1SQL::SplitStatement(3)User Contributed Perl DocumentationSQL::SplitStatement(3)
2
3
4

NAME

6       SQL::SplitStatement - Split any SQL code into atomic statements
7

SYNOPSIS

9           # Multiple SQL statements in a single string
10           my $sql_code = <<'SQL';
11           CREATE TABLE parent(a, b, c   , d    );
12           CREATE TABLE child (x, y, "w;", "z;z");
13           /* C-style comment; */
14           CREATE TRIGGER "check;delete;parent;" BEFORE DELETE ON parent WHEN
15               EXISTS (SELECT 1 FROM child WHERE old.a = x AND old.b = y)
16           BEGIN
17               SELECT RAISE(ABORT, 'constraint failed;'); -- Inline SQL comment
18           END;
19           -- Standalone SQL; comment; with semicolons;
20           INSERT INTO parent (a, b, c, d) VALUES ('pippo;', 'pluto;', NULL, NULL);
21           SQL
22
23           use SQL::SplitStatement;
24
25           my $sql_splitter = SQL::SplitStatement->new;
26           my @statements = $sql_splitter->split($sql_code);
27
28           # @statements now is:
29           #
30           # (
31           #     'CREATE TABLE parent(a, b, c   , d    )',
32           #     'CREATE TABLE child (x, y, "w;", "z;z")',
33           #     'CREATE TRIGGER "check;delete;parent;" BEFORE DELETE ON parent WHEN
34           #     EXISTS (SELECT 1 FROM child WHERE old.a = x AND old.b = y)
35           # BEGIN
36           #     SELECT RAISE(ABORT, \'constraint failed;\');
37           # END',
38           #     'INSERT INTO parent (a, b, c, d) VALUES (\'pippo;\', \'pluto;\', NULL, NULL)'
39           # )
40

DESCRIPTION

42       This is a simple module which tries to split any SQL code, even
43       including non-standard extensions (for the details see the "SUPPORTED
44       DBMSs" section below), into the atomic statements it is composed of.
45
46       The logic used to split the SQL code is more sophisticated than a raw
47       "split" on the ";" (semicolon) character: first, various different
48       statement terminator tokens are recognized (see below for the list),
49       then this module is able to correctly handle the presence of said
50       tokens inside identifiers, values, comments, "BEGIN ... END" blocks
51       (even nested), dollar-quoted strings, MySQL custom "DELIMITER"s,
52       procedural code etc., as (partially) exemplified in the "SYNOPSIS"
53       above.
54
55       Consider however that this is by no means a validating parser
56       (technically speaking, it's just a context-sensitive tokenizer). It
57       should rather be seen as an in-progress heuristic approach, which will
58       gradually improve as test cases will be reported. This also means that,
59       except for the "LIMITATIONS" detailed below, there is no known (to the
60       author) SQL code the most current release of this module can't
61       correctly split.
62
63       The test suite bundled with the distribution (which now includes the
64       popular Sakila and Pagila sample db schemata, as detailed in the
65       "SHOWCASE" section below) should give you an idea of the capabilities
66       of this module
67
68       If your atomic statements are to be fed to a DBMS, you are encouraged
69       to use DBIx::MultiStatementDo instead, which uses this module and also
70       (optionally) offers automatic transactions support, so that you'll have
71       the all-or-nothing behavior you would probably want.
72

METHODS

74   "new"
75       •   "SQL::SplitStatement->new( %options )"
76
77       •   "SQL::SplitStatement->new( \%options )"
78
79       It creates and returns a new SQL::SplitStatement object. It accepts its
80       options either as a hash or a hashref.
81
82       "new" takes the following Boolean options, which for documentation
83       purposes can be grouped in two sets: "Formatting Options" and "DBMSs
84       Specific Options".
85
86       Formatting Options
87
88       •   "keep_terminators"
89
90           A Boolean option which causes, when set to a false value (which is
91           the default), the trailing terminator token to be discarded in the
92           returned atomic statements.  When set to a true value, the
93           terminators are kept instead.
94
95           The possible terminators (which are treated as such depending on
96           the context) are:
97
98           •   ";" (the semicolon character);
99
100           •   any string defined by the MySQL "DELIMITER" command;
101
102           •   an ";" followed by an "/" (forward-slash character) on its own
103               line;
104
105           •   an ";" followed by an "." (dot character) on its own line,
106               followed by an "/" on its own line;
107
108           •   an "/" on its own line regardless of the preceding characters
109               (only if the "slash_terminates" option, explained below, is
110               set).
111
112           The multi-line terminators above are always treated as a single
113           token, that is they are discarded (or returned) as a whole
114           (regardless of the "slash_terminates" option value).
115
116           If your statements are to be fed to a DBMS, you are advised to keep
117           this option to its default (false) value, since some drivers/DBMSs
118           don't want the terminator to be present at the end of the (single)
119           statement.
120
121           (Note that the last, possibly empty, statement of a given SQL text,
122           never has a trailing terminator. See below for an example.)
123
124       •   "keep_terminator"
125
126           An alias for the the "keep_terminators" option explained above.
127           Note that if "keep_terminators" and "keep_terminator" are both
128           passed to "new", an exception is thrown.
129
130       •   "keep_extra_spaces"
131
132           A Boolean option which causes, when set to a false value (which is
133           the default), the spaces ("\s") around the statements to be
134           trimmed.  When set to a true value, these spaces are kept instead.
135
136           When "keep_terminators" is set to false as well, the terminator is
137           discarded first (regardless of the spaces around it) and the
138           trailing spaces are trimmed then. This ensures that if
139           "keep_extra_spaces" is set to false, the returned statements will
140           never have trailing (nor leading) spaces, regardless of the
141           "keep_terminators" value.
142
143       •   "keep_comments"
144
145           A Boolean option which causes, when set to a false value (which is
146           the default), the comments to be discarded in the returned
147           statements. When set to a true value, they are kept with the
148           statements instead.
149
150           Both SQL and multi-line C-style comments are recognized.
151
152           When kept, each comment is returned in the same string with the
153           atomic statement it belongs to. A comment belongs to a statement if
154           it appears, in the original SQL code, before the end of that
155           statement and after the terminator of the previous statement (if it
156           exists), as shown in this pseudo-SQL snippet:
157
158               /* This comment
159               will be returned
160               together with statement1 */
161
162               <statement1>; -- This will go with statement2
163                             -- (note the semicolon which closes statement1)
164
165               <statement2>
166               -- This with statement2 as well
167
168       •   "keep_empty_statements"
169
170           A Boolean option which causes, when set to a false value (which is
171           the default), the empty statements to be discarded. When set to a
172           true value, the empty statements are returned instead.
173
174           A statement is considered empty when it contains no characters
175           other than the terminator and space characters ("\s").
176
177           A statement composed solely of comments is not recognized as empty
178           and may therefore be returned even when "keep_empty_statements" is
179           false. To avoid this, it is sufficient to leave "keep_comments" to
180           false as well.
181
182           Note instead that an empty statement is recognized as such
183           regardless of the value of the options "keep_terminators" and
184           "keep_extra_spaces".
185
186       These options are basically to be kept to their default (false) values,
187       especially if the atomic statements are to be given to a DBMS.
188
189       They are intended mainly for cosmetic reasons, or if you want to count
190       by how many atomic statements, including the empty ones, your original
191       SQL code was composed of.
192
193       Another situation where they are useful (in the general case necessary,
194       really), is when you want to retain the ability to verbatim rebuild the
195       original SQL string from the returned statements:
196
197           my $verbatim_splitter = SQL::SplitStatement->new(
198               keep_terminators      => 1,
199               keep_extra_spaces     => 1,
200               keep_comments         => 1,
201               keep_empty_statements => 1
202           );
203
204           my @verbatim_statements = $verbatim_splitter->split($sql_string);
205
206           $sql_string eq join '', @verbatim_statements; # Always true, given the constructor above.
207
208       Other than this, again, you are recommended to stick with the defaults.
209
210       DBMSs Specific Options
211
212       The same syntactic structure can have different semantics across
213       different SQL dialects, so sometimes it is necessary to help the parser
214       to make the right decision. This is the function of these options.
215
216       •   "slash_terminates"
217
218           A Boolean option which causes, when set to a true value (which is
219           the default), a "/" (forward-slash) on its own line, even without a
220           preceding semicolon, to be admitted as a (possible) terminator.
221
222           If set to false, a forward-slash on its own line is treated as a
223           statement terminator only if preceded by a semicolon or by a dot
224           and a semicolon.
225
226           If you are dealing with Oracle's SQL, you should let this option
227           set, since a slash (alone, without a preceding semicolon) is
228           sometimes used as a terminator, as it is permitted by SQL*Plus (on
229           non-block statements).
230
231           With SQL dialects other than Oracle, there is the (theoretical)
232           possibility that a slash on its own line can pass the additional
233           checks and be considered a terminator (while it shouldn't). This
234           chance should be really tiny (it has never been observed in real
235           world code indeed). Though negligible, by setting this option to
236           false that risk can anyway be ruled out.
237
238   "split"
239       •   "$sql_splitter->split( $sql_string )"
240
241       This is the method which actually splits the SQL code into its atomic
242       components.
243
244       It returns a list containing the atomic statements, in the same order
245       they appear in the original SQL code. The atomic statements are
246       returned according to the options explained above.
247
248       Note that, as mentioned above, an SQL string which terminates with a
249       terminator token (for example a semicolon), contains a trailing empty
250       statement: this is correct and it is treated accordingly (if
251       "keep_empty_statements" is set to a true value):
252
253           my $sql_splitter = SQL::SplitStatement->new(
254               keep_empty_statements => 1
255           );
256
257           my @statements = $sql_splitter->split( 'SELECT 1;' );
258
259           print 'The SQL code contains ' . scalar(@statements) . ' statements.';
260           # The SQL code contains 2 statements.
261
262   "split_with_placeholders"
263       •   "$sql_splitter->split_with_placeholders( $sql_string )"
264
265       It works exactly as the "split" method explained above, except that it
266       returns also a list of integers, each of which is the number of the
267       placeholders contained in the corresponding atomic statement.
268
269       More precisely, its return value is a list of two elements, the first
270       of which is a reference to the list of the atomic statements exactly as
271       returned by the "split" method, while the second is a reference to the
272       list of the number of placeholders as explained above.
273
274       Here is an example:
275
276           # 4 statements (valid SQLite SQL)
277           my $sql_code = <<'SQL';
278           CREATE TABLE state (id, name);
279           INSERT INTO  state (id, name) VALUES (?, ?);
280           CREATE TABLE city  (id, name, state_id);
281           INSERT INTO  city  (id, name, state_id) VALUES (?, ?, ?)
282           SQL
283
284           my $splitter = SQL::SplitStatement->new;
285
286           my ( $statements, $placeholders )
287               = $splitter->split_with_placeholders( $sql_code );
288
289           # $placeholders now is: [0, 2, 0, 3]
290
291       where the returned $placeholders list(ref) is to be read as follows:
292       the first statement contains 0 placeholders, the second 2, the third 0
293       and the fourth 3.
294
295       The recognized placeholders are:
296
297       •   question mark placeholders, represented by the "?" character;
298
299       •   dollar sign numbered placeholders, represented by the "$1, $2, ...,
300           $n" strings;
301
302       •   named parameters, such as ":foo", ":bar", ":baz" etc.
303
304   "keep_terminators"
305       •   "$sql_splitter->keep_terminators"
306
307       •   "$sql_splitter->keep_terminators( $boolean )"
308
309           Getter/setter method for the "keep_terminators" option explained
310           above.
311
312   "keep_terminator"
313       An alias for the "keep_terminators" method explained above.
314
315   "keep_extra_spaces"
316       •   "$sql_splitter->keep_extra_spaces"
317
318       •   "$sql_splitter->keep_extra_spaces( $boolean )"
319
320           Getter/setter method for the "keep_extra_spaces" option explained
321           above.
322
323   "keep_comments"
324       •   "$sql_splitter->keep_comments"
325
326       •   "$sql_splitter->keep_comments( $boolean )"
327
328           Getter/setter method for the "keep_comments" option explained
329           above.
330
331   "keep_empty_statements"
332       •   "$sql_splitter->keep_empty_statements"
333
334       •   "$sql_splitter->keep_empty_statements( $boolean )"
335
336           Getter/setter method for the "keep_empty_statements" option
337           explained above.
338
339   "slash_terminates"
340       •   "$sql_splitter->slash_terminates"
341
342       •   "$sql_splitter->slash_terminates( $boolean )"
343
344           Getter/setter method for the "slash_terminates" option explained
345           above.
346

SUPPORTED DBMSs

348       SQL::SplitStatement aims to cover the widest possible range of DBMSs,
349       SQL dialects and extensions (even proprietary), in a (nearly) fully
350       transparent way for the user.
351
352       Currently it has been tested mainly on SQLite, PostgreSQL, MySQL and
353       Oracle.
354
355   Procedural Extensions
356       Procedural code is by far the most complex to handle.
357
358       Currently any block of code which start with "FUNCTION", "PROCEDURE",
359       "DECLARE", "CREATE" or "CALL" is correctly recognized, as well as
360       anonymous "BEGIN ... END" blocks, dollar quoted blocks and blocks
361       delimited by a "DELIMITER"-defined custom terminator, therefore a wide
362       range of procedural extensions should be handled correctly. However,
363       only PL/SQL, PL/PgSQL and MySQL code has been tested so far.
364
365       If you need also other procedural languages to be recognized, please
366       let me know (possibly with some test cases).
367

LIMITATIONS

369       Bound to be plenty, given the heuristic nature of this module (and its
370       ambitious goals). However, no limitations are currently known.
371
372       Please report any problematic test case.
373
374   Non-limitations
375       To be split correctly, the given input must, in general, be
376       syntactically valid SQL. For example, an unbalanced "BEGIN" or a
377       misspelled keyword could, under certain circumstances, confuse the
378       parser and make it trip over the next statement terminator, thus
379       returning non-split statements.  This should not be seen as a
380       limitation though, as the original (invalid) SQL code would have been
381       unusable anyway (remember that this is NOT a validating parser!)
382

SHOWCASE

384       To test the capabilities of this module, you can run it (or rather run
385       sql-split) on the files t/data/sakila-schema.sql and
386       t/data/pagila-schema.sql included in the distribution, which contain
387       two quite large and complex real world db schemata, for MySQL and
388       PostgreSQL respectively.
389
390       For more information:
391
392       •   Sakila db: <http://dev.mysql.com/doc/sakila/en/sakila.html>
393
394       •   Pagila db: <http://pgfoundry.org/projects/dbsamples>
395

DEPENDENCIES

397       SQL::SplitStatement depends on the following modules:
398
399       •   Carp
400
401       •   Class::Accessor::Fast
402
403       •   List::MoreUtils
404
405       •   Regexp::Common
406
407       •   SQL::Tokenizer 0.22 or newer
408

AUTHOR

410       Emanuele Zeppieri, "<emazep@cpan.org>"
411

BUGS

413       No known bugs.
414
415       Please report any bugs or feature requests to "bug-sql-SplitStatement
416       at rt.cpan.org", or through the web interface at
417       <http://rt.cpan.org/NoAuth/ReportBug.html?Queue=SQL-SplitStatement>.  I
418       will be notified, and then you'll automatically be notified of progress
419       on your bug as I make changes.
420

SUPPORT

422       You can find documentation for this module with the perldoc command:
423
424           perldoc SQL::SplitStatement
425
426       You can also look for information at:
427
428       •   AnnoCPAN: Annotated CPAN documentation
429
430           <http://annocpan.org/dist/SQL-SplitStatement>
431
432       •   CPAN Ratings
433
434           <http://cpanratings.perl.org/d/SQL-SplitStatement>
435
436       •   On MetaCPAN
437
438           <https://metacpan.org/pod/SQL::SplitStatement/>
439

ACKNOWLEDGEMENTS

441       Igor Sutton for his excellent SQL::Tokenizer, which made writing this
442       module a joke.
443

SEE ALSO

445       •   DBIx::MultiStatementDo
446
447       •   sql-split
448

LICENSE AND COPYRIGHT

450       Copyright 2010-2011 Emanuele Zeppieri.
451
452       This program is free software; you can redistribute it and/or modify it
453       under the terms of either: the GNU General Public License as published
454       by the Free Software Foundation, or the Artistic License.
455
456       See http://dev.perl.org/licenses/ for more information.
457
458
459
460perl v5.34.0                      2022-01-21            SQL::SplitStatement(3)