1CGI(3pm) Perl Programmers Reference Guide CGI(3pm)
2
6 CGI - Handle Common Gateway Interface requests and responses
7
9 use CGI;
10 my $q = CGI->new;
12 # Process an HTTP request
14 @values = $q->param('form_field');
15 $fh = $q->upload('file_field');
17 $riddle = $query->cookie('riddle_name');
19 %answers = $query->cookie('answers');
20 # Prepare various HTTP responses
22 print $q->header();
23 print $q->header('application/json');
24 $cookie1 = $q->cookie(-name=>'riddle_name', -value=>"The Sphynx's Question");
26 $cookie2 = $q->cookie(-name=>'answers', -value=>\%answers);
27 print $q->header(
28 -type => 'image/gif',
29 -expires => '+3d',
30 -cookie => [$cookie1,$cookie2]
31 );
32 print $q->redirect('http://somewhere.else/in/movie/land');
34
36 CGI.pm is a stable, complete and mature solution for processing and
37 preparing HTTP requests and responses. Major features including
38 processing form submissions, file uploads, reading and writing cookies,
39 query string generation and manipulation, and processing and preparing
40 HTTP headers. Some HTML generation utilities are included as well.
41 CGI.pm performs very well in in a vanilla CGI.pm environment and also
43 comes with built-in support for mod_perl and mod_perl2 as well as
44 FastCGI.
45 It has the benefit of having developed and refined over 10 years with
47 input from dozens of contributors and being deployed on thousands of
48 websites. CGI.pm has been included in the Perl distribution since Perl
49 5.4, and has become a de-facto standard.
50 PROGRAMMING STYLE
52 There are two styles of programming with CGI.pm, an object-oriented
53 style and a function-oriented style. In the object-oriented style you
54 create one or more CGI objects and then use object methods to create
55 the various elements of the page. Each CGI object starts out with the
56 list of named parameters that were passed to your CGI script by the
57 server. You can modify the objects, save them to a file or database
58 and recreate them. Because each object corresponds to the "state" of
59 the CGI script, and because each object's parameter list is independent
60 of the others, this allows you to save the state of the script and
61 restore it later.
62 For example, using the object oriented style, here is how you create a
64 simple "Hello World" HTML page:
65 #!/usr/local/bin/perl -w
67 use CGI; # load CGI routines
68 $q = CGI->new; # create new CGI object
69 print $q->header, # create the HTTP header
70 $q->start_html('hello world'), # start the HTML
71 $q->h1('hello world'), # level 1 header
72 $q->end_html; # end the HTML
73 In the function-oriented style, there is one default CGI object that
75 you rarely deal with directly. Instead you just call functions to
76 retrieve CGI parameters, create HTML tags, manage cookies, and so on.
77 This provides you with a cleaner programming interface, but limits you
78 to using one CGI object at a time. The following example prints the
79 same page, but uses the function-oriented interface. The main
80 differences are that we now need to import a set of functions into our
81 name space (usually the "standard" functions), and we don't need to
82 create the CGI object.
83 #!/usr/local/bin/perl
85 use CGI qw/:standard/; # load standard CGI routines
86 print header, # create the HTTP header
87 start_html('hello world'), # start the HTML
88 h1('hello world'), # level 1 header
89 end_html; # end the HTML
90 The examples in this document mainly use the object-oriented style.
92 See HOW TO IMPORT FUNCTIONS for important information on function-
93 oriented programming in CGI.pm
94 CALLING CGI.PM ROUTINES
96 Most CGI.pm routines accept several arguments, sometimes as many as 20
97 optional ones! To simplify this interface, all routines use a named
98 argument calling style that looks like this:
99 print $q->header(-type=>'image/gif',-expires=>'+3d');
101 Each argument name is preceded by a dash. Neither case nor order
103 matters in the argument list. -type, -Type, and -TYPE are all
104 acceptable. In fact, only the first argument needs to begin with a
105 dash. If a dash is present in the first argument, CGI.pm assumes
106 dashes for the subsequent ones.
107 Several routines are commonly called with just one argument. In the
109 case of these routines you can provide the single argument without an
110 argument name. header() happens to be one of these routines. In this
111 case, the single argument is the document type.
112 print $q->header('text/html');
114 Other such routines are documented below.
116 Sometimes named arguments expect a scalar, sometimes a reference to an
118 array, and sometimes a reference to a hash. Often, you can pass any
119 type of argument and the routine will do whatever is most appropriate.
120 For example, the param() routine is used to set a CGI parameter to a
121 single or a multi-valued value. The two cases are shown below:
122 $q->param(-name=>'veggie',-value=>'tomato');
124 $q->param(-name=>'veggie',-value=>['tomato','tomahto','potato','potahto']);
125 A large number of routines in CGI.pm actually aren't specifically
127 defined in the module, but are generated automatically as needed.
128 These are the "HTML shortcuts," routines that generate HTML tags for
129 use in dynamically-generated pages. HTML tags have both attributes
130 (the attribute="value" pairs within the tag itself) and contents (the
131 part between the opening and closing pairs.) To distinguish between
132 attributes and contents, CGI.pm uses the convention of passing HTML
133 attributes as a hash reference as the first argument, and the contents,
134 if any, as any subsequent arguments. It works out like this:
135 Code Generated HTML
137 ---- --------------
138 h1() <h1>
139 h1('some','contents'); <h1>some contents</h1>
140 h1({-align=>left}); <h1 align="LEFT">
141 h1({-align=>left},'contents'); <h1 align="LEFT">contents</h1>
142 HTML tags are described in more detail later.
144 Many newcomers to CGI.pm are puzzled by the difference between the
146 calling conventions for the HTML shortcuts, which require curly braces
147 around the HTML tag attributes, and the calling conventions for other
148 routines, which manage to generate attributes without the curly
149 brackets. Don't be confused. As a convenience the curly braces are
150 optional in all but the HTML shortcuts. If you like, you can use curly
151 braces when calling any routine that takes named arguments. For
152 example:
153 print $q->header( {-type=>'image/gif',-expires=>'+3d'} );
155 If you use the -w switch, you will be warned that some CGI.pm argument
157 names conflict with built-in Perl functions. The most frequent of
158 these is the -values argument, used to create multi-valued menus, radio
159 button clusters and the like. To get around this warning, you have
160 several choices:
161 1. Use another name for the argument, if one is available. For
163 example, -value is an alias for -values.
164 2. Change the capitalization, e.g. -Values
166 3. Put quotes around the argument name, e.g. '-values'
168 Many routines will do something useful with a named argument that it
170 doesn't recognize. For example, you can produce non-standard HTTP
171 header fields by providing them as named arguments:
172 print $q->header(-type => 'text/html',
174 -cost => 'Three smackers',
175 -annoyance_level => 'high',
176 -complaints_to => 'bit bucket');
177 This will produce the following nonstandard HTTP header:
179 HTTP/1.0 200 OK
181 Cost: Three smackers
182 Annoyance-level: high
183 Complaints-to: bit bucket
184 Content-type: text/html
185 Notice the way that underscores are translated automatically into
187 hyphens. HTML-generating routines perform a different type of
188 translation.
189 This feature allows you to keep up with the rapidly changing HTTP and
191 HTML "standards".
192 CREATING A NEW QUERY OBJECT (OBJECT-ORIENTED STYLE):
194 $query = CGI->new;
195 This will parse the input (from both POST and GET methods) and store it
197 into a perl5 object called $query.
198 Any filehandles from file uploads will have their position reset to the
200 beginning of the file.
201 CREATING A NEW QUERY OBJECT FROM AN INPUT FILE
203 $query = CGI->new(INPUTFILE);
204 If you provide a file handle to the new() method, it will read
206 parameters from the file (or STDIN, or whatever). The file can be in
207 any of the forms describing below under debugging (i.e. a series of
208 newline delimited TAG=VALUE pairs will work). Conveniently, this type
209 of file is created by the save() method (see below). Multiple records
210 can be saved and restored.
211 Perl purists will be pleased to know that this syntax accepts
213 references to file handles, or even references to filehandle globs,
214 which is the "official" way to pass a filehandle:
215 $query = CGI->new(\*STDIN);
217 You can also initialize the CGI object with a FileHandle or IO::File
219 object.
220 If you are using the function-oriented interface and want to initialize
222 CGI state from a file handle, the way to do this is with
223 restore_parameters(). This will (re)initialize the default CGI object
224 from the indicated file handle.
225 open (IN,"test.in") || die;
227 restore_parameters(IN);
228 close IN;
229 You can also initialize the query object from a hash reference:
231 $query = CGI->new( {'dinosaur'=>'barney',
233 'song'=>'I love you',
234 'friends'=>[qw/Jessica George Nancy/]}
235 );
236 or from a properly formatted, URL-escaped query string:
238 $query = CGI->new('dinosaur=barney&color=purple');
240 or from a previously existing CGI object (currently this clones the
242 parameter list, but none of the other object-specific fields, such as
243 autoescaping):
244 $old_query = CGI->new;
246 $new_query = CGI->new($old_query);
247 To create an empty query, initialize it from an empty string or hash:
249 $empty_query = CGI->new("");
251 -or-
253 $empty_query = CGI->new({});
255 FETCHING A LIST OF KEYWORDS FROM THE QUERY:
257 @keywords = $query->keywords
258 If the script was invoked as the result of an <ISINDEX> search, the
260 parsed keywords can be obtained as an array using the keywords()
261 method.
262 FETCHING THE NAMES OF ALL THE PARAMETERS PASSED TO YOUR SCRIPT:
264 @names = $query->param
265 If the script was invoked with a parameter list (e.g.
267 "name1=value1&name2=value2&name3=value3"), the param() method will
268 return the parameter names as a list. If the script was invoked as an
269 <ISINDEX> script and contains a string without ampersands (e.g.
270 "value1+value2+value3") , there will be a single parameter named
271 "keywords" containing the "+"-delimited keywords.
272 NOTE: As of version 1.5, the array of parameter names returned will be
274 in the same order as they were submitted by the browser. Usually this
275 order is the same as the order in which the parameters are defined in
276 the form (however, this isn't part of the spec, and so isn't
277 guaranteed).
278 FETCHING THE VALUE OR VALUES OF A SINGLE NAMED PARAMETER:
280 @values = $query->param('foo');
281 -or-
283 $value = $query->param('foo');
285 Pass the param() method a single argument to fetch the value of the
287 named parameter. If the parameter is multivalued (e.g. from multiple
288 selections in a scrolling list), you can ask to receive an array.
289 Otherwise the method will return a single value.
290 If a value is not given in the query string, as in the queries
292 "name1=&name2=", it will be returned as an empty string.
293 If the parameter does not exist at all, then param() will return undef
295 in a scalar context, and the empty list in a list context.
296 SETTING THE VALUE(S) OF A NAMED PARAMETER:
298 $query->param('foo','an','array','of','values');
299 This sets the value for the named parameter 'foo' to an array of
301 values. This is one way to change the value of a field AFTER the
302 script has been invoked once before. (Another way is with the
303 -override parameter accepted by all methods that generate form
304 elements.)
305 param() also recognizes a named parameter style of calling described in
307 more detail later:
308 $query->param(-name=>'foo',-values=>['an','array','of','values']);
310 -or-
312 $query->param(-name=>'foo',-value=>'the value');
314 APPENDING ADDITIONAL VALUES TO A NAMED PARAMETER:
316 $query->append(-name=>'foo',-values=>['yet','more','values']);
317 This adds a value or list of values to the named parameter. The values
319 are appended to the end of the parameter if it already exists.
320 Otherwise the parameter is created. Note that this method only
321 recognizes the named argument calling syntax.
322 IMPORTING ALL PARAMETERS INTO A NAMESPACE:
324 $query->import_names('R');
325 This creates a series of variables in the 'R' namespace. For example,
327 $R::foo, @R:foo. For keyword lists, a variable @R::keywords will
328 appear. If no namespace is given, this method will assume 'Q'.
329 WARNING: don't import anything into 'main'; this is a major security
330 risk!!!!
331 NOTE 1: Variable names are transformed as necessary into legal Perl
333 variable names. All non-legal characters are transformed into
334 underscores. If you need to keep the original names, you should use
335 the param() method instead to access CGI variables by name.
336 NOTE 2: In older versions, this method was called import(). As of
338 version 2.20, this name has been removed completely to avoid conflict
339 with the built-in Perl module import operator.
340 DELETING A PARAMETER COMPLETELY:
342 $query->delete('foo','bar','baz');
343 This completely clears a list of parameters. It sometimes useful for
345 resetting parameters that you don't want passed down between script
346 invocations.
347 If you are using the function call interface, use "Delete()" instead to
349 avoid conflicts with Perl's built-in delete operator.
350 DELETING ALL PARAMETERS:
352 $query->delete_all();
353 This clears the CGI object completely. It might be useful to ensure
355 that all the defaults are taken when you create a fill-out form.
356 Use Delete_all() instead if you are using the function call interface.
358 HANDLING NON-URLENCODED ARGUMENTS
360 If POSTed data is not of type application/x-www-form-urlencoded or
361 multipart/form-data, then the POSTed data will not be processed, but
362 instead be returned as-is in a parameter named POSTDATA. To retrieve
363 it, use code like this:
364 my $data = $query->param('POSTDATA');
366 Likewise if PUTed data can be retrieved with code like this:
368 my $data = $query->param('PUTDATA');
370 (If you don't know what the preceding means, don't worry about it. It
372 only affects people trying to use CGI for XML processing and other
373 specialized tasks.)
374 DIRECT ACCESS TO THE PARAMETER LIST:
376 $q->param_fetch('address')->[1] = '1313 Mockingbird Lane';
377 unshift @{$q->param_fetch(-name=>'address')},'George Munster';
378 If you need access to the parameter list in a way that isn't covered by
380 the methods given in the previous sections, you can obtain a direct
381 reference to it by calling the param_fetch() method with the name of
382 the parameter. This will return an array reference to the named
383 parameter, which you then can manipulate in any way you like.
384 You can also use a named argument style using the -name argument.
386 FETCHING THE PARAMETER LIST AS A HASH:
388 $params = $q->Vars;
389 print $params->{'address'};
390 @foo = split("\0",$params->{'foo'});
391 %params = $q->Vars;
392 use CGI ':cgi-lib';
394 $params = Vars;
395 Many people want to fetch the entire parameter list as a hash in which
397 the keys are the names of the CGI parameters, and the values are the
398 parameters' values. The Vars() method does this. Called in a scalar
399 context, it returns the parameter list as a tied hash reference.
400 Changing a key changes the value of the parameter in the underlying CGI
401 parameter list. Called in a list context, it returns the parameter
402 list as an ordinary hash. This allows you to read the contents of the
403 parameter list, but not to change it.
404 When using this, the thing you must watch out for are multivalued CGI
406 parameters. Because a hash cannot distinguish between scalar and list
407 context, multivalued parameters will be returned as a packed string,
408 separated by the "\0" (null) character. You must split this packed
409 string in order to get at the individual values. This is the
410 convention introduced long ago by Steve Brenner in his cgi-lib.pl
411 module for Perl version 4.
412 If you wish to use Vars() as a function, import the :cgi-lib set of
414 function calls (also see the section on CGI-LIB compatibility).
415 SAVING THE STATE OF THE SCRIPT TO A FILE:
417 $query->save(\*FILEHANDLE)
418 This will write the current state of the form to the provided
420 filehandle. You can read it back in by providing a filehandle to the
421 new() method. Note that the filehandle can be a file, a pipe, or
422 whatever!
423 The format of the saved file is:
425 NAME1=VALUE1
427 NAME1=VALUE1'
428 NAME2=VALUE2
429 NAME3=VALUE3
430 =
431 Both name and value are URL escaped. Multi-valued CGI parameters are
433 represented as repeated names. A session record is delimited by a
434 single = symbol. You can write out multiple records and read them back
435 in with several calls to new. You can do this across several sessions
436 by opening the file in append mode, allowing you to create primitive
437 guest books, or to keep a history of users' queries. Here's a short
438 example of creating multiple session records:
439 use CGI;
441 open (OUT,'>>','test.out') || die;
443 $records = 5;
444 for (0..$records) {
445 my $q = CGI->new;
446 $q->param(-name=>'counter',-value=>$_);
447 $q->save(\*OUT);
448 }
449 close OUT;
450 # reopen for reading
452 open (IN,'<','test.out') || die;
453 while (!eof(IN)) {
454 my $q = CGI->new(\*IN);
455 print $q->param('counter'),"\n";
456 }
457 The file format used for save/restore is identical to that used by the
459 Whitehead Genome Center's data exchange format "Boulderio", and can be
460 manipulated and even databased using Boulderio utilities. See
461 http://stein.cshl.org/boulder/
463 for further details.
465 If you wish to use this method from the function-oriented (non-OO)
467 interface, the exported name for this method is save_parameters().
468 RETRIEVING CGI ERRORS
470 Errors can occur while processing user input, particularly when
471 processing uploaded files. When these errors occur, CGI will stop
472 processing and return an empty parameter list. You can test for the
473 existence and nature of errors using the cgi_error() function. The
474 error messages are formatted as HTTP status codes. You can either
475 incorporate the error text into an HTML page, or use it as the value of
476 the HTTP status:
477 my $error = $q->cgi_error;
479 if ($error) {
480 print $q->header(-status=>$error),
481 $q->start_html('Problems'),
482 $q->h2('Request not processed'),
483 $q->strong($error);
484 exit 0;
485 }
486 When using the function-oriented interface (see the next section),
488 errors may only occur the first time you call param(). Be ready for
489 this!
490 USING THE FUNCTION-ORIENTED INTERFACE
492 To use the function-oriented interface, you must specify which CGI.pm
493 routines or sets of routines to import into your script's namespace.
494 There is a small overhead associated with this importation, but it
495 isn't much.
496 use CGI <list of methods>;
498 The listed methods will be imported into the current package; you can
500 call them directly without creating a CGI object first. This example
501 shows how to import the param() and header() methods, and then use them
502 directly:
503 use CGI 'param','header';
505 print header('text/plain');
506 $zipcode = param('zipcode');
507 More frequently, you'll import common sets of functions by referring to
509 the groups by name. All function sets are preceded with a ":"
510 character as in ":html3" (for tags defined in the HTML 3 standard).
511 Here is a list of the function sets you can import:
513 :cgi
515 Import all CGI-handling methods, such as param(), path_info() and
516 the like.
517 :form
519 Import all fill-out form generating methods, such as textfield().
520 :html2
522 Import all methods that generate HTML 2.0 standard elements.
523 :html3
525 Import all methods that generate HTML 3.0 elements (such as
526 <table>, <super> and <sub>).
527 :html4
529 Import all methods that generate HTML 4 elements (such as <abbrev>,
530 <acronym> and <thead>).
531 :netscape
533 Import the <blink>, <fontsize> and <center> tags.
534 :html
536 Import all HTML-generating shortcuts (i.e. 'html2', 'html3',
537 'html4' and 'netscape')
538 :standard
540 Import "standard" features, 'html2', 'html3', 'html4', 'form' and
541 'cgi'.
542 :all
544 Import all the available methods. For the full list, see the
545 CGI.pm code, where the variable %EXPORT_TAGS is defined.
546 If you import a function name that is not part of CGI.pm, the module
548 will treat it as a new HTML tag and generate the appropriate
549 subroutine. You can then use it like any other HTML tag. This is to
550 provide for the rapidly-evolving HTML "standard." For example, say
551 Microsoft comes out with a new tag called <gradient> (which causes the
552 user's desktop to be flooded with a rotating gradient fill until his
553 machine reboots). You don't need to wait for a new version of CGI.pm
554 to start using it immediately:
555 use CGI qw/:standard :html3 gradient/;
557 print gradient({-start=>'red',-end=>'blue'});
558 Note that in the interests of execution speed CGI.pm does not use the
560 standard Exporter syntax for specifying load symbols. This may change
561 in the future.
562 If you import any of the state-maintaining CGI or form-generating
564 methods, a default CGI object will be created and initialized
565 automatically the first time you use any of the methods that require
566 one to be present. This includes param(), textfield(), submit() and
567 the like. (If you need direct access to the CGI object, you can find
568 it in the global variable $CGI::Q). By importing CGI.pm methods, you
569 can create visually elegant scripts:
570 use CGI qw/:standard/;
572 print
573 header,
574 start_html('Simple Script'),
575 h1('Simple Script'),
576 start_form,
577 "What's your name? ",textfield('name'),p,
578 "What's the combination?",
579 checkbox_group(-name=>'words',
580 -values=>['eenie','meenie','minie','moe'],
581 -defaults=>['eenie','moe']),p,
582 "What's your favorite color?",
583 popup_menu(-name=>'color',
584 -values=>['red','green','blue','chartreuse']),p,
585 submit,
586 end_form,
587 hr,"\n";
588 if (param) {
590 print
591 "Your name is ",em(param('name')),p,
592 "The keywords are: ",em(join(", ",param('words'))),p,
593 "Your favorite color is ",em(param('color')),".\n";
594 }
595 print end_html;
596 PRAGMAS
598 In addition to the function sets, there are a number of pragmas that
599 you can import. Pragmas, which are always preceded by a hyphen, change
600 the way that CGI.pm functions in various ways. Pragmas, function sets,
601 and individual functions can all be imported in the same use() line.
602 For example, the following use statement imports the standard set of
603 functions and enables debugging mode (pragma -debug):
604 use CGI qw/:standard -debug/;
606 The current list of pragmas is as follows:
608 -any
610 When you use CGI -any, then any method that the query object
611 doesn't recognize will be interpreted as a new HTML tag. This
612 allows you to support the next ad hoc HTML extension. This lets
613 you go wild with new and unsupported tags:
614 use CGI qw(-any);
616 $q=CGI->new;
617 print $q->gradient({speed=>'fast',start=>'red',end=>'blue'});
618 Since using <cite>any</cite> causes any mistyped method name to be
620 interpreted as an HTML tag, use it with care or not at all.
621 -compile
623 This causes the indicated autoloaded methods to be compiled up
624 front, rather than deferred to later. This is useful for scripts
625 that run for an extended period of time under FastCGI or mod_perl,
626 and for those destined to be crunched by Malcolm Beattie's Perl
627 compiler. Use it in conjunction with the methods or method
628 families you plan to use.
629 use CGI qw(-compile :standard :html3);
631 or even
633 use CGI qw(-compile :all);
635 Note that using the -compile pragma in this way will always have
637 the effect of importing the compiled functions into the current
638 namespace. If you want to compile without importing use the
639 compile() method instead:
640 use CGI();
642 CGI->compile();
643 This is particularly useful in a mod_perl environment, in which you
645 might want to precompile all CGI routines in a startup script, and
646 then import the functions individually in each mod_perl script.
647 -nosticky
649 By default the CGI module implements a state-preserving behavior
650 called "sticky" fields. The way this works is that if you are
651 regenerating a form, the methods that generate the form field
652 values will interrogate param() to see if similarly-named
653 parameters are present in the query string. If they find a like-
654 named parameter, they will use it to set their default values.
655 Sometimes this isn't what you want. The -nosticky pragma prevents
657 this behavior. You can also selectively change the sticky behavior
658 in each element that you generate.
659 -tabindex
661 Automatically add tab index attributes to each form field. With
662 this option turned off, you can still add tab indexes manually by
663 passing a -tabindex option to each field-generating method.
664 -no_undef_params
666 This keeps CGI.pm from including undef params in the parameter
667 list.
668 -no_xhtml
670 By default, CGI.pm versions 2.69 and higher emit XHTML
671 (http://www.w3.org/TR/xhtml1/). The -no_xhtml pragma disables this
672 feature. Thanks to Michalis Kabrianis <kabrianis@hellug.gr> for
673 this feature.
674 If start_html()'s -dtd parameter specifies an HTML 2.0, 3.2, 4.0 or
676 4.01 DTD, XHTML will automatically be disabled without needing to
677 use this pragma.
678 -utf8
680 This makes CGI.pm treat all parameters as UTF-8 strings. Use this
681 with care, as it will interfere with the processing of binary
682 uploads. It is better to manually select which fields are expected
683 to return utf-8 strings and convert them using code like this:
684 use Encode;
686 my $arg = decode utf8=>param('foo');
687 -nph
689 This makes CGI.pm produce a header appropriate for an NPH (no
690 parsed header) script. You may need to do other things as well to
691 tell the server that the script is NPH. See the discussion of NPH
692 scripts below.
693 -newstyle_urls
695 Separate the name=value pairs in CGI parameter query strings with
696 semicolons rather than ampersands. For example:
697 ?name=fred;age=24;favorite_color=3
699 Semicolon-delimited query strings are always accepted, and will be
701 emitted by self_url() and query_string(). newstyle_urls became the
702 default in version 2.64.
703 -oldstyle_urls
705 Separate the name=value pairs in CGI parameter query strings with
706 ampersands rather than semicolons. This is no longer the default.
707 -autoload
709 This overrides the autoloader so that any function in your program
710 that is not recognized is referred to CGI.pm for possible
711 evaluation. This allows you to use all the CGI.pm functions
712 without adding them to your symbol table, which is of concern for
713 mod_perl users who are worried about memory consumption. Warning:
714 when -autoload is in effect, you cannot use "poetry mode"
715 (functions without the parenthesis). Use hr() rather than hr, or
716 add something like use subs qw/hr p header/ to the top of your
717 script.
718 -no_debug
720 This turns off the command-line processing features. If you want
721 to run a CGI.pm script from the command line to produce HTML, and
722 you don't want it to read CGI parameters from the command line or
723 STDIN, then use this pragma:
724 use CGI qw(-no_debug :standard);
726 -debug
728 This turns on full debugging. In addition to reading CGI arguments
729 from the command-line processing, CGI.pm will pause and try to read
730 arguments from STDIN, producing the message "(offline mode: enter
731 name=value pairs on standard input)" features.
732 See the section on debugging for more details.
734 -private_tempfiles
736 CGI.pm can process uploaded file. Ordinarily it spools the uploaded
737 file to a temporary directory, then deletes the file when done.
738 However, this opens the risk of eavesdropping as described in the
739 file upload section. Another CGI script author could peek at this
740 data during the upload, even if it is confidential information. On
741 Unix systems, the -private_tempfiles pragma will cause the
742 temporary file to be unlinked as soon as it is opened and before
743 any data is written into it, reducing, but not eliminating the risk
744 of eavesdropping (there is still a potential race condition). To
745 make life harder for the attacker, the program chooses tempfile
746 names by calculating a 32 bit checksum of the incoming HTTP
747 headers.
748 To ensure that the temporary file cannot be read by other CGI
750 scripts, use suEXEC or a CGI wrapper program to run your script.
751 The temporary file is created with mode 0600 (neither world nor
752 group readable).
753 The temporary directory is selected using the following algorithm:
755 1. if $CGITempFile::TMPDIRECTORY is already set, use that
757 2. if the environment variable TMPDIR exists, use the location
759 indicated.
760 3. Otherwise try the locations /usr/tmp, /var/tmp, C:\temp,
762 /tmp, /temp, ::Temporary Items, and \WWW_ROOT.
763 Each of these locations is checked that it is a directory and is
765 writable. If not, the algorithm tries the next choice.
766 SPECIAL FORMS FOR IMPORTING HTML-TAG FUNCTIONS
768 Many of the methods generate HTML tags. As described below, tag
769 functions automatically generate both the opening and closing tags.
770 For example:
771 print h1('Level 1 Header');
773 produces
775 <h1>Level 1 Header</h1>
777 There will be some times when you want to produce the start and end
779 tags yourself. In this case, you can use the form start_tag_name and
780 end_tag_name, as in:
781 print start_h1,'Level 1 Header',end_h1;
783 With a few exceptions (described below), start_tag_name and
785 end_tag_name functions are not generated automatically when you use
786 CGI. However, you can specify the tags you want to generate start/end
787 functions for by putting an asterisk in front of their name, or,
788 alternatively, requesting either "start_tag_name" or "end_tag_name" in
789 the import list.
790 Example:
792 use CGI qw/:standard *table start_ul/;
794 In this example, the following functions are generated in addition to
796 the standard ones:
797 1. start_table() (generates a <table> tag)
799 2. end_table() (generates a </table> tag)
800 3. start_ul() (generates a <ul> tag)
801 4. end_ul() (generates a </ul> tag)
802
804 Most of CGI.pm's functions deal with creating documents on the fly.
805 Generally you will produce the HTTP header first, followed by the
806 document itself. CGI.pm provides functions for generating HTTP headers
807 of various types as well as for generating HTML. For creating GIF
808 images, see the GD.pm module.
809 Each of these functions produces a fragment of HTML or HTTP which you
811 can print out directly so that it displays in the browser window,
812 append to a string, or save to a file for later use.
813 CREATING A STANDARD HTTP HEADER:
815 Normally the first thing you will do in any CGI script is print out an
816 HTTP header. This tells the browser what type of document to expect,
817 and gives other optional information, such as the language, expiration
818 date, and whether to cache the document. The header can also be
819 manipulated for special purposes, such as server push and pay per view
820 pages.
821 print header;
823 -or-
825 print header('image/gif');
827 -or-
829 print header('text/html','204 No response');
831 -or-
833 print header(-type=>'image/gif',
835 -nph=>1,
836 -status=>'402 Payment required',
837 -expires=>'+3d',
838 -cookie=>$cookie,
839 -charset=>'utf-7',
840 -attachment=>'foo.gif',
841 -Cost=>'$2.00');
842 header() returns the Content-type: header. You can provide your own
844 MIME type if you choose, otherwise it defaults to text/html. An
845 optional second parameter specifies the status code and a human-
846 readable message. For example, you can specify 204, "No response" to
847 create a script that tells the browser to do nothing at all.
848 The last example shows the named argument style for passing arguments
850 to the CGI methods using named parameters. Recognized parameters are
851 -type, -status, -expires, and -cookie. Any other named parameters will
852 be stripped of their initial hyphens and turned into header fields,
853 allowing you to specify any HTTP header you desire. Internal
854 underscores will be turned into hyphens:
855 print header(-Content_length=>3002);
857 Most browsers will not cache the output from CGI scripts. Every time
859 the browser reloads the page, the script is invoked anew. You can
860 change this behavior with the -expires parameter. When you specify an
861 absolute or relative expiration interval with this parameter, some
862 browsers and proxy servers will cache the script's output until the
863 indicated expiration date. The following forms are all valid for the
864 -expires field:
865 +30s 30 seconds from now
867 +10m ten minutes from now
868 +1h one hour from now
869 -1d yesterday (i.e. "ASAP!")
870 now immediately
871 +3M in three months
872 +10y in ten years time
873 Thursday, 25-Apr-1999 00:40:33 GMT at the indicated time & date
874 The -cookie parameter generates a header that tells the browser to
876 provide a "magic cookie" during all subsequent transactions with your
877 script. Some cookies have a special format that includes interesting
878 attributes such as expiration time. Use the cookie() method to create
879 and retrieve session cookies.
880 The -nph parameter, if set to a true value, will issue the correct
882 headers to work with a NPH (no-parse-header) script. This is important
883 to use with certain servers that expect all their scripts to be NPH.
884 The -charset parameter can be used to control the character set sent to
886 the browser. If not provided, defaults to ISO-8859-1. As a side
887 effect, this sets the charset() method as well.
888 The -attachment parameter can be used to turn the page into an
890 attachment. Instead of displaying the page, some browsers will prompt
891 the user to save it to disk. The value of the argument is the
892 suggested name for the saved file. In order for this to work, you may
893 have to set the -type to "application/octet-stream".
894 The -p3p parameter will add a P3P tag to the outgoing header. The
896 parameter can be an arrayref or a space-delimited string of P3P tags.
897 For example:
898 print header(-p3p=>[qw(CAO DSP LAW CURa)]);
900 print header(-p3p=>'CAO DSP LAW CURa');
901 In either case, the outgoing header will be formatted as:
903 P3P: policyref="/w3c/p3p.xml" cp="CAO DSP LAW CURa"
905 Note that if a header value contains a carriage return, a leading space
907 will be added to each new line that doesn't already have one as
908 specified by RFC2616 section 4.2. For example:
909 print header( -ingredients => "ham\neggs\nbacon" );
911 will generate
913 Ingredients: ham
915 eggs
916 bacon
917 GENERATING A REDIRECTION HEADER
919 print $q->redirect('http://somewhere.else/in/movie/land');
920 Sometimes you don't want to produce a document yourself, but simply
922 redirect the browser elsewhere, perhaps choosing a URL based on the
923 time of day or the identity of the user.
924 The redirect() method redirects the browser to a different URL. If you
926 use redirection like this, you should not print out a header as well.
927 You should always use full URLs (including the http: or ftp: part) in
929 redirection requests. Relative URLs will not work correctly.
930 You can also use named arguments:
932 print $q->redirect(
934 -uri=>'http://somewhere.else/in/movie/land',
935 -nph=>1,
936 -status=>301);
937 All names arguments recognized by header() are also recognized by
939 redirect(). However, most HTTP headers, including those generated by
940 -cookie and -target, are ignored by the browser.
941 The -nph parameter, if set to a true value, will issue the correct
943 headers to work with a NPH (no-parse-header) script. This is important
944 to use with certain servers, such as Microsoft IIS, which expect all
945 their scripts to be NPH.
946 The -status parameter will set the status of the redirect. HTTP
948 defines three different possible redirection status codes:
949 301 Moved Permanently
951 302 Found
952 303 See Other
953 The default if not specified is 302, which means "moved temporarily."
955 You may change the status to another status code if you wish. Be
956 advised that changing the status to anything other than 301, 302 or 303
957 will probably break redirection.
958 CREATING THE HTML DOCUMENT HEADER
960 print start_html(-title=>'Secrets of the Pyramids',
961 -author=>'fred@capricorn.org',
962 -base=>'true',
963 -target=>'_blank',
964 -meta=>{'keywords'=>'pharaoh secret mummy',
965 'copyright'=>'copyright 1996 King Tut'},
966 -style=>{'src'=>'/styles/style1.css'},
967 -BGCOLOR=>'blue');
968 The start_html() routine creates the top of the page, along with a lot
970 of optional information that controls the page's appearance and
971 behavior.
972 This method returns a canned HTML header and the opening <body> tag.
974 All parameters are optional. In the named parameter form, recognized
975 parameters are -title, -author, -base, -xbase, -dtd, -lang and -target
976 (see below for the explanation). Any additional parameters you
977 provide, such as the unofficial BGCOLOR attribute, are added to the
978 <body> tag. Additional parameters must be proceeded by a hyphen.
979 The argument -xbase allows you to provide an HREF for the <base> tag
981 different from the current location, as in
982 -xbase=>"http://home.mcom.com/"
984 All relative links will be interpreted relative to this tag.
986 The argument -target allows you to provide a default target frame for
988 all the links and fill-out forms on the page. This is a non-standard
989 HTTP feature which only works with some browsers!
990 -target=>"answer_window"
992 All relative links will be interpreted relative to this tag. You add
994 arbitrary meta information to the header with the -meta argument. This
995 argument expects a reference to a hash containing name/value pairs of
996 meta information. These will be turned into a series of header <meta>
997 tags that look something like this:
998 <meta name="keywords" content="pharaoh secret mummy">
1000 <meta name="description" content="copyright 1996 King Tut">
1001 To create an HTTP-EQUIV type of <meta> tag, use -head, described below.
1003 The -style argument is used to incorporate cascading stylesheets into
1005 your code. See the section on CASCADING STYLESHEETS for more
1006 information.
1007 The -lang argument is used to incorporate a language attribute into the
1009 <html> tag. For example:
1010 print $q->start_html(-lang=>'fr-CA');
1012 The default if not specified is "en-US" for US English, unless the -dtd
1014 parameter specifies an HTML 2.0 or 3.2 DTD, in which case the lang
1015 attribute is left off. You can force the lang attribute to left off in
1016 other cases by passing an empty string (-lang=>'').
1017 The -encoding argument can be used to specify the character set for
1019 XHTML. It defaults to iso-8859-1 if not specified.
1020 The -dtd argument can be used to specify a public DTD identifier
1022 string. For example:
1023 -dtd => '-//W3C//DTD HTML 4.01 Transitional//EN')
1025 Alternatively, it can take public and system DTD identifiers as an
1027 array:
1028 dtd => [ '-//W3C//DTD HTML 4.01 Transitional//EN', 'http://www.w3.org/TR/html4/loose.dtd' ]
1030 For the public DTD identifier to be considered, it must be valid.
1032 Otherwise it will be replaced by the default DTD. If the public DTD
1033 contains 'XHTML', CGI.pm will emit XML.
1034 The -declare_xml argument, when used in conjunction with XHTML, will
1036 put a <?xml> declaration at the top of the HTML header. The sole
1037 purpose of this declaration is to declare the character set encoding.
1038 In the absence of -declare_xml, the output HTML will contain a <meta>
1039 tag that specifies the encoding, allowing the HTML to pass most
1040 validators. The default for -declare_xml is false.
1041 You can place other arbitrary HTML elements to the <head> section with
1043 the -head tag. For example, to place a <link> element in the head
1044 section, use this:
1045 print start_html(-head=>Link({-rel=>'shortcut icon',
1047 -href=>'favicon.ico'}));
1048 To incorporate multiple HTML elements into the <head> section, just
1050 pass an array reference:
1051 print start_html(-head=>[
1053 Link({-rel=>'next',
1054 -href=>'http://www.capricorn.com/s2.html'}),
1055 Link({-rel=>'previous',
1056 -href=>'http://www.capricorn.com/s1.html'})
1057 ]
1058 );
1059 And here's how to create an HTTP-EQUIV <meta> tag:
1061 print start_html(-head=>meta({-http_equiv => 'Content-Type',
1063 -content => 'text/html'}))
1064 JAVASCRIPTING: The -script, -noScript, -onLoad, -onMouseOver,
1066 -onMouseOut and -onUnload parameters are used to add JavaScript calls
1067 to your pages. -script should point to a block of text containing
1068 JavaScript function definitions. This block will be placed within a
1069 <script> block inside the HTML (not HTTP) header. The block is placed
1070 in the header in order to give your page a fighting chance of having
1071 all its JavaScript functions in place even if the user presses the stop
1072 button before the page has loaded completely. CGI.pm attempts to
1073 format the script in such a way that JavaScript-naive browsers will not
1074 choke on the code: unfortunately there are some browsers, such as
1075 Chimera for Unix, that get confused by it nevertheless.
1076 The -onLoad and -onUnload parameters point to fragments of JavaScript
1078 code to execute when the page is respectively opened and closed by the
1079 browser. Usually these parameters are calls to functions defined in
1080 the -script field:
1081 $query = CGI->new;
1083 print header;
1084 $JSCRIPT=<<END;
1085 // Ask a silly question
1086 function riddle_me_this() {
1087 var r = prompt("What walks on four legs in the morning, " +
1088 "two legs in the afternoon, " +
1089 "and three legs in the evening?");
1090 response(r);
1091 }
1092 // Get a silly answer
1093 function response(answer) {
1094 if (answer == "man")
1095 alert("Right you are!");
1096 else
1097 alert("Wrong! Guess again.");
1098 }
1099 END
1100 print start_html(-title=>'The Riddle of the Sphinx',
1101 -script=>$JSCRIPT);
1102 Use the -noScript parameter to pass some HTML text that will be
1104 displayed on browsers that do not have JavaScript (or browsers where
1105 JavaScript is turned off).
1106 The <script> tag, has several attributes including "type", "charset"
1108 and "src". "src" allows you to keep JavaScript code in an external
1109 file. To use these attributes pass a HASH reference in the -script
1110 parameter containing one or more of -type, -src, or -code:
1111 print $q->start_html(-title=>'The Riddle of the Sphinx',
1113 -script=>{-type=>'JAVASCRIPT',
1114 -src=>'/javascript/sphinx.js'}
1115 );
1116 print $q->(-title=>'The Riddle of the Sphinx',
1118 -script=>{-type=>'PERLSCRIPT',
1119 -code=>'print "hello world!\n;"'}
1120 );
1121 A final feature allows you to incorporate multiple <script> sections
1123 into the header. Just pass the list of script sections as an array
1124 reference. this allows you to specify different source files for
1125 different dialects of JavaScript. Example:
1126 print $q->start_html(-title=>'The Riddle of the Sphinx',
1128 -script=>[
1129 { -type => 'text/javascript',
1130 -src => '/javascript/utilities10.js'
1131 },
1132 { -type => 'text/javascript',
1133 -src => '/javascript/utilities11.js'
1134 },
1135 { -type => 'text/jscript',
1136 -src => '/javascript/utilities12.js'
1137 },
1138 { -type => 'text/ecmascript',
1139 -src => '/javascript/utilities219.js'
1140 }
1141 ]
1142 );
1143 The option "-language" is a synonym for -type, and is supported for
1145 backwards compatibility.
1146 The old-style positional parameters are as follows:
1148 Parameters:
1150 1. The title
1151 2. The author's e-mail address (will create a <link rev="MADE"> tag if
1153 present
1154 3. A 'true' flag if you want to include a <base> tag in the header.
1156 This helps resolve relative addresses to absolute ones when the
1157 document is moved, but makes the document hierarchy non-portable.
1158 Use with care!
1159 4, 5, 6...
1161 Any other parameters you want to include in the <body> tag. This
1162 is a good place to put HTML extensions, such as colors and
1163 wallpaper patterns.
1164 ENDING THE HTML DOCUMENT:
1166 print end_html
1167 This ends an HTML document by printing the </body></html> tags.
1169 CREATING A SELF-REFERENCING URL THAT PRESERVES STATE INFORMATION:
1171 $myself = self_url;
1172 print q(<a href="$myself">I'm talking to myself.</a>);
1173 self_url() will return a URL, that, when selected, will reinvoke this
1175 script with all its state information intact. This is most useful when
1176 you want to jump around within the document using internal anchors but
1177 you don't want to disrupt the current contents of the form(s).
1178 Something like this will do the trick.
1179 $myself = self_url;
1181 print "<a href=\"$myself#table1\">See table 1</a>";
1182 print "<a href=\"$myself#table2\">See table 2</a>";
1183 print "<a href=\"$myself#yourself\">See for yourself</a>";
1184 If you want more control over what's returned, using the url() method
1186 instead.
1187 You can also retrieve the unprocessed query string with query_string():
1189 $the_string = query_string;
1191 OBTAINING THE SCRIPT'S URL
1193 $full_url = url();
1194 $full_url = url(-full=>1); #alternative syntax
1195 $relative_url = url(-relative=>1);
1196 $absolute_url = url(-absolute=>1);
1197 $url_with_path = url(-path_info=>1);
1198 $url_with_path_and_query = url(-path_info=>1,-query=>1);
1199 $netloc = url(-base => 1);
1200 url() returns the script's URL in a variety of formats. Called without
1202 any arguments, it returns the full form of the URL, including host name
1203 and port number
1204 http://your.host.com/path/to/script.cgi
1206 You can modify this format with the following named arguments:
1208 -absolute
1210 If true, produce an absolute URL, e.g.
1211 /path/to/script.cgi
1213 -relative
1215 Produce a relative URL. This is useful if you want to reinvoke
1216 your script with different parameters. For example:
1217 script.cgi
1219 -full
1221 Produce the full URL, exactly as if called without any arguments.
1222 This overrides the -relative and -absolute arguments.
1223 -path (-path_info)
1225 Append the additional path information to the URL. This can be
1226 combined with -full, -absolute or -relative. -path_info is
1227 provided as a synonym.
1228 -query (-query_string)
1230 Append the query string to the URL. This can be combined with
1231 -full, -absolute or -relative. -query_string is provided as a
1232 synonym.
1233 -base
1235 Generate just the protocol and net location, as in
1236 http://www.foo.com:8000
1237 -rewrite
1239 If Apache's mod_rewrite is turned on, then the script name and path
1240 info probably won't match the request that the user sent. Set
1241 -rewrite=>1 (default) to return URLs that match what the user sent
1242 (the original request URI). Set -rewrite=>0 to return URLs that
1243 match the URL after mod_rewrite's rules have run. Because the
1244 additional path information only makes sense in the context of the
1245 rewritten URL, -rewrite is set to false when you request path info
1246 in the URL.
1247 MIXING POST AND URL PARAMETERS
1249 $color = url_param('color');
1250 It is possible for a script to receive CGI parameters in the URL as
1252 well as in the fill-out form by creating a form that POSTs to a URL
1253 containing a query string (a "?" mark followed by arguments). The
1254 param() method will always return the contents of the POSTed fill-out
1255 form, ignoring the URL's query string. To retrieve URL parameters,
1256 call the url_param() method. Use it in the same way as param(). The
1257 main difference is that it allows you to read the parameters, but not
1258 set them.
1259 Under no circumstances will the contents of the URL query string
1261 interfere with similarly-named CGI parameters in POSTed forms. If you
1262 try to mix a URL query string with a form submitted with the GET
1263 method, the results will not be what you expect.
1264
1266 CGI.pm defines general HTML shortcut methods for many HTML tags. HTML
1267 shortcuts are named after a single HTML element and return a fragment
1268 of HTML text. Example:
1269 print $q->blockquote(
1271 "Many years ago on the island of",
1272 $q->a({href=>"http://crete.org/"},"Crete"),
1273 "there lived a Minotaur named",
1274 $q->strong("Fred."),
1275 ),
1276 $q->hr;
1277 This results in the following HTML code (extra newlines have been added
1279 for readability):
1280 <blockquote>
1282 Many years ago on the island of
1283 <a href="http://crete.org/">Crete</a> there lived
1284 a minotaur named <strong>Fred.</strong>
1285 </blockquote>
1286 <hr>
1287 If you find the syntax for calling the HTML shortcuts awkward, you can
1289 import them into your namespace and dispense with the object syntax
1290 completely (see the next section for more details):
1291 use CGI ':standard';
1293 print blockquote(
1294 "Many years ago on the island of",
1295 a({href=>"http://crete.org/"},"Crete"),
1296 "there lived a minotaur named",
1297 strong("Fred."),
1298 ),
1299 hr;
1300 PROVIDING ARGUMENTS TO HTML SHORTCUTS
1302 The HTML methods will accept zero, one or multiple arguments. If you
1303 provide no arguments, you get a single tag:
1304 print hr; # <hr>
1306 If you provide one or more string arguments, they are concatenated
1308 together with spaces and placed between opening and closing tags:
1309 print h1("Chapter","1"); # <h1>Chapter 1</h1>"
1311 If the first argument is a hash reference, then the keys and values of
1313 the hash become the HTML tag's attributes:
1314 print a({-href=>'fred.html',-target=>'_new'},
1316 "Open a new frame");
1317 <a href="fred.html",target="_new">Open a new frame</a>
1319 You may dispense with the dashes in front of the attribute names if you
1321 prefer:
1322 print img {src=>'fred.gif',align=>'LEFT'};
1324 <img align="LEFT" src="fred.gif">
1326 Sometimes an HTML tag attribute has no argument. For example, ordered
1328 lists can be marked as COMPACT. The syntax for this is an argument
1329 that that points to an undef string:
1330 print ol({compact=>undef},li('one'),li('two'),li('three'));
1332 Prior to CGI.pm version 2.41, providing an empty ('') string as an
1334 attribute argument was the same as providing undef. However, this has
1335 changed in order to accommodate those who want to create tags of the
1336 form <img alt="">. The difference is shown in these two pieces of
1337 code:
1338 CODE RESULT
1340 img({alt=>undef}) <img alt>
1341 img({alt=>''}) <img alt="">
1342 THE DISTRIBUTIVE PROPERTY OF HTML SHORTCUTS
1344 One of the cool features of the HTML shortcuts is that they are
1345 distributive. If you give them an argument consisting of a reference
1346 to a list, the tag will be distributed across each element of the list.
1347 For example, here's one way to make an ordered list:
1348 print ul(
1350 li({-type=>'disc'},['Sneezy','Doc','Sleepy','Happy'])
1351 );
1352 This example will result in HTML output that looks like this:
1354 <ul>
1356 <li type="disc">Sneezy</li>
1357 <li type="disc">Doc</li>
1358 <li type="disc">Sleepy</li>
1359 <li type="disc">Happy</li>
1360 </ul>
1361 This is extremely useful for creating tables. For example:
1363 print table({-border=>undef},
1365 caption('When Should You Eat Your Vegetables?'),
1366 Tr({-align=>'CENTER',-valign=>'TOP'},
1367 [
1368 th(['Vegetable', 'Breakfast','Lunch','Dinner']),
1369 td(['Tomatoes' , 'no', 'yes', 'yes']),
1370 td(['Broccoli' , 'no', 'no', 'yes']),
1371 td(['Onions' , 'yes','yes', 'yes'])
1372 ]
1373 )
1374 );
1375 HTML SHORTCUTS AND LIST INTERPOLATION
1377 Consider this bit of code:
1378 print blockquote(em('Hi'),'mom!'));
1380 It will ordinarily return the string that you probably expect, namely:
1382 <blockquote><em>Hi</em> mom!</blockquote>
1384 Note the space between the element "Hi" and the element "mom!". CGI.pm
1386 puts the extra space there using array interpolation, which is
1387 controlled by the magic $" variable. Sometimes this extra space is not
1388 what you want, for example, when you are trying to align a series of
1389 images. In this case, you can simply change the value of $" to an
1390 empty string.
1391 {
1393 local($") = '';
1394 print blockquote(em('Hi'),'mom!'));
1395 }
1396 I suggest you put the code in a block as shown here. Otherwise the
1398 change to $" will affect all subsequent code until you explicitly reset
1399 it.
1400 NON-STANDARD HTML SHORTCUTS
1402 A few HTML tags don't follow the standard pattern for various reasons.
1403 comment() generates an HTML comment (<!-- comment -->). Call it like
1405 print comment('here is my comment');
1407 Because of conflicts with built-in Perl functions, the following
1409 functions begin with initial caps:
1410 Select
1412 Tr
1413 Link
1414 Delete
1415 Accept
1416 Sub
1417 In addition, start_html(), end_html(), start_form(), end_form(),
1419 start_multipart_form() and all the fill-out form tags are special. See
1420 their respective sections.
1421 AUTOESCAPING HTML
1423 By default, all HTML that is emitted by the form-generating functions
1424 is passed through a function called escapeHTML():
1425 $escaped_string = escapeHTML("unescaped string");
1427 Escape HTML formatting characters in a string.
1428 Provided that you have specified a character set of ISO-8859-1 (the
1430 default), the standard HTML escaping rules will be used. The "<"
1431 character becomes "<", ">" becomes ">", "&" becomes "&", and
1432 the quote character becomes """. In addition, the hexadecimal
1433 0x8b and 0x9b characters, which some browsers incorrectly interpret as
1434 the left and right angle-bracket characters, are replaced by their
1435 numeric character entities ("‹" and "›"). If you manually
1436 change the charset, either by calling the charset() method explicitly
1437 or by passing a -charset argument to header(), then all characters will
1438 be replaced by their numeric entities, since CGI.pm has no lookup table
1439 for all the possible encodings.
1440 "escapeHTML()" expects the supplied string to be a character string.
1442 This means you should Encode::decode data received from "outside" and
1443 Encode::encode your strings before sending them back outside. If your
1444 source code UTF-8 encoded and you want to upgrade string literals in
1445 your source to character strings, you can use "use utf8". See
1446 perlunitut, perlunifaq and perlunicode for more information on how Perl
1447 handles the difference between bytes and characters.
1448 The automatic escaping does not apply to other shortcuts, such as h1().
1450 You should call escapeHTML() yourself on untrusted data in order to
1451 protect your pages against nasty tricks that people may enter into
1452 guestbooks, etc.. To change the character set, use charset(). To turn
1453 autoescaping off completely, use autoEscape(0):
1454 $charset = charset([$charset]);
1456 Get or set the current character set.
1457 $flag = autoEscape([$flag]);
1459 Get or set the value of the autoescape flag.
1460 PRETTY-PRINTING HTML
1462 By default, all the HTML produced by these functions comes out as one
1463 long line without carriage returns or indentation. This is yuck, but it
1464 does reduce the size of the documents by 10-20%. To get pretty-printed
1465 output, please use CGI::Pretty, a subclass contributed by Brian
1466 Paulsen.
1467
1469 General note The various form-creating methods all return strings to
1470 the caller, containing the tag or tags that will create the requested
1471 form element. You are responsible for actually printing out these
1472 strings. It's set up this way so that you can place formatting tags
1473 around the form elements.
1474 Another note The default values that you specify for the forms are only
1476 used the first time the script is invoked (when there is no query
1477 string). On subsequent invocations of the script (when there is a
1478 query string), the former values are used even if they are blank.
1479 If you want to change the value of a field from its previous value, you
1481 have two choices:
1482 (1) call the param() method to set it.
1484 (2) use the -override (alias -force) parameter (a new feature in
1486 version 2.15). This forces the default value to be used, regardless of
1487 the previous value:
1488 print textfield(-name=>'field_name',
1490 -default=>'starting value',
1491 -override=>1,
1492 -size=>50,
1493 -maxlength=>80);
1494 Yet another note By default, the text and labels of form elements are
1496 escaped according to HTML rules. This means that you can safely use
1497 "<CLICK ME>" as the label for a button. However, it also interferes
1498 with your ability to incorporate special HTML character sequences, such
1499 as Á, into your fields. If you wish to turn off automatic
1500 escaping, call the autoEscape() method with a false value immediately
1501 after creating the CGI object:
1502 $query = CGI->new;
1504 $query->autoEscape(0);
1505 Note that autoEscape() is exclusively used to effect the behavior of
1507 how some CGI.pm HTML generation functions handle escaping. Calling
1508 escapeHTML() explicitly will always escape the HTML.
1509 A Lurking Trap! Some of the form-element generating methods return
1511 multiple tags. In a scalar context, the tags will be concatenated
1512 together with spaces, or whatever is the current value of the $"
1513 global. In a list context, the methods will return a list of elements,
1514 allowing you to modify them if you wish. Usually you will not notice
1515 this behavior, but beware of this:
1516 printf("%s\n",end_form())
1518 end_form() produces several tags, and only the first of them will be
1520 printed because the format only expects one value.
1521 <p>
1523 CREATING AN ISINDEX TAG
1525 print isindex(-action=>$action);
1526 -or-
1528 print isindex($action);
1530 Prints out an <isindex> tag. Not very exciting. The parameter -action
1532 specifies the URL of the script to process the query. The default is
1533 to process the query with the current script.
1534 STARTING AND ENDING A FORM
1536 print start_form(-method=>$method,
1537 -action=>$action,
1538 -enctype=>$encoding);
1539 <... various form stuff ...>
1540 print end_form;
1541 -or-
1543 print start_form($method,$action,$encoding);
1545 <... various form stuff ...>
1546 print end_form;
1547 start_form() will return a <form> tag with the optional method, action
1549 and form encoding that you specify. The defaults are:
1550 method: POST
1552 action: this script
1553 enctype: application/x-www-form-urlencoded for non-XHTML
1554 multipart/form-data for XHTML, see multipart/form-data below.
1555 end_form() returns the closing </form> tag.
1557 Start_form()'s enctype argument tells the browser how to package the
1559 various fields of the form before sending the form to the server. Two
1560 values are possible:
1561 Note: These methods were previously named startform() and endform().
1563 These methods are now DEPRECATED. Please use start_form() and
1564 end_form() instead.
1565 application/x-www-form-urlencoded
1567 This is the older type of encoding. It is compatible with many CGI
1568 scripts and is suitable for short fields containing text data. For
1569 your convenience, CGI.pm stores the name of this encoding type in
1570 &CGI::URL_ENCODED.
1571 multipart/form-data
1573 This is the newer type of encoding. It is suitable for forms that
1574 contain very large fields or that are intended for transferring
1575 binary data. Most importantly, it enables the "file upload"
1576 feature. For your convenience, CGI.pm stores the name of this
1577 encoding type in &CGI::MULTIPART
1578 Forms that use this type of encoding are not easily interpreted by
1580 CGI scripts unless they use CGI.pm or another library designed to
1581 handle them.
1582 If XHTML is activated (the default), then forms will be
1584 automatically created using this type of encoding.
1585 The start_form() method uses the older form of encoding by default
1587 unless XHTML is requested. If you want to use the newer form of
1588 encoding by default, you can call start_multipart_form() instead of
1589 start_form(). The method end_multipart_form() is an alias to
1590 end_form().
1591 JAVASCRIPTING: The -name and -onSubmit parameters are provided for use
1593 with JavaScript. The -name parameter gives the form a name so that it
1594 can be identified and manipulated by JavaScript functions. -onSubmit
1595 should point to a JavaScript function that will be executed just before
1596 the form is submitted to your server. You can use this opportunity to
1597 check the contents of the form for consistency and completeness. If
1598 you find something wrong, you can put up an alert box or maybe fix
1599 things up yourself. You can abort the submission by returning false
1600 from this function.
1601 Usually the bulk of JavaScript functions are defined in a <script>
1603 block in the HTML header and -onSubmit points to one of these function
1604 call. See start_html() for details.
1605 FORM ELEMENTS
1607 After starting a form, you will typically create one or more
1608 textfields, popup menus, radio groups and other form elements. Each of
1609 these elements takes a standard set of named arguments. Some elements
1610 also have optional arguments. The standard arguments are as follows:
1611 -name
1613 The name of the field. After submission this name can be used to
1614 retrieve the field's value using the param() method.
1615 -value, -values
1617 The initial value of the field which will be returned to the script
1618 after form submission. Some form elements, such as text fields,
1619 take a single scalar -value argument. Others, such as popup menus,
1620 take a reference to an array of values. The two arguments are
1621 synonyms.
1622 -tabindex
1624 A numeric value that sets the order in which the form element
1625 receives focus when the user presses the tab key. Elements with
1626 lower values receive focus first.
1627 -id A string identifier that can be used to identify this element to
1629 JavaScript and DHTML.
1630 -override
1632 A boolean, which, if true, forces the element to take on the value
1633 specified by -value, overriding the sticky behavior described
1634 earlier for the -nosticky pragma.
1635 -onChange, -onFocus, -onBlur, -onMouseOver, -onMouseOut, -onSelect
1637 These are used to assign JavaScript event handlers. See the
1638 JavaScripting section for more details.
1639 Other common arguments are described in the next section. In addition
1641 to these, all attributes described in the HTML specifications are
1642 supported.
1643 CREATING A TEXT FIELD
1645 print textfield(-name=>'field_name',
1646 -value=>'starting value',
1647 -size=>50,
1648 -maxlength=>80);
1649 -or-
1650 print textfield('field_name','starting value',50,80);
1652 textfield() will return a text input field.
1654 Parameters
1656 1. The first parameter is the required name for the field (-name).
1657 2. The optional second parameter is the default starting value for the
1659 field contents (-value, formerly known as -default).
1660 3. The optional third parameter is the size of the field in
1662 characters (-size).
1663 4. The optional fourth parameter is the maximum number of characters
1665 the
1666 field will accept (-maxlength).
1667 As with all these methods, the field will be initialized with its
1669 previous contents from earlier invocations of the script. When the
1670 form is processed, the value of the text field can be retrieved with:
1671 $value = param('foo');
1673 If you want to reset it from its initial value after the script has
1675 been called once, you can do so like this:
1676 param('foo',"I'm taking over this value!");
1678 CREATING A BIG TEXT FIELD
1680 print textarea(-name=>'foo',
1681 -default=>'starting value',
1682 -rows=>10,
1683 -columns=>50);
1684 -or
1686 print textarea('foo','starting value',10,50);
1688 textarea() is just like textfield, but it allows you to specify rows
1690 and columns for a multiline text entry box. You can provide a starting
1691 value for the field, which can be long and contain multiple lines.
1692 CREATING A PASSWORD FIELD
1694 print password_field(-name=>'secret',
1695 -value=>'starting value',
1696 -size=>50,
1697 -maxlength=>80);
1698 -or-
1699 print password_field('secret','starting value',50,80);
1701 password_field() is identical to textfield(), except that its contents
1703 will be starred out on the web page.
1704 CREATING A FILE UPLOAD FIELD
1706 print filefield(-name=>'uploaded_file',
1707 -default=>'starting value',
1708 -size=>50,
1709 -maxlength=>80);
1710 -or-
1711 print filefield('uploaded_file','starting value',50,80);
1713 filefield() will return a file upload field. In order to take full
1715 advantage of this you must use the new multipart encoding scheme for
1716 the form. You can do this either by calling start_form() with an
1717 encoding type of &CGI::MULTIPART, or by calling the new method
1718 start_multipart_form() instead of vanilla start_form().
1719 Parameters
1721 1. The first parameter is the required name for the field (-name).
1722 2. The optional second parameter is the starting value for the field
1724 contents to be used as the default file name (-default).
1725 For security reasons, browsers don't pay any attention to this
1727 field, and so the starting value will always be blank. Worse, the
1728 field loses its "sticky" behavior and forgets its previous
1729 contents. The starting value field is called for in the HTML
1730 specification, however, and possibly some browser will eventually
1731 provide support for it.
1732 3. The optional third parameter is the size of the field in characters
1734 (-size).
1735 4. The optional fourth parameter is the maximum number of characters
1737 the field will accept (-maxlength).
1738 JAVASCRIPTING: The -onChange, -onFocus, -onBlur, -onMouseOver,
1740 -onMouseOut and -onSelect parameters are recognized. See textfield()
1741 for details.
1742 PROCESSING A FILE UPLOAD FIELD
1744 Basics
1745 When the form is processed, you can retrieve an IO::Handle compatible
1747 handle for a file upload field like this:
1748 $lightweight_fh = $q->upload('field_name');
1750 # undef may be returned if it's not a valid file handle
1752 if (defined $lightweight_fh) {
1753 # Upgrade the handle to one compatible with IO::Handle:
1754 my $io_handle = $lightweight_fh->handle;
1755 open (OUTFILE,'>>','/usr/local/web/users/feedback');
1757 while ($bytesread = $io_handle->read($buffer,1024)) {
1758 print OUTFILE $buffer;
1759 }
1760 }
1761 In a list context, upload() will return an array of filehandles. This
1763 makes it possible to process forms that use the same name for multiple
1764 upload fields.
1765 If you want the entered file name for the file, you can just call
1767 param():
1768 $filename = $q->param('field_name');
1770 Different browsers will return slightly different things for the name.
1772 Some browsers return the filename only. Others return the full path to
1773 the file, using the path conventions of the user's machine.
1774 Regardless, the name returned is always the name of the file on the
1775 user's machine, and is unrelated to the name of the temporary file that
1776 CGI.pm creates during upload spooling (see below).
1777 When a file is uploaded the browser usually sends along some
1779 information along with it in the format of headers. The information
1780 usually includes the MIME content type. To retrieve this information,
1781 call uploadInfo(). It returns a reference to a hash containing all the
1782 document headers.
1783 $filename = $q->param('uploaded_file');
1785 $type = $q->uploadInfo($filename)->{'Content-Type'};
1786 unless ($type eq 'text/html') {
1787 die "HTML FILES ONLY!";
1788 }
1789 If you are using a machine that recognizes "text" and "binary" data
1791 modes, be sure to understand when and how to use them (see the Camel
1792 book). Otherwise you may find that binary files are corrupted during
1793 file uploads.
1794 Accessing the temp files directly
1796 When processing an uploaded file, CGI.pm creates a temporary file on
1798 your hard disk and passes you a file handle to that file. After you are
1799 finished with the file handle, CGI.pm unlinks (deletes) the temporary
1800 file. If you need to you can access the temporary file directly. You
1801 can access the temp file for a file upload by passing the file name to
1802 the tmpFileName() method:
1803 $filename = $query->param('uploaded_file');
1805 $tmpfilename = $query->tmpFileName($filename);
1806 The temporary file will be deleted automatically when your program
1808 exits unless you manually rename it. On some operating systems (such as
1809 Windows NT), you will need to close the temporary file's filehandle
1810 before your program exits. Otherwise the attempt to delete the
1811 temporary file will fail.
1812 Handling interrupted file uploads
1814 There are occasionally problems involving parsing the uploaded file.
1816 This usually happens when the user presses "Stop" before the upload is
1817 finished. In this case, CGI.pm will return undef for the name of the
1818 uploaded file and set cgi_error() to the string "400 Bad request
1819 (malformed multipart POST)". This error message is designed so that
1820 you can incorporate it into a status code to be sent to the browser.
1821 Example:
1822 $file = $q->upload('uploaded_file');
1824 if (!$file && $q->cgi_error) {
1825 print $q->header(-status=>$q->cgi_error);
1826 exit 0;
1827 }
1828 You are free to create a custom HTML page to complain about the error,
1830 if you wish.
1831 Progress bars for file uploads and avoiding temp files
1833 CGI.pm gives you low-level access to file upload management through a
1835 file upload hook. You can use this feature to completely turn off the
1836 temp file storage of file uploads, or potentially write your own file
1837 upload progress meter.
1838 This is much like the UPLOAD_HOOK facility available in
1840 Apache::Request, with the exception that the first argument to the
1841 callback is an Apache::Upload object, here it's the remote filename.
1842 $q = CGI->new(\&hook [,$data [,$use_tempfile]]);
1844 sub hook {
1846 my ($filename, $buffer, $bytes_read, $data) = @_;
1847 print "Read $bytes_read bytes of $filename\n";
1848 }
1849 The $data field is optional; it lets you pass configuration information
1851 (e.g. a database handle) to your hook callback.
1852 The $use_tempfile field is a flag that lets you turn on and off
1854 CGI.pm's use of a temporary disk-based file during file upload. If you
1855 set this to a FALSE value (default true) then
1856 $q->param('uploaded_file') will no longer work, and the only way to get
1857 at the uploaded data is via the hook you provide.
1858 If using the function-oriented interface, call the CGI::upload_hook()
1860 method before calling param() or any other CGI functions:
1861 CGI::upload_hook(\&hook [,$data [,$use_tempfile]]);
1863 This method is not exported by default. You will have to import it
1865 explicitly if you wish to use it without the CGI:: prefix.
1866 Troubleshooting file uploads on Windows
1868 If you are using CGI.pm on a Windows platform and find that binary
1870 files get slightly larger when uploaded but that text files remain the
1871 same, then you have forgotten to activate binary mode on the output
1872 filehandle. Be sure to call binmode() on any handle that you create to
1873 write the uploaded file to disk.
1874 Older ways to process file uploads
1876 ( This section is here for completeness. if you are building a new
1878 application with CGI.pm, you can skip it. )
1879 The original way to process file uploads with CGI.pm was to use
1881 param(). The value it returns has a dual nature as both a file name and
1882 a lightweight filehandle. This dual nature is problematic if you
1883 following the recommended practice of having "use strict" in your code.
1884 Perl will complain when you try to use a string as a filehandle. More
1885 seriously, it is possible for the remote user to type garbage into the
1886 upload field, in which case what you get from param() is not a
1887 filehandle at all, but a string.
1888 To solve this problem the upload() method was added, which always
1890 returns a lightweight filehandle. This generally works well, but will
1891 have trouble interoperating with some other modules because the file
1892 handle is not derived from IO::Handle. So that brings us to current
1893 recommendation given above, which is to call the handle() method on the
1894 file handle returned by upload(). That upgrades the handle to an
1895 IO::Handle. It's a big win for compatibility for a small penalty of
1896 loading IO::Handle the first time you call it.
1897 CREATING A POPUP MENU
1899 print popup_menu('menu_name',
1900 ['eenie','meenie','minie'],
1901 'meenie');
1902 -or-
1904 %labels = ('eenie'=>'your first choice',
1906 'meenie'=>'your second choice',
1907 'minie'=>'your third choice');
1908 %attributes = ('eenie'=>{'class'=>'class of first choice'});
1909 print popup_menu('menu_name',
1910 ['eenie','meenie','minie'],
1911 'meenie',\%labels,\%attributes);
1912 -or (named parameter style)-
1914 print popup_menu(-name=>'menu_name',
1916 -values=>['eenie','meenie','minie'],
1917 -default=>['meenie','minie'],
1918 -labels=>\%labels,
1919 -attributes=>\%attributes);
1920 popup_menu() creates a menu.
1922 1. The required first argument is the menu's name (-name).
1924 2. The required second argument (-values) is an array reference
1926 containing the list of menu items in the menu. You can pass the
1927 method an anonymous array, as shown in the example, or a reference
1928 to a named array, such as "\@foo".
1929 3. The optional third parameter (-default) is the name of the default
1931 menu choice. If not specified, the first item will be the default.
1932 The values of the previous choice will be maintained across
1933 queries. Pass an array reference to select multiple defaults.
1934 4. The optional fourth parameter (-labels) is provided for people who
1936 want to use different values for the user-visible label inside the
1937 popup menu and the value returned to your script. It's a pointer
1938 to an hash relating menu values to user-visible labels. If you
1939 leave this parameter blank, the menu values will be displayed by
1940 default. (You can also leave a label undefined if you want to).
1941 5. The optional fifth parameter (-attributes) is provided to assign
1943 any of the common HTML attributes to an individual menu item. It's
1944 a pointer to a hash relating menu values to another hash with the
1945 attribute's name as the key and the attribute's value as the value.
1946 When the form is processed, the selected value of the popup menu can be
1948 retrieved using:
1949 $popup_menu_value = param('menu_name');
1951 CREATING AN OPTION GROUP
1953 Named parameter style
1954 print popup_menu(-name=>'menu_name',
1956 -values=>[qw/eenie meenie minie/,
1957 optgroup(-name=>'optgroup_name',
1958 -values => ['moe','catch'],
1959 -attributes=>{'catch'=>{'class'=>'red'}})],
1960 -labels=>{'eenie'=>'one',
1961 'meenie'=>'two',
1962 'minie'=>'three'},
1963 -default=>'meenie');
1964 Old style
1966 print popup_menu('menu_name',
1967 ['eenie','meenie','minie',
1968 optgroup('optgroup_name', ['moe', 'catch'],
1969 {'catch'=>{'class'=>'red'}})],'meenie',
1970 {'eenie'=>'one','meenie'=>'two','minie'=>'three'});
1971 optgroup() creates an option group within a popup menu.
1973 1. The required first argument (-name) is the label attribute of the
1975 optgroup and is not inserted in the parameter list of the query.
1976 2. The required second argument (-values) is an array reference
1978 containing the list of menu items in the menu. You can pass the
1979 method an anonymous array, as shown in the example, or a reference
1980 to a named array, such as \@foo. If you pass a HASH reference, the
1981 keys will be used for the menu values, and the values will be used
1982 for the menu labels (see -labels below).
1983 3. The optional third parameter (-labels) allows you to pass a
1985 reference to a hash containing user-visible labels for one or more
1986 of the menu items. You can use this when you want the user to see
1987 one menu string, but have the browser return your program a
1988 different one. If you don't specify this, the value string will be
1989 used instead ("eenie", "meenie" and "minie" in this example). This
1990 is equivalent to using a hash reference for the -values parameter.
1991 4. An optional fourth parameter (-labeled) can be set to a true value
1993 and indicates that the values should be used as the label attribute
1994 for each option element within the optgroup.
1995 5. An optional fifth parameter (-novals) can be set to a true value
1997 and indicates to suppress the val attribute in each option element
1998 within the optgroup.
1999 See the discussion on optgroup at W3C
2001 (http://www.w3.org/TR/REC-html40/interact/forms.html#edef-OPTGROUP)
2002 for details.
2003 6. An optional sixth parameter (-attributes) is provided to assign any
2005 of the common HTML attributes to an individual menu item. It's a
2006 pointer to a hash relating menu values to another hash with the
2007 attribute's name as the key and the attribute's value as the value.
2008 CREATING A SCROLLING LIST
2010 print scrolling_list('list_name',
2011 ['eenie','meenie','minie','moe'],
2012 ['eenie','moe'],5,'true',{'moe'=>{'class'=>'red'}});
2013 -or-
2014 print scrolling_list('list_name',
2016 ['eenie','meenie','minie','moe'],
2017 ['eenie','moe'],5,'true',
2018 \%labels,%attributes);
2019 -or-
2021 print scrolling_list(-name=>'list_name',
2023 -values=>['eenie','meenie','minie','moe'],
2024 -default=>['eenie','moe'],
2025 -size=>5,
2026 -multiple=>'true',
2027 -labels=>\%labels,
2028 -attributes=>\%attributes);
2029 scrolling_list() creates a scrolling list.
2031 Parameters:
2033 1. The first and second arguments are the list name (-name) and values
2034 (-values). As in the popup menu, the second argument should be an
2035 array reference.
2036 2. The optional third argument (-default) can be either a reference to
2038 a list containing the values to be selected by default, or can be a
2039 single value to select. If this argument is missing or undefined,
2040 then nothing is selected when the list first appears. In the named
2041 parameter version, you can use the synonym "-defaults" for this
2042 parameter.
2043 3. The optional fourth argument is the size of the list (-size).
2045 4. The optional fifth argument can be set to true to allow multiple
2047 simultaneous selections (-multiple). Otherwise only one selection
2048 will be allowed at a time.
2049 5. The optional sixth argument is a pointer to a hash containing long
2051 user-visible labels for the list items (-labels). If not provided,
2052 the values will be displayed.
2053 6. The optional sixth parameter (-attributes) is provided to assign
2055 any of the common HTML attributes to an individual menu item. It's
2056 a pointer to a hash relating menu values to another hash with the
2057 attribute's name as the key and the attribute's value as the value.
2058 When this form is processed, all selected list items will be
2060 returned as a list under the parameter name 'list_name'. The
2061 values of the selected items can be retrieved with:
2062 @selected = param('list_name');
2064 CREATING A GROUP OF RELATED CHECKBOXES
2066 print checkbox_group(-name=>'group_name',
2067 -values=>['eenie','meenie','minie','moe'],
2068 -default=>['eenie','moe'],
2069 -linebreak=>'true',
2070 -disabled => ['moe'],
2071 -labels=>\%labels,
2072 -attributes=>\%attributes);
2073 print checkbox_group('group_name',
2075 ['eenie','meenie','minie','moe'],
2076 ['eenie','moe'],'true',\%labels,
2077 {'moe'=>{'class'=>'red'}});
2078 HTML3-COMPATIBLE BROWSERS ONLY:
2080 print checkbox_group(-name=>'group_name',
2082 -values=>['eenie','meenie','minie','moe'],
2083 -rows=2,-columns=>2);
2084 checkbox_group() creates a list of checkboxes that are related by the
2086 same name.
2087 Parameters:
2089 1. The first and second arguments are the checkbox name and values,
2090 respectively (-name and -values). As in the popup menu, the second
2091 argument should be an array reference. These values are used for
2092 the user-readable labels printed next to the checkboxes as well as
2093 for the values passed to your script in the query string.
2094 2. The optional third argument (-default) can be either a reference to
2096 a list containing the values to be checked by default, or can be a
2097 single value to checked. If this argument is missing or undefined,
2098 then nothing is selected when the list first appears.
2099 3. The optional fourth argument (-linebreak) can be set to true to
2101 place line breaks between the checkboxes so that they appear as a
2102 vertical list. Otherwise, they will be strung together on a
2103 horizontal line.
2104 The optional -labels argument is a pointer to a hash relating the
2106 checkbox values to the user-visible labels that will be printed next to
2107 them. If not provided, the values will be used as the default.
2108 The optional parameters -rows, and -columns cause checkbox_group() to
2110 return an HTML3 compatible table containing the checkbox group
2111 formatted with the specified number of rows and columns. You can
2112 provide just the -columns parameter if you wish; checkbox_group will
2113 calculate the correct number of rows for you.
2114 The option -disabled takes an array of checkbox values and disables
2116 them by greying them out (this may not be supported by all browsers).
2117 The optional -attributes argument is provided to assign any of the
2119 common HTML attributes to an individual menu item. It's a pointer to a
2120 hash relating menu values to another hash with the attribute's name as
2121 the key and the attribute's value as the value.
2122 The optional -tabindex argument can be used to control the order in
2124 which radio buttons receive focus when the user presses the tab button.
2125 If passed a scalar numeric value, the first element in the group will
2126 receive this tab index and subsequent elements will be incremented by
2127 one. If given a reference to an array of radio button values, then the
2128 indexes will be jiggered so that the order specified in the array will
2129 correspond to the tab order. You can also pass a reference to a hash
2130 in which the hash keys are the radio button values and the values are
2131 the tab indexes of each button. Examples:
2132 -tabindex => 100 # this group starts at index 100 and counts up
2134 -tabindex => ['moe','minie','eenie','meenie'] # tab in this order
2135 -tabindex => {meenie=>100,moe=>101,minie=>102,eenie=>200} # tab in this order
2136 The optional -labelattributes argument will contain attributes attached
2138 to the <label> element that surrounds each button.
2139 When the form is processed, all checked boxes will be returned as a
2141 list under the parameter name 'group_name'. The values of the "on"
2142 checkboxes can be retrieved with:
2143 @turned_on = param('group_name');
2145 The value returned by checkbox_group() is actually an array of button
2147 elements. You can capture them and use them within tables, lists, or
2148 in other creative ways:
2149 @h = checkbox_group(-name=>'group_name',-values=>\@values);
2151 &use_in_creative_way(@h);
2152 CREATING A STANDALONE CHECKBOX
2154 print checkbox(-name=>'checkbox_name',
2155 -checked=>1,
2156 -value=>'ON',
2157 -label=>'CLICK ME');
2158 -or-
2160 print checkbox('checkbox_name','checked','ON','CLICK ME');
2162 checkbox() is used to create an isolated checkbox that isn't logically
2164 related to any others.
2165 Parameters:
2167 1. The first parameter is the required name for the checkbox (-name).
2168 It will also be used for the user-readable label printed next to
2169 the checkbox.
2170 2. The optional second parameter (-checked) specifies that the
2172 checkbox is turned on by default. Synonyms are -selected and -on.
2173 3. The optional third parameter (-value) specifies the value of the
2175 checkbox when it is checked. If not provided, the word "on" is
2176 assumed.
2177 4. The optional fourth parameter (-label) is the user-readable label
2179 to be attached to the checkbox. If not provided, the checkbox name
2180 is used.
2181 The value of the checkbox can be retrieved using:
2183 $turned_on = param('checkbox_name');
2185 CREATING A RADIO BUTTON GROUP
2187 print radio_group(-name=>'group_name',
2188 -values=>['eenie','meenie','minie'],
2189 -default=>'meenie',
2190 -linebreak=>'true',
2191 -labels=>\%labels,
2192 -attributes=>\%attributes);
2193 -or-
2195 print radio_group('group_name',['eenie','meenie','minie'],
2197 'meenie','true',\%labels,\%attributes);
2198 HTML3-COMPATIBLE BROWSERS ONLY:
2201 print radio_group(-name=>'group_name',
2203 -values=>['eenie','meenie','minie','moe'],
2204 -rows=2,-columns=>2);
2205 radio_group() creates a set of logically-related radio buttons (turning
2207 one member of the group on turns the others off)
2208 Parameters:
2210 1. The first argument is the name of the group and is required
2211 (-name).
2212 2. The second argument (-values) is the list of values for the radio
2214 buttons. The values and the labels that appear on the page are
2215 identical. Pass an array reference in the second argument, either
2216 using an anonymous array, as shown, or by referencing a named array
2217 as in "\@foo".
2218 3. The optional third parameter (-default) is the name of the default
2220 button to turn on. If not specified, the first item will be the
2221 default. You can provide a nonexistent button name, such as "-" to
2222 start up with no buttons selected.
2223 4. The optional fourth parameter (-linebreak) can be set to 'true' to
2225 put line breaks between the buttons, creating a vertical list.
2226 5. The optional fifth parameter (-labels) is a pointer to an
2228 associative array relating the radio button values to user-visible
2229 labels to be used in the display. If not provided, the values
2230 themselves are displayed.
2231 All modern browsers can take advantage of the optional parameters
2233 -rows, and -columns. These parameters cause radio_group() to return an
2234 HTML3 compatible table containing the radio group formatted with the
2235 specified number of rows and columns. You can provide just the
2236 -columns parameter if you wish; radio_group will calculate the correct
2237 number of rows for you.
2238 To include row and column headings in the returned table, you can use
2240 the -rowheaders and -colheaders parameters. Both of these accept a
2241 pointer to an array of headings to use. The headings are just
2242 decorative. They don't reorganize the interpretation of the radio
2243 buttons -- they're still a single named unit.
2244 The optional -tabindex argument can be used to control the order in
2246 which radio buttons receive focus when the user presses the tab button.
2247 If passed a scalar numeric value, the first element in the group will
2248 receive this tab index and subsequent elements will be incremented by
2249 one. If given a reference to an array of radio button values, then the
2250 indexes will be jiggered so that the order specified in the array will
2251 correspond to the tab order. You can also pass a reference to a hash
2252 in which the hash keys are the radio button values and the values are
2253 the tab indexes of each button. Examples:
2254 -tabindex => 100 # this group starts at index 100 and counts up
2256 -tabindex => ['moe','minie','eenie','meenie'] # tab in this order
2257 -tabindex => {meenie=>100,moe=>101,minie=>102,eenie=>200} # tab in this order
2258 The optional -attributes argument is provided to assign any of the
2260 common HTML attributes to an individual menu item. It's a pointer to a
2261 hash relating menu values to another hash with the attribute's name as
2262 the key and the attribute's value as the value.
2263 The optional -labelattributes argument will contain attributes attached
2265 to the <label> element that surrounds each button.
2266 When the form is processed, the selected radio button can be retrieved
2268 using:
2269 $which_radio_button = param('group_name');
2271 The value returned by radio_group() is actually an array of button
2273 elements. You can capture them and use them within tables, lists, or
2274 in other creative ways:
2275 @h = radio_group(-name=>'group_name',-values=>\@values);
2277 &use_in_creative_way(@h);
2278 CREATING A SUBMIT BUTTON
2280 print submit(-name=>'button_name',
2281 -value=>'value');
2282 -or-
2284 print submit('button_name','value');
2286 submit() will create the query submission button. Every form should
2288 have one of these.
2289 Parameters:
2291 1. The first argument (-name) is optional. You can give the button a
2292 name if you have several submission buttons in your form and you
2293 want to distinguish between them.
2294 2. The second argument (-value) is also optional. This gives the
2296 button a value that will be passed to your script in the query
2297 string. The name will also be used as the user-visible label.
2298 3. You can use -label as an alias for -value. I always get confused
2300 about which of -name and -value changes the user-visible label on
2301 the button.
2302 You can figure out which button was pressed by using different values
2304 for each one:
2305 $which_one = param('button_name');
2307 CREATING A RESET BUTTON
2309 print reset
2310 reset() creates the "reset" button. Note that it restores the form to
2312 its value from the last time the script was called, NOT necessarily to
2313 the defaults.
2314 Note that this conflicts with the Perl reset() built-in. Use
2316 CORE::reset() to get the original reset function.
2317 CREATING A DEFAULT BUTTON
2319 print defaults('button_label')
2320 defaults() creates a button that, when invoked, will cause the form to
2322 be completely reset to its defaults, wiping out all the changes the
2323 user ever made.
2324 CREATING A HIDDEN FIELD
2326 print hidden(-name=>'hidden_name',
2327 -default=>['value1','value2'...]);
2328 -or-
2330 print hidden('hidden_name','value1','value2'...);
2332 hidden() produces a text field that can't be seen by the user. It is
2334 useful for passing state variable information from one invocation of
2335 the script to the next.
2336 Parameters:
2338 1. The first argument is required and specifies the name of this field
2339 (-name).
2340 2. The second argument is also required and specifies its value
2342 (-default). In the named parameter style of calling, you can
2343 provide a single value here or a reference to a whole list
2344 Fetch the value of a hidden field this way:
2346 $hidden_value = param('hidden_name');
2348 Note, that just like all the other form elements, the value of a hidden
2350 field is "sticky". If you want to replace a hidden field with some
2351 other values after the script has been called once you'll have to do it
2352 manually:
2353 param('hidden_name','new','values','here');
2355 CREATING A CLICKABLE IMAGE BUTTON
2357 print image_button(-name=>'button_name',
2358 -src=>'/source/URL',
2359 -align=>'MIDDLE');
2360 -or-
2362 print image_button('button_name','/source/URL','MIDDLE');
2364 image_button() produces a clickable image. When it's clicked on the
2366 position of the click is returned to your script as "button_name.x" and
2367 "button_name.y", where "button_name" is the name you've assigned to it.
2368 Parameters:
2370 1. The first argument (-name) is required and specifies the name of
2371 this field.
2372 2. The second argument (-src) is also required and specifies the URL
2374 3. The third option (-align, optional) is an alignment type, and may be
2376 TOP, BOTTOM or MIDDLE
2377 Fetch the value of the button this way:
2379 $x = param('button_name.x');
2380 $y = param('button_name.y');
2381 CREATING A JAVASCRIPT ACTION BUTTON
2383 print button(-name=>'button_name',
2384 -value=>'user visible label',
2385 -onClick=>"do_something()");
2386 -or-
2388 print button('button_name',"user visible value","do_something()");
2390 button() produces an "<input>" tag with "type="button"". When it's
2392 pressed the fragment of JavaScript code pointed to by the -onClick
2393 parameter will be executed.
2394
2396 Browsers support a so-called "cookie" designed to help maintain state
2397 within a browser session. CGI.pm has several methods that support
2398 cookies.
2399 A cookie is a name=value pair much like the named parameters in a CGI
2401 query string. CGI scripts create one or more cookies and send them to
2402 the browser in the HTTP header. The browser maintains a list of
2403 cookies that belong to a particular Web server, and returns them to the
2404 CGI script during subsequent interactions.
2405 In addition to the required name=value pair, each cookie has several
2407 optional attributes:
2408 1. an expiration time
2410 This is a time/date string (in a special GMT format) that indicates
2411 when a cookie expires. The cookie will be saved and returned to
2412 your script until this expiration date is reached if the user exits
2413 the browser and restarts it. If an expiration date isn't
2414 specified, the cookie will remain active until the user quits the
2415 browser.
2416 2. a domain
2418 This is a partial or complete domain name for which the cookie is
2419 valid. The browser will return the cookie to any host that matches
2420 the partial domain name. For example, if you specify a domain name
2421 of ".capricorn.com", then the browser will return the cookie to Web
2422 servers running on any of the machines "www.capricorn.com",
2423 "www2.capricorn.com", "feckless.capricorn.com", etc. Domain names
2424 must contain at least two periods to prevent attempts to match on
2425 top level domains like ".edu". If no domain is specified, then the
2426 browser will only return the cookie to servers on the host the
2427 cookie originated from.
2428 3. a path
2430 If you provide a cookie path attribute, the browser will check it
2431 against your script's URL before returning the cookie. For
2432 example, if you specify the path "/cgi-bin", then the cookie will
2433 be returned to each of the scripts "/cgi-bin/tally.pl",
2434 "/cgi-bin/order.pl", and "/cgi-bin/customer_service/complain.pl",
2435 but not to the script "/cgi-private/site_admin.pl". By default,
2436 path is set to "/", which causes the cookie to be sent to any CGI
2437 script on your site.
2438 4. a "secure" flag
2440 If the "secure" attribute is set, the cookie will only be sent to
2441 your script if the CGI request is occurring on a secure channel,
2442 such as SSL.
2443 The interface to HTTP cookies is the cookie() method:
2445 $cookie = cookie(-name=>'sessionID',
2447 -value=>'xyzzy',
2448 -expires=>'+1h',
2449 -path=>'/cgi-bin/database',
2450 -domain=>'.capricorn.org',
2451 -secure=>1);
2452 print header(-cookie=>$cookie);
2453 cookie() creates a new cookie. Its parameters include:
2455 -name
2457 The name of the cookie (required). This can be any string at all.
2458 Although browsers limit their cookie names to non-whitespace
2459 alphanumeric characters, CGI.pm removes this restriction by
2460 escaping and unescaping cookies behind the scenes.
2461 -value
2463 The value of the cookie. This can be any scalar value, array
2464 reference, or even hash reference. For example, you can store an
2465 entire hash into a cookie this way:
2466 $cookie=cookie(-name=>'family information',
2468 -value=>\%childrens_ages);
2469 -path
2471 The optional partial path for which this cookie will be valid, as
2472 described above.
2473 -domain
2475 The optional partial domain for which this cookie will be valid, as
2476 described above.
2477 -expires
2479 The optional expiration date for this cookie. The format is as
2480 described in the section on the header() method:
2481 "+1h" one hour from now
2483 -secure
2485 If set to true, this cookie will only be used within a secure SSL
2486 session.
2487 The cookie created by cookie() must be incorporated into the HTTP
2489 header within the string returned by the header() method:
2490 use CGI ':standard';
2492 print header(-cookie=>$my_cookie);
2493 To create multiple cookies, give header() an array reference:
2495 $cookie1 = cookie(-name=>'riddle_name',
2497 -value=>"The Sphynx's Question");
2498 $cookie2 = cookie(-name=>'answers',
2499 -value=>\%answers);
2500 print header(-cookie=>[$cookie1,$cookie2]);
2501 To retrieve a cookie, request it by name by calling cookie() method
2503 without the -value parameter. This example uses the object-oriented
2504 form:
2505 use CGI;
2507 $query = CGI->new;
2508 $riddle = $query->cookie('riddle_name');
2509 %answers = $query->cookie('answers');
2510 Cookies created with a single scalar value, such as the "riddle_name"
2512 cookie, will be returned in that form. Cookies with array and hash
2513 values can also be retrieved.
2514 The cookie and CGI namespaces are separate. If you have a parameter
2516 named 'answers' and a cookie named 'answers', the values retrieved by
2517 param() and cookie() are independent of each other. However, it's
2518 simple to turn a CGI parameter into a cookie, and vice-versa:
2519 # turn a CGI parameter into a cookie
2521 $c=cookie(-name=>'answers',-value=>[param('answers')]);
2522 # vice-versa
2523 param(-name=>'answers',-value=>[cookie('answers')]);
2524 If you call cookie() without any parameters, it will return a list of
2526 the names of all cookies passed to your script:
2527 @cookies = cookie();
2529 See the cookie.cgi example script for some ideas on how to use cookies
2531 effectively.
2532
2534 It's possible for CGI.pm scripts to write into several browser panels
2535 and windows using the HTML 4 frame mechanism. There are three
2536 techniques for defining new frames programmatically:
2537 1. Create a <Frameset> document
2539 After writing out the HTTP header, instead of creating a standard
2540 HTML document using the start_html() call, create a <frameset>
2541 document that defines the frames on the page. Specify your
2542 script(s) (with appropriate parameters) as the SRC for each of the
2543 frames.
2544 There is no specific support for creating <frameset> sections in
2546 CGI.pm, but the HTML is very simple to write.
2547 2. Specify the destination for the document in the HTTP header
2549 You may provide a -target parameter to the header() method:
2550 print header(-target=>'ResultsWindow');
2552 This will tell the browser to load the output of your script into
2554 the frame named "ResultsWindow". If a frame of that name doesn't
2555 already exist, the browser will pop up a new window and load your
2556 script's document into that. There are a number of magic names
2557 that you can use for targets. See the HTML "<frame>" documentation
2558 for details.
2559 3. Specify the destination for the document in the <form> tag
2561 You can specify the frame to load in the FORM tag itself. With
2562 CGI.pm it looks like this:
2563 print start_form(-target=>'ResultsWindow');
2565 When your script is reinvoked by the form, its output will be
2567 loaded into the frame named "ResultsWindow". If one doesn't
2568 already exist a new window will be created.
2569 The script "frameset.cgi" in the examples directory shows one way to
2571 create pages in which the fill-out form and the response live in side-
2572 by-side frames.
2573
2575 The usual way to use JavaScript is to define a set of functions in a
2576 <SCRIPT> block inside the HTML header and then to register event
2577 handlers in the various elements of the page. Events include such
2578 things as the mouse passing over a form element, a button being
2579 clicked, the contents of a text field changing, or a form being
2580 submitted. When an event occurs that involves an element that has
2581 registered an event handler, its associated JavaScript code gets
2582 called.
2583 The elements that can register event handlers include the <BODY> of an
2585 HTML document, hypertext links, all the various elements of a fill-out
2586 form, and the form itself. There are a large number of events, and each
2587 applies only to the elements for which it is relevant. Here is a
2588 partial list:
2589 onLoad
2591 The browser is loading the current document. Valid in:
2592 + The HTML <BODY> section only.
2594 onUnload
2596 The browser is closing the current page or frame. Valid for:
2597 + The HTML <BODY> section only.
2599 onSubmit
2601 The user has pressed the submit button of a form. This event
2602 happens just before the form is submitted, and your function can
2603 return a value of false in order to abort the submission. Valid
2604 for:
2605 + Forms only.
2607 onClick
2609 The mouse has clicked on an item in a fill-out form. Valid for:
2610 + Buttons (including submit, reset, and image buttons)
2612 + Checkboxes
2613 + Radio buttons
2614 onChange
2616 The user has changed the contents of a field. Valid for:
2617 + Text fields
2619 + Text areas
2620 + Password fields
2621 + File fields
2622 + Popup Menus
2623 + Scrolling lists
2624 onFocus
2626 The user has selected a field to work with. Valid for:
2627 + Text fields
2629 + Text areas
2630 + Password fields
2631 + File fields
2632 + Popup Menus
2633 + Scrolling lists
2634 onBlur
2636 The user has deselected a field (gone to work somewhere else).
2637 Valid for:
2638 + Text fields
2640 + Text areas
2641 + Password fields
2642 + File fields
2643 + Popup Menus
2644 + Scrolling lists
2645 onSelect
2647 The user has changed the part of a text field that is selected.
2648 Valid for:
2649 + Text fields
2651 + Text areas
2652 + Password fields
2653 + File fields
2654 onMouseOver
2656 The mouse has moved over an element.
2657 + Text fields
2659 + Text areas
2660 + Password fields
2661 + File fields
2662 + Popup Menus
2663 + Scrolling lists
2664 onMouseOut
2666 The mouse has moved off an element.
2667 + Text fields
2669 + Text areas
2670 + Password fields
2671 + File fields
2672 + Popup Menus
2673 + Scrolling lists
2674 In order to register a JavaScript event handler with an HTML element,
2676 just use the event name as a parameter when you call the corresponding
2677 CGI method. For example, to have your validateAge() JavaScript code
2678 executed every time the textfield named "age" changes, generate the
2679 field like this:
2680 print textfield(-name=>'age',-onChange=>"validateAge(this)");
2682 This example assumes that you've already declared the validateAge()
2684 function by incorporating it into a <SCRIPT> block. The CGI.pm
2685 start_html() method provides a convenient way to create this section.
2686 Similarly, you can create a form that checks itself over for
2688 consistency and alerts the user if some essential value is missing by
2689 creating it this way:
2690 print start_form(-onSubmit=>"validateMe(this)");
2691 See the javascript.cgi script for a demonstration of how this all
2693 works.
2694
2696 CGI.pm has limited support for HTML3's cascading style sheets (css).
2697 To incorporate a stylesheet into your document, pass the start_html()
2698 method a -style parameter. The value of this parameter may be a
2699 scalar, in which case it is treated as the source URL for the
2700 stylesheet, or it may be a hash reference. In the latter case you
2701 should provide the hash with one or more of -src or -code. -src points
2702 to a URL where an externally-defined stylesheet can be found. -code
2703 points to a scalar value to be incorporated into a <style> section.
2704 Style definitions in -code override similarly-named ones in -src, hence
2705 the name "cascading."
2706 You may also specify the type of the stylesheet by adding the optional
2708 -type parameter to the hash pointed to by -style. If not specified,
2709 the style defaults to 'text/css'.
2710 To refer to a style within the body of your document, add the -class
2712 parameter to any HTML element:
2713 print h1({-class=>'Fancy'},'Welcome to the Party');
2715 Or define styles on the fly with the -style parameter:
2717 print h1({-style=>'Color: red;'},'Welcome to Hell');
2719 You may also use the new span() element to apply a style to a section
2721 of text:
2722 print span({-style=>'Color: red;'},
2724 h1('Welcome to Hell'),
2725 "Where did that handbasket get to?"
2726 );
2727 Note that you must import the ":html3" definitions to have the span()
2729 method available. Here's a quick and dirty example of using CSS's.
2730 See the CSS specification at http://www.w3.org/Style/CSS/ for more
2731 information.
2732 use CGI qw/:standard :html3/;
2734 #here's a stylesheet incorporated directly into the page
2736 $newStyle=<<END;
2737 <!--
2738 P.Tip {
2739 margin-right: 50pt;
2740 margin-left: 50pt;
2741 color: red;
2742 }
2743 P.Alert {
2744 font-size: 30pt;
2745 font-family: sans-serif;
2746 color: red;
2747 }
2748 -->
2749 END
2750 print header();
2751 print start_html( -title=>'CGI with Style',
2752 -style=>{-src=>'http://www.capricorn.com/style/st1.css',
2753 -code=>$newStyle}
2754 );
2755 print h1('CGI with Style'),
2756 p({-class=>'Tip'},
2757 "Better read the cascading style sheet spec before playing with this!"),
2758 span({-style=>'color: magenta'},
2759 "Look Mom, no hands!",
2760 p(),
2761 "Whooo wee!"
2762 );
2763 print end_html;
2764 Pass an array reference to -code or -src in order to incorporate
2766 multiple stylesheets into your document.
2767 Should you wish to incorporate a verbatim stylesheet that includes
2769 arbitrary formatting in the header, you may pass a -verbatim tag to the
2770 -style hash, as follows:
2771 print start_html (-style => {-verbatim => '@import
2773 url("/server-common/css/'.$cssFile.'");',
2774 -src => '/server-common/css/core.css'});
2775 This will generate an HTML header that contains this:
2777 <link rel="stylesheet" type="text/css" href="/server-common/css/core.css">
2779 <style type="text/css">
2780 @import url("/server-common/css/main.css");
2781 </style>
2782 Any additional arguments passed in the -style value will be
2784 incorporated into the <link> tag. For example:
2785 start_html(-style=>{-src=>['/styles/print.css','/styles/layout.css'],
2787 -media => 'all'});
2788 This will give:
2790 <link rel="stylesheet" type="text/css" href="/styles/print.css" media="all"/>
2792 <link rel="stylesheet" type="text/css" href="/styles/layout.css" media="all"/>
2793 <p>
2795 To make more complicated <link> tags, use the Link() function and pass
2797 it to start_html() in the -head argument, as in:
2798 @h = (Link({-rel=>'stylesheet',-type=>'text/css',-src=>'/ss/ss.css',-media=>'all'}),
2800 Link({-rel=>'stylesheet',-type=>'text/css',-src=>'/ss/fred.css',-media=>'paper'}));
2801 print start_html({-head=>\@h})
2802 To create primary and "alternate" stylesheet, use the -alternate
2804 option:
2805 start_html(-style=>{-src=>[
2807 {-src=>'/styles/print.css'},
2808 {-src=>'/styles/alt.css',-alternate=>1}
2809 ]
2810 });
2811
2813 If you are running the script from the command line or in the perl
2814 debugger, you can pass the script a list of keywords or parameter=value
2815 pairs on the command line or from standard input (you don't have to
2816 worry about tricking your script into reading from environment
2817 variables). You can pass keywords like this:
2818 your_script.pl keyword1 keyword2 keyword3
2820 or this:
2822 your_script.pl keyword1+keyword2+keyword3
2824 or this:
2826 your_script.pl name1=value1 name2=value2
2828 or this:
2830 your_script.pl name1=value1&name2=value2
2832 To turn off this feature, use the -no_debug pragma.
2834 To test the POST method, you may enable full debugging with the -debug
2836 pragma. This will allow you to feed newline-delimited name=value pairs
2837 to the script on standard input.
2838 When debugging, you can use quotes and backslashes to escape characters
2840 in the familiar shell manner, letting you place spaces and other funny
2841 characters in your parameter=value pairs:
2842 your_script.pl "name1='I am a long value'" "name2=two\ words"
2844 Finally, you can set the path info for the script by prefixing the
2846 first name/value parameter with the path followed by a question mark
2847 (?):
2848 your_script.pl /your/path/here?name1=value1&name2=value2
2850 DUMPING OUT ALL THE NAME/VALUE PAIRS
2852 The Dump() method produces a string consisting of all the query's
2853 name/value pairs formatted nicely as a nested list. This is useful for
2854 debugging purposes:
2855 print Dump
2857 Produces something that looks like:
2859 <ul>
2861 <li>name1
2862 <ul>
2863 <li>value1
2864 <li>value2
2865 </ul>
2866 <li>name2
2867 <ul>
2868 <li>value1
2869 </ul>
2870 </ul>
2871 As a shortcut, you can interpolate the entire CGI object into a string
2873 and it will be replaced with the a nice HTML dump shown above:
2874 $query=CGI->new;
2876 print "<h2>Current Values</h2> $query\n";
2877
2879 Some of the more useful environment variables can be fetched through
2880 this interface. The methods are as follows:
2881 Accept()
2883 Return a list of MIME types that the remote browser accepts. If you
2884 give this method a single argument corresponding to a MIME type, as
2885 in Accept('text/html'), it will return a floating point value
2886 corresponding to the browser's preference for this type from 0.0
2887 (don't want) to 1.0. Glob types (e.g. text/*) in the browser's
2888 accept list are handled correctly.
2889 Note that the capitalization changed between version 2.43 and 2.44
2891 in order to avoid conflict with Perl's accept() function.
2892 raw_cookie()
2894 Returns the HTTP_COOKIE variable. Cookies have a special format,
2895 and this method call just returns the raw form (?cookie dough).
2896 See cookie() for ways of setting and retrieving cooked cookies.
2897 Called with no parameters, raw_cookie() returns the packed cookie
2899 structure. You can separate it into individual cookies by
2900 splitting on the character sequence "; ". Called with the name of
2901 a cookie, retrieves the unescaped form of the cookie. You can use
2902 the regular cookie() method to get the names, or use the
2903 raw_fetch() method from the CGI::Cookie module.
2904 user_agent()
2906 Returns the HTTP_USER_AGENT variable. If you give this method a
2907 single argument, it will attempt to pattern match on it, allowing
2908 you to do something like user_agent(Mozilla);
2909 path_info()
2911 Returns additional path information from the script URL. E.G.
2912 fetching /cgi-bin/your_script/additional/stuff will result in
2913 path_info() returning "/additional/stuff".
2914 NOTE: The Microsoft Internet Information Server is broken with
2916 respect to additional path information. If you use the Perl DLL
2917 library, the IIS server will attempt to execute the additional path
2918 information as a Perl script. If you use the ordinary file
2919 associations mapping, the path information will be present in the
2920 environment, but incorrect. The best thing to do is to avoid using
2921 additional path information in CGI scripts destined for use with
2922 IIS.
2923 path_translated()
2925 As per path_info() but returns the additional path information
2926 translated into a physical path, e.g.
2927 "/usr/local/etc/httpd/htdocs/additional/stuff".
2928 The Microsoft IIS is broken with respect to the translated path as
2930 well.
2931 remote_host()
2933 Returns either the remote host name or IP address. if the former
2934 is unavailable.
2935 remote_addr()
2937 Returns the remote host IP address, or 127.0.0.1 if the address is
2938 unavailable.
2939 script_name() Return the script name as a partial URL, for self-
2941 referring scripts.
2942 referer()
2943 Return the URL of the page the browser was viewing prior to
2944 fetching your script. Not available for all browsers.
2945 auth_type ()
2947 Return the authorization/verification method in use for this
2948 script, if any.
2949 server_name ()
2951 Returns the name of the server, usually the machine's host name.
2952 virtual_host ()
2954 When using virtual hosts, returns the name of the host that the
2955 browser attempted to contact
2956 server_port ()
2958 Return the port that the server is listening on.
2959 virtual_port ()
2961 Like server_port() except that it takes virtual hosts into account.
2962 Use this when running with virtual hosts.
2963 server_software ()
2965 Returns the server software and version number.
2966 remote_user ()
2968 Return the authorization/verification name used for user
2969 verification, if this script is protected.
2970 user_name ()
2972 Attempt to obtain the remote user's name, using a variety of
2973 different techniques. This only works with older browsers such as
2974 Mosaic. Newer browsers do not report the user name for privacy
2975 reasons!
2976 request_method()
2978 Returns the method used to access your script, usually one of
2979 'POST', 'GET' or 'HEAD'.
2980 content_type()
2982 Returns the content_type of data submitted in a POST, generally
2983 multipart/form-data or application/x-www-form-urlencoded
2984 http()
2986 Called with no arguments returns the list of HTTP environment
2987 variables, including such things as HTTP_USER_AGENT,
2988 HTTP_ACCEPT_LANGUAGE, and HTTP_ACCEPT_CHARSET, corresponding to the
2989 like-named HTTP header fields in the request. Called with the name
2990 of an HTTP header field, returns its value. Capitalization and the
2991 use of hyphens versus underscores are not significant.
2992 For example, all three of these examples are equivalent:
2994 $requested_language = http('Accept-language');
2996 $requested_language = http('Accept_language');
2997 $requested_language = http('HTTP_ACCEPT_LANGUAGE');
2998 https()
3000 The same as http(), but operates on the HTTPS environment variables
3001 present when the SSL protocol is in effect. Can be used to
3002 determine whether SSL is turned on.
3003
3005 NPH, or "no-parsed-header", scripts bypass the server completely by
3006 sending the complete HTTP header directly to the browser. This has
3007 slight performance benefits, but is of most use for taking advantage of
3008 HTTP extensions that are not directly supported by your server, such as
3009 server push and PICS headers.
3010 Servers use a variety of conventions for designating CGI scripts as
3012 NPH. Many Unix servers look at the beginning of the script's name for
3013 the prefix "nph-". The Macintosh WebSTAR server and Microsoft's
3014 Internet Information Server, in contrast, try to decide whether a
3015 program is an NPH script by examining the first line of script output.
3016 CGI.pm supports NPH scripts with a special NPH mode. When in this
3018 mode, CGI.pm will output the necessary extra header information when
3019 the header() and redirect() methods are called.
3020 The Microsoft Internet Information Server requires NPH mode. As of
3022 version 2.30, CGI.pm will automatically detect when the script is
3023 running under IIS and put itself into this mode. You do not need to do
3024 this manually, although it won't hurt anything if you do. However,
3025 note that if you have applied Service Pack 6, much of the functionality
3026 of NPH scripts, including the ability to redirect while setting a
3027 cookie, do not work at all on IIS without a special patch from
3028 Microsoft. See
3029 http://web.archive.org/web/20010812012030/http://support.microsoft.com/support/kb/articles/Q280/3/41.ASP
3030 Non-Parsed Headers Stripped From CGI Applications That Have nph- Prefix
3031 in Name.
3032 In the use statement
3034 Simply add the "-nph" pragma to the list of symbols to be imported
3035 into your script:
3036 use CGI qw(:standard -nph)
3038 By calling the nph() method:
3040 Call nph() with a non-zero parameter at any point after using
3041 CGI.pm in your program.
3042 CGI->nph(1)
3044 By using -nph parameters
3046 in the header() and redirect() statements:
3047 print header(-nph=>1);
3049
3051 CGI.pm provides four simple functions for producing multipart documents
3052 of the type needed to implement server push. These functions were
3053 graciously provided by Ed Jordan <ed@fidalgo.net>. To import these
3054 into your namespace, you must import the ":push" set. You are also
3055 advised to put the script into NPH mode and to set $| to 1 to avoid
3056 buffering problems.
3057 Here is a simple script that demonstrates server push:
3059 #!/usr/local/bin/perl
3061 use CGI qw/:push -nph/;
3062 $| = 1;
3063 print multipart_init(-boundary=>'----here we go!');
3064 for (0 .. 4) {
3065 print multipart_start(-type=>'text/plain'),
3066 "The current time is ",scalar(localtime),"\n";
3067 if ($_ < 4) {
3068 print multipart_end;
3069 } else {
3070 print multipart_final;
3071 }
3072 sleep 1;
3073 }
3074 This script initializes server push by calling multipart_init(). It
3076 then enters a loop in which it begins a new multipart section by
3077 calling multipart_start(), prints the current local time, and ends a
3078 multipart section with multipart_end(). It then sleeps a second, and
3079 begins again. On the final iteration, it ends the multipart section
3080 with multipart_final() rather than with multipart_end().
3081 multipart_init()
3083 multipart_init(-boundary=>$boundary);
3084 Initialize the multipart system. The -boundary argument specifies
3086 what MIME boundary string to use to separate parts of the document.
3087 If not provided, CGI.pm chooses a reasonable boundary for you.
3088 multipart_start()
3090 multipart_start(-type=>$type)
3091 Start a new part of the multipart document using the specified MIME
3093 type. If not specified, text/html is assumed.
3094 multipart_end()
3096 multipart_end()
3097 End a part. You must remember to call multipart_end() once for
3099 each multipart_start(), except at the end of the last part of the
3100 multipart document when multipart_final() should be called instead
3101 of multipart_end().
3102 multipart_final()
3104 multipart_final()
3105 End all parts. You should call multipart_final() rather than
3107 multipart_end() at the end of the last part of the multipart
3108 document.
3109 Users interested in server push applications should also have a look at
3111 the CGI::Push module.
3112
3114 A potential problem with CGI.pm is that, by default, it attempts to
3115 process form POSTings no matter how large they are. A wily hacker
3116 could attack your site by sending a CGI script a huge POST of many
3117 megabytes. CGI.pm will attempt to read the entire POST into a
3118 variable, growing hugely in size until it runs out of memory. While
3119 the script attempts to allocate the memory the system may slow down
3120 dramatically. This is a form of denial of service attack.
3121 Another possible attack is for the remote user to force CGI.pm to
3123 accept a huge file upload. CGI.pm will accept the upload and store it
3124 in a temporary directory even if your script doesn't expect to receive
3125 an uploaded file. CGI.pm will delete the file automatically when it
3126 terminates, but in the meantime the remote user may have filled up the
3127 server's disk space, causing problems for other programs.
3128 The best way to avoid denial of service attacks is to limit the amount
3130 of memory, CPU time and disk space that CGI scripts can use. Some Web
3131 servers come with built-in facilities to accomplish this. In other
3132 cases, you can use the shell limit or ulimit commands to put ceilings
3133 on CGI resource usage.
3134 CGI.pm also has some simple built-in protections against denial of
3136 service attacks, but you must activate them before you can use them.
3137 These take the form of two global variables in the CGI name space:
3138 $CGI::POST_MAX
3140 If set to a non-negative integer, this variable puts a ceiling on
3141 the size of POSTings, in bytes. If CGI.pm detects a POST that is
3142 greater than the ceiling, it will immediately exit with an error
3143 message. This value will affect both ordinary POSTs and multipart
3144 POSTs, meaning that it limits the maximum size of file uploads as
3145 well. You should set this to a reasonably high value, such as 1
3146 megabyte.
3147 $CGI::DISABLE_UPLOADS
3149 If set to a non-zero value, this will disable file uploads
3150 completely. Other fill-out form values will work as usual.
3151 You can use these variables in either of two ways.
3153 1. On a script-by-script basis
3155 Set the variable at the top of the script, right after the "use"
3156 statement:
3157 use CGI qw/:standard/;
3159 use CGI::Carp 'fatalsToBrowser';
3160 $CGI::POST_MAX=1024 * 100; # max 100K posts
3161 $CGI::DISABLE_UPLOADS = 1; # no uploads
3162 2. Globally for all scripts
3164 Open up CGI.pm, find the definitions for $POST_MAX and
3165 $DISABLE_UPLOADS, and set them to the desired values. You'll find
3166 them towards the top of the file in a subroutine named
3167 initialize_globals().
3168 An attempt to send a POST larger than $POST_MAX bytes will cause
3170 param() to return an empty CGI parameter list. You can test for this
3171 event by checking cgi_error(), either after you create the CGI object
3172 or, if you are using the function-oriented interface, call <param()>
3173 for the first time. If the POST was intercepted, then cgi_error() will
3174 return the message "413 POST too large".
3175 This error message is actually defined by the HTTP protocol, and is
3177 designed to be returned to the browser as the CGI script's status
3178 code. For example:
3179 $uploaded_file = param('upload');
3181 if (!$uploaded_file && cgi_error()) {
3182 print header(-status=>cgi_error());
3183 exit 0;
3184 }
3185 However it isn't clear that any browser currently knows what to do with
3187 this status code. It might be better just to create an HTML page that
3188 warns the user of the problem.
3189
3191 To make it easier to port existing programs that use cgi-lib.pl the
3192 compatibility routine "ReadParse" is provided. Porting is simple:
3193 OLD VERSION
3195 require "cgi-lib.pl";
3197 &ReadParse;
3198 print "The value of the antique is $in{antique}.\n";
3199 NEW VERSION
3201 use CGI;
3203 CGI::ReadParse();
3204 print "The value of the antique is $in{antique}.\n";
3205 CGI.pm's ReadParse() routine creates a tied variable named %in, which
3207 can be accessed to obtain the query variables. Like ReadParse, you can
3208 also provide your own variable. Infrequently used features of
3209 ReadParse, such as the creation of @in and $in variables, are not
3210 supported.
3211 Once you use ReadParse, you can retrieve the query object itself this
3213 way:
3214 $q = $in{CGI};
3216 print $q->textfield(-name=>'wow',
3217 -value=>'does this really work?');
3218 This allows you to start using the more interesting features of CGI.pm
3220 without rewriting your old scripts from scratch.
3221 An even simpler way to mix cgi-lib calls with CGI.pm calls is to import
3223 both the ":cgi-lib" and ":standard" method:
3224 use CGI qw(:cgi-lib :standard);
3226 &ReadParse;
3227 print "The price of your purchase is $in{price}.\n";
3228 print textfield(-name=>'price', -default=>'$1.99');
3229 Cgi-lib functions that are available in CGI.pm
3231 In compatability mode, the following cgi-lib.pl functions are available
3232 for your use:
3233 ReadParse()
3235 PrintHeader()
3236 HtmlTop()
3237 HtmlBot()
3238 SplitParam()
3239 MethGet()
3240 MethPost()
3241 Cgi-lib functions that are not available in CGI.pm
3243 * Extended form of ReadParse()
3244 The extended form of ReadParse() that provides for file upload
3245 spooling, is not available.
3246 * MyBaseURL()
3248 This function is not available. Use CGI.pm's url() method instead.
3249 * MyFullURL()
3251 This function is not available. Use CGI.pm's self_url() method
3252 instead.
3253 * CgiError(), CgiDie()
3255 These functions are not supported. Look at CGI::Carp for the way I
3256 prefer to handle error messages.
3257 * PrintVariables()
3259 This function is not available. To achieve the same effect,
3260 just print out the CGI object:
3261 use CGI qw(:standard);
3263 $q = CGI->new;
3264 print h1("The Variables Are"),$q;
3265 * PrintEnv()
3267 This function is not available. You'll have to roll your own if you really need it.
3268
3270 The CGI.pm distribution is copyright 1995-2007, Lincoln D. Stein. It
3271 is distributed under GPL and the Artistic License 2.0.
3272 Address bug reports and comments to: lstein@cshl.org. When sending bug
3274 reports, please provide the version of CGI.pm, the version of Perl, the
3275 name and version of your Web server, and the name and version of the
3276 operating system you are using. If the problem is even remotely
3277 browser dependent, please provide information about the affected
3278 browsers as well.
3279
3281 Thanks very much to:
3282 Matt Heffron (heffron@falstaff.css.beckman.com)
3284 James Taylor (james.taylor@srs.gov)
3285 Scott Anguish <sanguish@digifix.com>
3286 Mike Jewell (mlj3u@virginia.edu)
3287 Timothy Shimmin (tes@kbs.citri.edu.au)
3288 Joergen Haegg (jh@axis.se)
3289 Laurent Delfosse (delfosse@delfosse.com)
3290 Richard Resnick (applepi1@aol.com)
3291 Craig Bishop (csb@barwonwater.vic.gov.au)
3292 Tony Curtis (tc@vcpc.univie.ac.at)
3293 Tim Bunce (Tim.Bunce@ig.co.uk)
3294 Tom Christiansen (tchrist@convex.com)
3295 Andreas Koenig (k@franz.ww.TU-Berlin.DE)
3296 Tim MacKenzie (Tim.MacKenzie@fulcrum.com.au)
3297 Kevin B. Hendricks (kbhend@dogwood.tyler.wm.edu)
3298 Stephen Dahmen (joyfire@inxpress.net)
3299 Ed Jordan (ed@fidalgo.net)
3300 David Alan Pisoni (david@cnation.com)
3301 Doug MacEachern (dougm@opengroup.org)
3302 Robin Houston (robin@oneworld.org)
3303 ...and many many more...
3304 for suggestions and bug fixes.
3305
3307 #!/usr/local/bin/perl
3308 use CGI ':standard';
3310 print header;
3312 print start_html("Example CGI.pm Form");
3313 print "<h1> Example CGI.pm Form</h1>\n";
3314 print_prompt();
3315 do_work();
3316 print_tail();
3317 print end_html;
3318 sub print_prompt {
3320 print start_form;
3321 print "<em>What's your name?</em><br>";
3322 print textfield('name');
3323 print checkbox('Not my real name');
3324 print "<p><em>Where can you find English Sparrows?</em><br>";
3326 print checkbox_group(
3327 -name=>'Sparrow locations',
3328 -values=>[England,France,Spain,Asia,Hoboken],
3329 -linebreak=>'yes',
3330 -defaults=>[England,Asia]);
3331 print "<p><em>How far can they fly?</em><br>",
3333 radio_group(
3334 -name=>'how far',
3335 -values=>['10 ft','1 mile','10 miles','real far'],
3336 -default=>'1 mile');
3337 print "<p><em>What's your favorite color?</em> ";
3339 print popup_menu(-name=>'Color',
3340 -values=>['black','brown','red','yellow'],
3341 -default=>'red');
3342 print hidden('Reference','Monty Python and the Holy Grail');
3344 print "<p><em>What have you got there?</em><br>";
3346 print scrolling_list(
3347 -name=>'possessions',
3348 -values=>['A Coconut','A Grail','An Icon',
3349 'A Sword','A Ticket'],
3350 -size=>5,
3351 -multiple=>'true');
3352 print "<p><em>Any parting comments?</em><br>";
3354 print textarea(-name=>'Comments',
3355 -rows=>10,
3356 -columns=>50);
3357 print "<p>",reset;
3359 print submit('Action','Shout');
3360 print submit('Action','Scream');
3361 print end_form;
3362 print "<hr>\n";
3363 }
3364 sub do_work {
3366 print "<h2>Here are the current settings in this form</h2>";
3368 for my $key (param) {
3370 print "<strong>$key</strong> -> ";
3371 my @values = param($key);
3372 print join(", ",@values),"<br>\n";
3373 }
3374 }
3375 sub print_tail {
3377 print <<END;
3378 <hr>
3379 <address>Lincoln D. Stein</address><br>
3380 <a href="/">Home Page</a>
3381 END
3382 }
3383
3385 Please report them.
3386
3388 CGI::Carp - provides a Carp implementation tailored to the CGI
3389 environment.
3390 CGI::Fast - supports running CGI applications under FastCGI
3392 CGI::Pretty - pretty prints HTML generated by CGI.pm (with a
3394 performance penalty)
3395perl v5.10.1 2017-03-22 CGI(3pm)