changes. (But read "Speak before you patch"
6093       below.)
6094
6095   Browsing the source code repository
6096       Use http://svn.perl.org/modules/dbi/trunk (basic) or
6097       http://svn.perl.org/viewcvs/modules/ (more useful)
6098
6099   How to create a patch using Subversion
6100       The DBI source code is maintained using Subversion (a replacement for
6101       CVS, see <http://subversion.tigris.org/>). To access the source you'll
6102       need to install a Subversion client. Then, to get the source code, do:
6103
6104         svn checkout http://svn.perl.org/modules/dbi/trunk
6105
6106       If it prompts for a username and password use your perl.org account if
6107       you have one, else just 'guest' and 'guest'. The source code will be in
6108       a new subdirectory called "trunk".
6109
6110       To keep informed about changes to the source you can send an empty
6111       email to svn-commit-modules-dbi-subscribe@perl.org after which you'll
6112       get an email with the change log message and diff of each change
6113       checked-in to the source.
6114
6115       After making your changes you can generate a patch file, but before you
6116       do, make sure your source is still up to date using:
6117
6118         svn update
6119
6120       If you get any conflicts reported you'll need to fix them first.  Then
6121       generate the patch file from within the "trunk" directory using:
6122
6123         svn diff > foo.patch
6124
6125       Read the patch file, as a sanity check, and then email it to
6126       dbi-dev@perl.org.
6127
6128   How to create a patch without Subversion
6129       Unpack a fresh copy of the distribution:
6130
6131         tar xfz DBI-1.40.tar.gz
6132
6133       Rename the newly created top level directory:
6134
6135         mv DBI-1.40 DBI-1.40.your_foo
6136
6137       Edit the contents of DBI-1.40.your_foo/* till it does what you want.
6138
6139       Test your changes and then remove all temporary files:
6140
6141         make test && make distclean
6142
6143       Go back to the directory you originally unpacked the distribution:
6144
6145         cd ..
6146
6147       Unpack another copy of the original distribution you started with:
6148
6149         tar xfz DBI-1.40.tar.gz
6150
6151       Then create a patch file by performing a recursive "diff" on the two
6152       top level directories:
6153
6154         diff -r -u DBI-1.40 DBI-1.40.your_foo > DBI-1.40.your_foo.patch
6155
6156   Speak before you patch
6157       For anything non-trivial or possibly controversial it's a good idea to
6158       discuss (on dbi-dev@perl.org) the changes you propose before actually
6159       spending time working on them. Otherwise you run the risk of them being
6160       rejected because they don't fit into some larger plans you may not be
6161       aware of.
6162

TRANSLATIONS

6164       A German translation of this manual (possibly slightly out of date) is
6165       available, thanks to O'Reilly, at:
6166
6167         http://www.oreilly.de/catalog/perldbiger/
6168
6169       Some other translations:
6170
6171        http://cronopio.net/perl/                              - Spanish
6172        http://member.nifty.ne.jp/hippo2000/dbimemo.htm        - Japanese
6173

TRAINING

6175       References to DBI related training resources. No recommendation
6176       implied.
6177
6178         http://www.treepax.co.uk/
6179         http://www.keller.com/dbweb/
6180
6181       (If you offer professional DBI related training services, please send
6182       me your details so I can add them here.)
6183
6185       Apache::DBI by E.Mergl@bawue.de
6186           To be used with the Apache daemon together with an embedded Perl
6187           interpreter like "mod_perl". Establishes a database connection
6188           which remains open for the lifetime of the HTTP daemon. This way
6189           the CGI connect and disconnect for every database access becomes
6190           superfluous.
6191
6192       SQL Parser
6193           See also the SQL::Statement module, SQL parser and engine.
6194
6195
6196
6197perl v5.16.3                      2013-05-16                            DBI(3)
Impressum