1HTML::Mason::Request(3)User Contributed Perl DocumentatioHnTML::Mason::Request(3)
2
3
4
6 HTML::Mason::Request - Mason Request Class
7
9 $m->abort (...)
10 $m->comp (...)
11 etc.
12
14 The Request API is your gateway to all Mason features not provided by
15 syntactic tags. Mason creates a new Request object for every web
16 request. Inside a component you access the current request object via
17 the global $m. Outside of a component, you can use the class method
18 "instance".
19
21 The methods Request->comp, Request->comp_exists, and
22 Request->fetch_comp take a component path argument. Component paths
23 are like URL paths, and always use a forward slash (/) as the
24 separator, regardless of what your operating system uses.
25
26 · If the path is absolute (starting with a '/'), then the component
27 is found relative to the component root.
28
29 · If the path is relative (no leading '/'), then the component is
30 found relative to the current component directory.
31
32 · If the path matches both a subcomponent and file-based component,
33 the subcomponent takes precedence.
34
36 autoflush
37 True or false, default is false. Indicates whether to flush the
38 output buffer ("$m->flush_buffer") after every string is output.
39 Turn on autoflush if you need to send partial output to the client,
40 for example in a progress meter.
41
42 As of Mason 1.3, autoflush will only work if enable_autoflush has
43 been set. Components can be compiled more efficiently if they
44 don't have to check for autoflush. Before using autoflush you might
45 consider whether a few manual "$m->flush_buffer" calls would work
46 nearly as well.
47
48 data_cache_api
49 The "$m->cache" API to use:
50
51 · '1.1', the default, indicates a "Cache::Cache" based API.
52
53 · 'chi' indicates a "CHI" based API.
54
55 · '1.0' indicates the custom cache API used in Mason 1.0x and
56 earlier. This compatibility layer is provided as a convenience
57 for users upgrading from older versions of Mason, but will not
58 be supported indefinitely.
59
60 data_cache_defaults
61 A hash reference of default options to use for the "$m->cache"
62 command. For example, to use Cache::Cache's "MemoryCache"
63 implementation by default:
64
65 data_cache_defaults => {cache_class => 'MemoryCache'}
66
67 To use the CHI "FastMmap" driver by default:
68
69 data_cache_api => 'CHI',
70 data_cache_defaults => {driver => 'FastMmap'},
71
72 These settings are overriden by options given to particular
73 "$m->cache" calls.
74
75 dhandler_name
76 File name used for dhandlers. Default is "dhandler". If this is
77 set to an empty string ("") then dhandlers are turned off entirely.
78
79 error_format
80 Indicates how errors are formatted. The built-in choices are
81
82 · brief - just the error message with no trace information
83
84 · text - a multi-line text format
85
86 · line - a single-line text format, with different pieces of
87 information separated by tabs (useful for log files)
88
89 · html - a fancy html format
90
91 The default format under Apache and CGI is either line or html
92 depending on whether the error mode is fatal or output,
93 respectively. The default for standalone mode is text.
94
95 The formats correspond to "HTML::Mason::Exception" methods named
96 as_format. You can define your own format by creating an
97 appropriately named method; for example, to define an "xml" format,
98 create a method "HTML::Mason::Exception::as_xml" patterned after
99 one of the built-in methods.
100
101 error_mode
102 Indicates how errors are returned to the caller. The choices are
103 fatal, meaning die with the error, and output, meaning output the
104 error just like regular output.
105
106 The default under Apache and CGI is output, causing the error to be
107 displayed in the browser. The default for standalone mode is
108 fatal.
109
110 component_error_handler
111 A code reference used to handle errors thrown during component
112 compilation or runtime. By default, this is a subroutine that turns
113 non-exception object errors in components into exceptions. If this
114 parameter is set to a false value, these errors are simply rethrown
115 as-is.
116
117 Turning exceptions into objects can be expensive, since this will
118 cause the generation of a stack trace for each error. If you are
119 using strings or unblessed references as exceptions in your code,
120 you may want to turn this off as a performance boost.
121
122 max_recurse
123 The maximum recursion depth for the component stack, for the
124 request stack, and for the inheritance stack. An error is signalled
125 if the maximum is exceeded. Default is 32.
126
127 out_method
128 Indicates where to send output. If out_method is a reference to a
129 scalar, output is appended to the scalar. If out_method is a
130 reference to a subroutine, the subroutine is called with each
131 output string. For example, to send output to a file called
132 "mason.out":
133
134 my $fh = new IO::File ">mason.out";
135 ...
136 out_method => sub { $fh->print($_[0]) }
137
138 By default, out_method prints to standard output. Under Apache,
139 standard output is redirected to "$r->print".
140
141 plugins
142 An array of plugins that will be called at various stages of
143 request processing. Please see HTML::Mason::Plugin for details.
144
146 All of the above properties have standard accessor methods of the same
147 name. In general, no arguments retrieves the value, and one argument
148 sets and returns the value. For example:
149
150 my $max_recurse_level = $m->max_recurse;
151 $m->autoflush(1);
152
154 abort ([return value])
155 Ends the current request, finishing the page without returning
156 through components. The optional argument specifies the return
157 value from "Interp::exec"; in a web environment, this ultimately
158 becomes the HTTP status code.
159
160 "abort" is implemented by throwing an HTML::Mason::Exception::Abort
161 object and can thus be caught by eval(). The "aborted" method is a
162 shortcut for determining whether a caught error was generated by
163 "abort".
164
165 If "abort" is called from a component that has a "<%filter>", than
166 any output generated up to that point is filtered, unless "abort"
167 is called from a "<%shared>" block.
168
169 clear_and_abort ([return value])
170 This method is syntactic sugar for calling "clear_buffer()" and
171 then "abort()". If you are aborting the request because of an
172 error, you will often want to clear the buffer first so that any
173 output generated up to that point is not sent to the client.
174
175 aborted ([$err])
176 Returns true or undef indicating whether the specified $err was
177 generated by "abort". If no $err was passed, uses $@.
178
179 In this code, we catch and process fatal errors while letting
180 "abort" exceptions pass through:
181
182 eval { code_that_may_fail_or_abort() };
183 if ($@) {
184 die $@ if $m->aborted;
185
186 # handle fatal errors...
187
188 $@ can lose its value quickly, so if you are planning to call
189 $m->aborted more than a few lines after the eval, you should save
190 $@ to a temporary variable.
191
192 base_comp
193 Returns the current base component.
194
195 Here are the rules that determine base_comp as you move from
196 component to component.
197
198 · At the beginning of a request, the base component is
199 initialized to the requested component ("$m->request_comp()").
200
201 · When you call a regular component via a path, the base
202 component changes to the called component.
203
204 · When you call a component method via a path (/foo/bar:baz), the
205 base component changes to the method's owner.
206
207 · The base component does not change when:
208
209 · a component call is made to a component object
210
211 · a component call is made to SELF:x or PARENT:x or REQUEST:x
212
213 · a component call is made to a subcomponent (<%def>)
214
215 This may return nothing if the base component is not yet known, for
216 example inside a plugin's "start_request_hook()" method, where we
217 have created a request but it does not yet know anything about the
218 component being called.
219
220 cache
221 "$m->cache" returns a new cache object with a namespace specific to
222 this component. The parameters to and return value from "$m->cache"
223 differ depending on which data_cache_api you are using.
224
225 If data_cache_api = 1.1 (default)
226 cache_class specifies the class of cache object to create. It
227 defaults to "FileCache" in most cases, or "MemoryCache" if the
228 interpreter has no data directory, and must be a backend
229 subclass of "Cache::Cache". The prefix "Cache::" need not be
230 included. See the "Cache::Cache" package for a full list of
231 backend subclasses.
232
233 Beyond that, cache_options may include any valid options to the
234 new() method of the cache class. e.g. for "FileCache", valid
235 options include "default_expires_in" and "cache_depth".
236
237 See HTML::Mason::Cache::BaseCache for information about the
238 object returend from "$m->cache".
239
240 If data_cache_api = CHI
241 chi_root_class specifies the factory class that will be called
242 to create cache objects. The default is 'CHI'.
243
244 driver specifies the driver to use, for example "Memory" or
245 "FastMmap". The default is "File" in most cases, or "Memory"
246 if the interpreter has no data directory.
247
248 Beyond that, cache_options may include any valid options to the
249 new() method of the driver. e.g. for the "File" driver, valid
250 options include "expires_in" and "depth".
251
252 cache_self ([expires_in => '...'], [key => '...'], [get_options],
253 [cache_options])
254 "$m->cache_self" caches the entire output and return result of a
255 component.
256
257 "cache_self" either returns undef, or a list containing the return
258 value of the component followed by '1'. You should return
259 immediately upon getting the latter result, as this indicates that
260 you are inside the second invocation of the component.
261
262 "cache_self" takes any of parameters to "$m->cache" (e.g.
263 cache_depth), any of the optional parameters to "$cache->get"
264 (expire_if, busy_lock), and two additional options:
265
266 · expire_in or expires_in: Indicates when the cache expires - it
267 is passed as the third argument to "$cache->set". e.g. '10
268 sec', '5 min', '2 hours'.
269
270 · key: An identifier used to uniquely identify the cache results
271 - it is passed as the first argument to "$cache->get" and
272 "$cache->set". The default key is '__mason_cache_self__'.
273
274 To cache the component's output:
275
276 <%init>
277 return if $m->cache_self(expire_in => '10 sec'[, key => 'fookey']);
278 ... <rest of init> ...
279 </%init>
280
281 To cache the component's scalar return value:
282
283 <%init>
284 my ($result, $cached) = $m->cache_self(expire_in => '5 min'[, key => 'fookey']);
285
286 return $result if $cached;
287 ... <rest of init> ...
288 </%init>
289
290 To cache the component's list return value:
291
292 <%init>
293 my (@retval) = $m->cache_self(expire_in => '3 hours'[, key => 'fookey']);
294
295 return @retval if pop @retval;
296 ... <rest of init> ...
297 </%init>
298
299 We call "pop" on @retval to remove the mandatory '1' at the end of
300 the list.
301
302 If a component has a "<%filter>" block, then the filtered output is
303 cached.
304
305 Note: users upgrading from 1.0x and earlier can continue to use the
306 old "$m->cache_self" API by setting data_cache_api to '1.0'. This
307 support will be removed at a later date.
308
309 See the the DATA CACHING section of the developer's manual section
310 for more details on how to exercise finer control over caching.
311
312 caller_args
313 Returns the arguments passed by the component at the specified
314 stack level. Use a positive argument to count from the current
315 component and a negative argument to count from the component at
316 the bottom of the stack. e.g.
317
318 $m->caller_args(0) # arguments passed to current component
319 $m->caller_args(1) # arguments passed to component that called us
320 $m->caller_args(-1) # arguments passed to first component executed
321
322 When called in scalar context, a hash reference is returned. When
323 called in list context, a list of arguments (which may be assigned
324 to a hash) is returned. Returns undef or an empty list, depending
325 on context, if the specified stack level does not exist.
326
327 callers
328 With no arguments, returns the current component stack as a list of
329 component objects, starting with the current component and ending
330 with the top-level component. With one numeric argument, returns
331 the component object at that index in the list. Use a positive
332 argument to count from the current component and a negative
333 argument to count from the component at the bottom of the stack.
334 e.g.
335
336 my @comps = $m->callers # all components
337 $m->callers(0) # current component
338 $m->callers(1) # component that called us
339 $m->callers(-1) # first component executed
340
341 Returns undef or an empty list, depending on context, if the
342 specified stack level does not exist.
343
344 caller
345 A synonym for "$m->callers(1)", i.e. the component that called the
346 currently executing component.
347
348 call_next ([args...])
349 Calls the next component in the content wrapping chain; usually
350 called from an autohandler. With no arguments, the original
351 arguments are passed to the component. Any arguments specified
352 here serve to augment and override (in case of conflict) the
353 original arguments. Works like "$m->comp" in terms of return value
354 and scalar/list context. See the autohandlers section of the
355 developer's manual for examples.
356
357 call_self (output, return, error, tag)
358 This method allows a component to call itself so that it can filter
359 both its output and return values. It is fairly advanced; for most
360 purposes the "<%filter>" tag will be sufficient and simpler.
361
362 "$m->call_self" takes four arguments, all of them optional.
363
364 output - scalar reference that will be populated with the component
365 output.
366 return - scalar reference that will be populated with the component
367 return value.
368 error - scalar reference that will be populated with the error
369 thrown by the component, if any. If this parameter is not defined,
370 then call_self will not catch errors.
371 tag - a name for this call_self invocation; can almost always be
372 omitted.
373
374 "$m->call_self" acts like a "fork()" in the sense that it will
375 return twice with different values. When it returns 0, you allow
376 control to pass through to the rest of your component. When it
377 returns 1, that means the component has finished and you can
378 examine the output, return value and error. (Don't worry, it
379 doesn't really do a fork! See next section for explanation.)
380
381 The following examples would generally appear at the top of a
382 "<%init>" section. Here is a no-op "$m->call_self" that leaves the
383 output and return value untouched:
384
385 <%init>
386 my ($output, $retval);
387 if ($m->call_self(\$output, \$retval)) {
388 $m->print($output);
389 return $retval;
390 }
391 ...
392
393 Here is a simple output filter that makes the output all uppercase.
394 Note that we ignore both the original and the final return value.
395
396 <%init>
397 my ($output, $error);
398 if ($m->call_self(\$output, undef)) {
399 $m->print(uc $output);
400 return;
401 }
402 ...
403
404 Here is a piece of code that traps all errors occuring anywhere in
405 a component or its children, e.g. for the purpose of handling
406 application-specific exceptions. This is difficult to do with a
407 manual "eval" because it would have to span multiple code sections
408 and the main component body.
409
410 <%init>
411 my ($output, undef, $error);
412 if ($m->call_self(\$output, undef, \$error)) {
413 if ($error) {
414 # check $error and do something with it
415 }
416 $m->print($output);
417 return;
418 }
419 ...
420
421 clear_buffer
422 Clears the Mason output buffer. Any output sent before this line is
423 discarded. Useful for handling error conditions that can only be
424 detected in the middle of a request.
425
426 clear_buffer is, of course, thwarted by "flush_buffer".
427
428 comp (comp, args...)
429 Calls the component designated by comp with the specified
430 option/value pairs. comp may be a component path or a component
431 object.
432
433 Components work exactly like Perl subroutines in terms of return
434 values and context. A component can return any type of value, which
435 is then returned from the "$m->comp" call.
436
437 The <& &> tag provides a convenient shortcut for "$m->comp".
438
439 As of 1.10, component calls can accept an initial hash reference of
440 modifiers. The only currently supported modifier is "store", which
441 stores the component's output in a scalar reference. For example:
442
443 my $buf;
444 my $return = $m->comp( { store => \$buf }, '/some/comp', type => 'big' );
445
446 This mostly duplicates the behavior of scomp, but can be useful in
447 rare cases where you need to capture both a component's output and
448 return value.
449
450 This modifier can be used with the <& &> tag as well, for example:
451
452 <& { store => \$buf }, '/some/comp', size => 'medium' &>
453
454 comp_exists (comp_path)
455 Returns 1 if comp_path is the path of an existing component, 0
456 otherwise. comp_path may be any path accepted by comp or
457 fetch_comp, including method or subcomponent paths.
458
459 Depending on implementation, <comp_exists> may try to load the
460 component referred to by the path, and may throw an error if the
461 component contains a syntax error.
462
463 content
464 Evaluates the content (passed between <&| comp &> and </&> tags) of
465 the current component, and returns the resulting text.
466
467 Returns undef if there is no content.
468
469 has_content
470 Returns true if the component was called with content (i.e. with
471 <&| comp &> and </&> tags instead of a single <& comp &> tag). This
472 is generally better than checking the defined'ness of "$m->content"
473 because it will not try to evaluate the content.
474
475 count
476 Returns the number of this request, which is unique for a given
477 request and interpreter.
478
479 current_args
480 Returns the arguments passed to the current component. When called
481 in scalar context, a hash reference is returned. When called in
482 list context, a list of arguments (which may be assigned to a hash)
483 is returned.
484
485 current_comp
486 Returns the current component object.
487
488 decline
489 Used from a top-level component or dhandler, this method clears the
490 output buffer, aborts the current request and restarts with the
491 next applicable dhandler up the tree. If no dhandler is available,
492 a not-found error occurs.
493
494 This method bears no relation to the Apache DECLINED status except
495 in name.
496
497 declined ([$err])
498 Returns true or undef indicating whether the specified $err was
499 generated by "decline". If no $err was passed, uses $@.
500
501 depth
502 Returns the current size of the component stack. The lowest
503 possible value is 1, which indicates we are in the top-level
504 component.
505
506 dhandler_arg
507 If the request has been handled by a dhandler, this method returns
508 the remainder of the URI or "Interp::exec" path when the dhandler
509 directory is removed. Otherwise returns undef.
510
511 "dhandler_arg" may be called from any component in the request, not
512 just the dhandler.
513
514 exec (comp, args...)
515 Starts the request by executing the top-level component and
516 arguments. This is normally called for you on the main request, but
517 you can use it to execute subrequests.
518
519 A request can only be executed once; e.g. it is an error to call
520 this recursively on the same request.
521
522 fetch_comp (comp_path)
523 Given a comp_path, returns the corresponding component object or
524 undef if no such component exists.
525
526 fetch_next
527 Returns the next component in the content wrapping chain, or undef
528 if there is no next component. Usually called from an autohandler.
529 See the autohandlers section of the developer's manual for usage
530 and examples.
531
532 fetch_next_all
533 Returns a list of the remaining components in the content wrapping
534 chain. Usually called from an autohandler. See the autohandlers
535 section of the developer's manual for usage and examples.
536
537 file (filename)
538 Returns the contents of filename as a string. If filename is a
539 relative path, Mason prepends the current component directory.
540
541 flush_buffer
542 Flushes the Mason output buffer. Under mod_perl, also sends HTTP
543 headers if they haven't been sent and calls "$r->rflush" to flush
544 the Apache buffer. Flushing the initial bytes of output can make
545 your servers appear more responsive.
546
547 Attempts to flush the buffers are ignored within the context of a
548 call to "$m->scomp" or when output is being stored in a scalar
549 reference, as with the " { store => \$out } " component call
550 modifier.
551
552 "<%filter>" blocks will process the output whenever the buffers are
553 flushed. If "autoflush" is on, your data may be filtered in small
554 pieces.
555
556 instance
557 This class method returns the "HTML::Mason::Request" currently in
558 use. If called when no Mason request is active it will return
559 "undef".
560
561 If called inside a subrequest, it returns the subrequest object.
562
563 interp
564 Returns the Interp object associated with this request.
565
566 make_subrequest (comp => path, args => arrayref, other parameters)
567 This method creates a new Request object which inherits its
568 parent's settable properties, such as autoflush and out_method.
569 These values may be overridden by passing parameters to this
570 method.
571
572 The "comp" parameter is required, while all other parameters are
573 optional. It may be specified as an absolute path or as a path
574 relative to the current component.
575
576 See the subrequests section of the developer's manual for more
577 information about subrequests.
578
579 notes (key, value)
580 The "notes()" method provides a place to store application data,
581 giving developers a way to share data among multiple components.
582 Any data stored here persists for the duration of the request, i.e.
583 the same lifetime as the Request object.
584
585 Conceptually, "notes()" contains a hash of key-value pairs.
586 "notes($key, $value)" stores a new entry in this hash.
587 "notes($key)" returns a previously stored value. "notes()" without
588 any arguments returns a reference to the entire hash of key-value
589 pairs.
590
591 "notes()" is similar to the mod_perl method "$r->pnotes()". The
592 main differences are that this "notes()" can be used in a
593 non-mod_perl environment, and that its lifetime is tied to the
594 Mason request object, not the Apache request object. In
595 particular, a Mason subrequest has its own "notes()" structure, but
596 would access the same "$r->pnotes()" structure.
597
598 out (string)
599 A synonym for "$m->print".
600
601 print (string)
602 Print the given string. Rarely needed, since normally all text is
603 just placed in the component body and output implicitly.
604 "$m->print" is useful if you need to output something in the middle
605 of a Perl block.
606
607 In 1.1 and on, "print" and "$r->print" are remapped to "$m->print",
608 so they may be used interchangeably. Before 1.1, one should only
609 use "$m->print".
610
611 request_args
612 Returns the arguments originally passed to the top level component
613 (see request_comp for definition). When called in scalar context,
614 a hash reference is returned. When called in list context, a list
615 of arguments (which may be assigned to a hash) is returned.
616
617 request_comp
618 Returns the component originally called in the request. Without
619 autohandlers, this is the same as the first component executed.
620 With autohandlers, this is the component at the end of the
621 "$m->call_next" chain.
622
623 request_depth
624 Returns the current size of the request/subrequest stack. The
625 lowest possible value is 1, which indicates we are in the top-level
626 request. A value of 2 indicates we are inside a subrequest of the
627 top-level request, and so on.
628
629 scomp (comp, args...)
630 Like comp, but returns the component output as a string instead of
631 printing it. (Think sprintf versus printf.) The component's return
632 value is discarded.
633
634 subexec (comp, args...)
635 This method creates a new subrequest with the specified top-level
636 component and arguments, and executes it. This is most often used
637 to perform an "internal redirect" to a new component such that
638 autohandlers and dhandlers take effect.
639
640 time
641 Returns the interpreter's notion of the current time (deprecated).
642
644 These additional methods are available when running Mason with mod_perl
645 and the ApacheHandler.
646
647 ah Returns the ApacheHandler object associated with this request.
648
649 apache_req
650 Returns the Apache request object. This is also available in the
651 global $r.
652
653 auto_send_headers
654 True or false, default is true. Indicates whether Mason should
655 automatically send HTTP headers before sending content back to the
656 client. If you set to false, you should call "$r->send_http_header"
657 manually.
658
659 See the sending HTTP headers section of the developer's manual for
660 more details about the automatic header feature.
661
662 NOTE: This parameter has no effect under mod_perl-2, since calling
663 "$r->send_http_header" is no longer needed.
664
666 This additional method is available when running Mason with the
667 CGIHandler module.
668
669 cgi_request
670 Returns the Apache request emulation object, which is available as
671 $r inside components.
672
673 See the CGIHandler docs for more details.
674
676 This method is available when Mason is running under either the
677 ApacheHandler or CGIHandler modules.
678
679 cgi_object
680 Returns the CGI object used to parse any CGI parameters submitted
681 to the component, assuming that you have not changed the default
682 value of the ApacheHandler args_method parameter. If you are using
683 the 'mod_perl' args method, then calling this method is a fatal
684 error. See the ApacheHandler and CGIHandler documentation for more
685 details.
686
687 redirect ($url, [$status])
688 Given a url, this generates a proper HTTP redirect for that URL. It
689 uses "$m->clear_and_abort" to clear out any previous output, and
690 abort the request. By default, the status code used is 302, but
691 this can be overridden by the user.
692
693 Since this is implemented using "$m->abort", it will be trapped by
694 an " eval {} " block. If you are using an " eval {} " block in
695 your code to trap errors, you need to make sure to rethrow these
696 exceptions, like this:
697
698 eval {
699 ...
700 };
701
702 die $@ if $m->aborted;
703
704 # handle other exceptions
705
707 Jonathan Swartz <swartz@pobox.com>, Dave Rolsky <autarch@urth.org>, Ken
708 Williams <ken@mathforum.org>
709
711 HTML::Mason, HTML::Mason::Devel, HTML::Mason::Component
712
713
714
715perl v5.12.0 2010-05-03 HTML::Mason::Request(3)