File: Request.pm

package info (click to toggle)
libnet-async-fastcgi-perl 0.26-1
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid, trixie
  • size: 284 kB
  • sloc: perl: 2,308; makefile: 2
file content (907 lines) | stat: -rw-r--r-- 20,522 bytes parent folder | download
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
#  You may distribute under the terms of either the GNU General Public License
#  or the Artistic License (the same terms as Perl itself)
#
#  (C) Paul Evans, 2005-2024 -- leonerd@leonerd.org.uk

package Net::Async::FastCGI::Request 0.26;

use v5.14;
use warnings;

use Carp;

use Net::FastCGI::Constant qw( :type :flag :protocol_status );
use Net::FastCGI::Protocol qw( 
   parse_params
   build_end_request_body
);

# The largest amount of data we can fit in a FastCGI record - MUST NOT
# be greater than 2^16-1
use constant MAXRECORDDATA => 65535;

use Encode qw( find_encoding );
use POSIX qw( EAGAIN );

my $CRLF = "\x0d\x0a";

=head1 NAME

C<Net::Async::FastCGI::Request> - a single active FastCGI request

=head1 SYNOPSIS

   use Net::Async::FastCGI;
   use IO::Async::Loop;

   my $fcgi = Net::Async::FastCGI->new(
      on_request => sub {
         my ( $fcgi, $req ) = @_;

         my $path = $req->param( "PATH_INFO" );
         $req->print_stdout( "Status: 200 OK\r\n" .
                             "Content-type: text/plain\r\n" .
                             "\r\n" .
                             "You requested $path" );
         $req->finish();
      }
   );

   my $loop = IO::Async::Loop->new();

   $loop->add( $fcgi );

   $loop->run;

=head1 DESCRIPTION

Instances of this object class represent individual requests received from the
webserver that are currently in-progress, and have not yet been completed.
When given to the controlling program, each request will already have its
parameters (and, on servers without stdin streaming enabled, its STDIN data).
The program can then write response data to the STDOUT stream, messages to the
STDERR stream, and eventually finish it.

This module would not be used directly by a program using
C<Net::Async::FastCGI>, but rather, objects in this class are passed into the
C<on_request> event of the containing C<Net::Async::FastCGI> object.

=cut

sub new
{
   my $class = shift;
   my %args = @_;

   my $rec = $args{rec};

   my $self = bless {
      conn       => $args{conn},
      fcgi       => $args{fcgi},

      reqid      => $rec->{reqid},
      keepconn   => $rec->{flags} & FCGI_KEEP_CONN,

      stdin        => "",
      stdindone    => 0,
      stream_stdin => $args{stream_stdin},

      params     => {},
      paramsdone => 0,

      stdout     => "",
      stderr     => "",

      used_stderr => 0,
   }, $class;

   $self->set_encoding( $args{fcgi}->_default_encoding );

   return $self;
}

sub write_record
{
   my $self = shift;
   my ( $rec ) = @_;

   return if $self->is_aborted;

   my $content = $rec->{content};
   my $contentlen = length( $content );
   if( $contentlen > MAXRECORDDATA ) {
      warn __PACKAGE__."->write_record() called with content longer than ".MAXRECORDDATA." bytes - truncating";
      $content = substr( $content, 0, MAXRECORDDATA );
   }

   $rec->{reqid} = $self->{reqid} unless defined $rec->{reqid};

   my $conn = $self->{conn};

   $conn->write_record( $rec, $content );

}

sub incomingrecord
{
   my $self = shift;
   my ( $rec ) = @_;

   my $type    = $rec->{type};

   if( $type == FCGI_PARAMS ) {
      $self->incomingrecord_params( $rec );
   }
   elsif( $type == FCGI_STDIN ) {
      $self->incomingrecord_stdin( $rec );
   }
   else {
      warn "$self just received unknown record type";
   }
}

sub _ready_check
{
   my $self = shift;

   if( $self->{paramsdone} and ( $self->{stdindone} || $self->{stream_stdin} ) ) {
      $self->{fcgi}->_request_ready( $self );
   }
}

sub incomingrecord_params
{
   my $self = shift;
   my ( $rec ) = @_;

   my $content = $rec->{content};
   my $len     = $rec->{len};

   if( $len ) {
      no warnings 'uninitialized';
      $self->{paramscontent} .= $content;
      return;
   }
   else {
      $self->{params} = parse_params( delete $self->{paramscontent} );
      $self->{paramsdone} = 1;
   }

   $self->_ready_check;
}

sub incomingrecord_stdin
{
   my $self = shift;
   my ( $rec ) = @_;

   my $content = $rec->{content};
   my $len     = $rec->{len};

   if( $len ) {
      $self->{stdin} .= $content;
   }
   else {
      $self->{stdindone} = 1;
   }

   if( $self->{stream_stdin} ) {
      $self->_flush_stdin;
   }
   else {
      $self->_ready_check;
   }
}

sub _start
{
   my $self = shift;
   $self->{started} = 1;

   $self->_flush_stdin if $self->{stream_stdin} and length $self->{stdin};
}

sub _flush_stdin
{
   my $self = shift;

   my $on_stdin = $self->{on_stdin_read};
   if( !$on_stdin ) {
      warn "NaFastCGI::Request incoming STDIN data with on on_stdin_read\n" if $self->{started};
      return;
   }

   {
      my $ret = $self->$on_stdin( \$self->{stdin}, $self->{stdindone} );
      redo if $ret and length $self->{stdin};
   }
}

=head1 METHODS

=cut

=head2 params

   $hashref = $req->params;

This method returns a reference to a hash containing a copy of the request
parameters that had been sent by the webserver as part of the request.

=cut

sub params
{
   my $self = shift;

   my %p = %{$self->{params}};

   return \%p;
}

=head2 param

   $p = $req->param( $key );

This method returns the value of a single request parameter, or C<undef> if no
such key exists.

=cut

sub param
{
   my $self = shift;
   my ( $key ) = @_;

   return $self->{params}{$key};
}

=head2 method

   $method = $req->method;

Returns the value of the C<REQUEST_METHOD> parameter, or C<GET> if there is no
value set for it.

=cut

sub method
{
   my $self = shift;
   return $self->param( "REQUEST_METHOD" ) || "GET";
}

=head2 script_name

   $script_name = $req->script_name;

Returns the value of the C<SCRIPT_NAME> parameter.

=cut

sub script_name
{
   my $self = shift;
   return $self->param( "SCRIPT_NAME" );
}

=head2 path_info

   $path_info = $req->path_info;

Returns the value of the C<PATH_INFO> parameter.

=cut

sub path_info
{
   my $self = shift;
   return $self->param( "PATH_INFO" );
}

=head2 path

   $path = $req->path;

Returns the full request path by reconstructing it from C<script_name> and
C<path_info>.

=cut

sub path
{
   my $self = shift;

   my $path = join "", grep defined && length,
      $self->script_name,
      $self->path_info;
   $path = "/" if !length $path;

   return $path;
}

=head2 query_string

   $query_string = $req->query_string;

Returns the value of the C<QUERY_STRING> parameter.

=cut

sub query_string
{
   my $self = shift;
   return $self->param( "QUERY_STRING" ) || "";
}

=head2 protocol

   $protocol = $req->protocol;

Returns the value of the C<SERVER_PROTOCOL> parameter.

=cut

sub protocol
{
   my $self = shift;
   return $self->param( "SERVER_PROTOCOL" );
}

=head2 set_encoding

   $req->set_encoding( $encoding );

Sets the character encoding used by the request's STDIN, STDOUT and STDERR
streams. This method may be called at any time to change the encoding in
effect, which will be used the next time C<read_stdin_line>, C<read_stdin>,
C<print_stdout> or C<print_stderr> are called. This encoding will remain in
effect until changed again. The encoding of a new request is determined by the
C<default_encoding> parameter of the containing C<Net::Async::FastCGI> object.
If the value C<undef> is passed, the encoding will be removed, and the above
methods will work directly on bytes instead of encoded strings.

=cut

sub set_encoding
{
   my $self = shift;
   my ( $encoding ) = @_;

   if( defined $encoding ) {
      my $codec = find_encoding( $encoding );
      defined $codec or croak "Unrecognised encoding '$encoding'";
      $self->{codec} = $codec;
   }
   else {
      undef $self->{codec};
   }
}

=head2 read_stdin_line

   $line = $req->read_stdin_line;

This method works similarly to the C<< <HANDLE> >> operator. If at least one
line of data is available then it is returned, including the linefeed, and
removed from the buffer. If not, then any remaining partial line is returned
and removed from the buffer. If no data is available any more, then C<undef>
is returned instead.

=cut

sub read_stdin_line
{
   my $self = shift;
   croak "Cannot call ->read_stdin_line on streaming-stdin requests" if $self->{stream_stdin};

   my $codec = $self->{codec};

   if( $self->{stdin} =~ s/^(.*[\r\n])// ) {
      return $codec ? $codec->decode( $1 ) : $1;
   }
   elsif( $self->{stdin} =~ s/^(.+)// ) {
      return $codec ? $codec->decode( $1 ) : $1;
   }
   else {
      return undef;
   }
}

=head2 read_stdin

   $data = $req->read_stdin( $size );

This method works similarly to the C<read(HANDLE)> function. It returns the
next block of up to $size bytes from the STDIN buffer. If no data is available
any more, then C<undef> is returned instead. If $size is not defined, then it
will return all the available data.

=cut

sub read_stdin
{
   my $self = shift;
   croak "Cannot call ->read_stdin on streaming-stdin requests" if $self->{stream_stdin};
   my ( $size ) = @_;

   return undef unless length $self->{stdin};

   $size = length $self->{stdin} unless defined $size;

   my $codec = $self->{codec};

   # If $size is too big, substr() will cope
   my $bytes = substr( $self->{stdin}, 0, $size, "" );
   return $codec ? $codec->decode( $bytes ) : $bytes;
}

=head2 set_on_stdin_read

   $req->set_on_stdin_read( $on_stdin_read );

      $again = $on_stdin_read->( $req, $buffref, $eof );

I<Since version 0.26.>

Only valid on requests on servers with stdin streaming enabled.

This method should be called as part of the C<on_request> event on the server,
to set the callback function to invoke when new data is provided to the stdin
stream for this request.

The callback function is invoked in a similar style to the C<on_read> event
handler of an L<IO::Async::Stream>. It is passed the request itself, along
with a SCALAR reference to the buffer containing the stdin data, and a boolean
indicating if the end of stdin data has been reached.

It should inspect this buffer and remove some prefix of it that it wishes to
consume. Any remaining content will be present on the next call. If it returns
a true value, the callback will be invoked again immediately, to consume more
data. This continues until there is no more data left, or it returns false.

=cut

sub set_on_stdin_read
{
   my $self = shift;
   croak "Cannot call ->set_on_stdin_read except on streaming-stdin requests" unless $self->{stream_stdin};
   ( $self->{on_stdin_read} ) = @_;
}

sub _print_stream
{
   my $self = shift;
   my ( $data, $stream ) = @_;

   while( length $data ) {
      # Send chunks of up to MAXRECORDDATA bytes at once
      my $chunk = substr( $data, 0, MAXRECORDDATA, "" );
      $self->write_record( { type => $stream, content => $chunk } );
   }
}

sub _flush_streams
{
   my $self = shift;

   if( length $self->{stdout} ) {
      $self->_print_stream( $self->{stdout}, FCGI_STDOUT );
      $self->{stdout} = "";
   }
   elsif( my $cb = $self->{stdout_cb} ) {
      $cb->();
   }

   if( length $self->{stderr} ) {
      $self->_print_stream( $self->{stderr}, FCGI_STDERR );
      $self->{stderr} = "";
   }
}

sub _needs_flush
{
   my $self = shift;
   return defined $self->{stdout_cb};
}

=head2 print_stdout

   $req->print_stdout( $data );

This method appends the given data to the STDOUT stream of the FastCGI
request, sending it to the webserver to be sent to the client.

=cut

sub print_stdout
{
   my $self = shift;
   my ( $data ) = @_;

   my $codec = $self->{codec};

   $self->{stdout} .= $codec ? $codec->encode( $data ) : $data;

   $self->{conn}->_req_needs_flush( $self );
}

=head2 print_stderr

   $req->print_stderr( $data );

This method appends the given data to the STDERR stream of the FastCGI
request, sending it to the webserver.

=cut

sub print_stderr
{
   my $self = shift;
   my ( $data ) = @_;

   my $codec = $self->{codec};

   $self->{used_stderr} = 1;
   $self->{stderr} .= $codec ? $codec->encode( $data ) : $data;

   $self->{conn}->_req_needs_flush( $self );
}

=head2 stream_stdout_then_finish

   $req->stream_stdout_then_finish( $readfn, $exitcode );

This method installs a callback for streaming data to the STDOUT stream.
Whenever the output stream is otherwise-idle, the function will be called to
generate some more data to output. When this function returns C<undef> it
indicates the end of the stream, and the request will be finished with the
given exit code.

If this method is used, then care should be taken to ensure that the number of
bytes written to the server matches the number that was claimed in the
C<Content-Length>, if such was provided. This logic should be performed by the
containing application; C<Net::Async::FastCGI> will not track it.

=cut

sub stream_stdout_then_finish
{
   my $self = shift;
   my ( $readfn, $exitcode ) = @_;

   $self->{stdout_cb} = sub {
      my $data = $readfn->();

      if( defined $data ) {
         $self->print_stdout( $data );
      }
      else {
         delete $self->{stdout_cb};
         $self->finish( $exitcode );
      }
   };

   $self->{conn}->_req_needs_flush( $self );
}

=head2 stdin

   $stdin = $req->stdin;

Returns an IO handle representing the request's STDIN buffer. This may be read
from using the C<read> or C<readline> functions or the C<< <$stdin> >>
operator.

Note that this will be a tied IO handle, it will not be useable directly as an
OS-level filehandle.

=cut

sub stdin
{
   my $self = shift;

   return Net::Async::FastCGI::Request::TiedHandle->new(
      READ => sub { 
         $_[1] = $self->read_stdin( $_[2] );
         return defined $_[1] ? length $_[1] : 0;
      },
      READLINE => sub {
         return $self->read_stdin_line;
      },
   );
}

=head2 stdout

=head2 stderr

   $stdout = $req->stdout;

   $stderr = $req->stderr;

Returns an IO handle representing the request's STDOUT or STDERR streams
respectively. These may written to using C<print>, C<printf>, C<say>, etc..

Note that these will be tied IO handles, they will not be useable directly as
an OS-level filehandle.

=cut

sub _stdouterr
{
   my $self = shift;
   my ( $method ) = @_;

   return Net::Async::FastCGI::Request::TiedHandle->new(
      WRITE => sub { $self->$method( $_[1] ) },
   );
}

sub stdout
{
   return shift->_stdouterr( "print_stdout" );
}

sub stderr
{
   return shift->_stdouterr( "print_stderr" );
}

=head2 finish

   $req->finish( $exitcode );

When the request has been dealt with, this method should be called to indicate
to the webserver that it is finished. After calling this method, no more data
may be appended to the STDOUT stream. At some point after calling this method,
the request object will be removed from the containing C<Net::Async::FastCGI>
object, once all the buffered outbound data has been sent.

If present, C<$exitcode> should indicate the numeric status code to send to
the webserver. If absent, a value of C<0> is presumed.

=cut

sub finish
{
   my $self = shift;
   my ( $exitcode ) = @_;

   return if $self->is_aborted;

   $self->_flush_streams;

   # Signal the end of STDOUT
   $self->write_record( { type => FCGI_STDOUT, content => "" } );

   # Signal the end of STDERR if we used it
   $self->write_record( { type => FCGI_STDERR, content => "" } ) if $self->{used_stderr};

   $self->write_record( { type => FCGI_END_REQUEST, 
         content => build_end_request_body( $exitcode || 0, FCGI_REQUEST_COMPLETE )
   } );

   my $conn = $self->{conn};

   if( $self->{keepconn} ) {
      $conn->_removereq( $self->{reqid} );
   }
   else {
      $conn->close;
   }
}

=head2 stdout_with_close

   $stdout = $req->stdout_with_close;

Similar to the C<stdout> method, except that when the C<close> method is
called on the returned filehandle, the request will be finished by calling
C<finish>.

=cut

sub stdout_with_close
{
   my $self = shift;

   return Net::Async::FastCGI::Request::TiedHandle->new(
      WRITE => sub { $self->print_stdout( $_[1] ) },
      CLOSE => sub { $self->finish( 0 ) },
   );
}

sub _abort
{
   my $self = shift;
   $self->{aborted} = 1;

   my $conn = $self->{conn};
   $conn->_removereq( $self->{reqid} );

   delete $self->{stdout_cb};
}

=head2 is_aborted

   $req->is_aborted;

Returns true if the webserver has already closed the control connection. No
further work on this request is necessary, as it will be discarded.

It is not required to call this method; if the request is aborted then any
output will be discarded. It may however be useful to call just before
expensive operations, in case effort can be avoided if it would otherwise be
wasted.

=cut

sub is_aborted
{
   my $self = shift;
   return $self->{aborted};
}

=head1 HTTP::Request/Response Interface

The following pair of methods form an interface that allows the request to be
used as a source of L<HTTP::Request> objects, responding to them by sending
L<HTTP::Response> objects. This may be useful to fit it in to existing code
that already uses these.

=cut

=head2 as_http_request

   $http_req = $req->as_http_request;

Returns a new C<HTTP::Request> object that gives a reasonable approximation to
the request. Because the webserver has translated the original HTTP request
into FastCGI parameters, this may not be a perfect recreation of the request
as received by the webserver.

=cut

sub as_http_request
{
   my $self = shift;

   require HTTP::Request;

   my $params = $self->params;

   my $authority =
      ( $params->{HTTP_HOST} || $params->{SERVER_NAME} || "" ) . ":" .
      ( $params->{SERVER_PORT} || "80" );

   my $path = $self->path;
   my $query_string = $self->query_string;

   $path .= "?$query_string" if length $query_string;

   my $uri = URI->new( "http://$authority$path" )->canonical;

   my @headers;

   # Content-Type and Content-Length come specially
   push @headers, "Content-Type"   => $params->{CONTENT_TYPE}
      if exists $params->{CONTENT_TYPE};

   push @headers, "Content-Length" => $params->{CONTENT_LENGTH}
      if exists $params->{CONTENT_LENGTH};

   # Pull all the HTTP_FOO parameters as headers. These will be in all-caps
   # and use _ for word separators, but HTTP::Headers can cope
   foreach ( keys %$params ) {
      m/^HTTP_(.*)$/ and push @headers, $1 => $params->{$_};
   }

   my $content = $self->{stdin};

   my $req = HTTP::Request->new( $self->method, $uri, \@headers, $content );

   $req->protocol( $self->protocol );

   return $req;
}

=head2 send_http_response

   $req->send_http_response( $resp );

Sends the given C<HTTP::Response> object as the response to this request. The
status, headers and content are all written out to the request's STDOUT stream
and then the request is finished with 0 as the exit code.

=cut

sub send_http_response
{
   my $self = shift;
   my ( $resp ) = @_;

   # (Fast)CGI suggests this is the way to report the status
   $resp->header( Status => $resp->code );

   my $topline = $resp->protocol . " " . $resp->status_line;

   $self->print_stdout( $topline . $CRLF );
   $self->print_stdout( $resp->headers_as_string( $CRLF ) );

   $self->print_stdout( $CRLF );

   $self->print_stdout( $resp->content );
   $self->finish( 0 );
}

package # hide from CPAN
   Net::Async::FastCGI::Request::TiedHandle;
use base qw( Tie::Handle );

use Symbol qw( gensym );

sub new
{
   my $class = shift;

   my $handle = gensym;
   tie *$handle, $class, @_;

   return $handle;
}

sub TIEHANDLE
{
   my $class = shift;
   return bless { @_ }, $class;
}

sub CLOSE    { shift->{CLOSE}->( @_ ) }
sub READ     { shift->{READ}->( @_ ) }
sub READLINE { shift->{READLINE}->( @_ ) }
sub WRITE    { shift->{WRITE}->( @_ ) }

=head1 EXAMPLES

=head2 Streaming A File

To serve contents of files on disk, it may be more efficient to use
C<stream_stdout_then_finish>:

   use Net::Async::FastCGI;
   use IO::Async::Loop;

   my $fcgi = Net::Async::FastCGI->new(
      on_request => sub {
         my ( $fcgi, $req ) = @_;

         open( my $file, "<", "/path/to/file" );
         $req->print_stdout( "Status: 200 OK\r\n" .
                             "Content-type: application/octet-stream\r\n" .
                             "\r\n" );

         $req->stream_stdout_then_finish(
            sub { read( $file, my $buffer, 8192 ) or return undef; return $buffer },
            0
         );
      }

   my $loop = IO::Async::Loop->new();

   $loop->add( $fcgi );

   $loop->run;

It may be more efficient again to instead use the C<X-Sendfile> feature of
certain webservers, which allows the webserver itself to serve the file
efficiently. See your webserver's documentation for more detail.

=head1 AUTHOR

Paul Evans <leonerd@leonerd.org.uk>

=cut

0x55AA;