1 | package AnyEvent::Handle; |
1 | package AnyEvent::Handle; |
2 | |
2 | |
3 | no warnings; |
3 | no warnings; |
4 | use strict; |
4 | use strict qw(subs vars); |
5 | |
5 | |
6 | use AnyEvent (); |
6 | use AnyEvent (); |
7 | use AnyEvent::Util qw(WSAEWOULDBLOCK); |
7 | use AnyEvent::Util qw(WSAEWOULDBLOCK); |
8 | use Scalar::Util (); |
8 | use Scalar::Util (); |
9 | use Carp (); |
9 | use Carp (); |
… | |
… | |
14 | |
14 | |
15 | AnyEvent::Handle - non-blocking I/O on file handles via AnyEvent |
15 | AnyEvent::Handle - non-blocking I/O on file handles via AnyEvent |
16 | |
16 | |
17 | =cut |
17 | =cut |
18 | |
18 | |
19 | our $VERSION = 4.21; |
19 | our $VERSION = 4.3; |
20 | |
20 | |
21 | =head1 SYNOPSIS |
21 | =head1 SYNOPSIS |
22 | |
22 | |
23 | use AnyEvent; |
23 | use AnyEvent; |
24 | use AnyEvent::Handle; |
24 | use AnyEvent::Handle; |
… | |
… | |
49 | |
49 | |
50 | This module is a helper module to make it easier to do event-based I/O on |
50 | This module is a helper module to make it easier to do event-based I/O on |
51 | filehandles. For utility functions for doing non-blocking connects and accepts |
51 | filehandles. For utility functions for doing non-blocking connects and accepts |
52 | on sockets see L<AnyEvent::Util>. |
52 | on sockets see L<AnyEvent::Util>. |
53 | |
53 | |
|
|
54 | The L<AnyEvent::Intro> tutorial contains some well-documented |
|
|
55 | AnyEvent::Handle examples. |
|
|
56 | |
54 | In the following, when the documentation refers to of "bytes" then this |
57 | In the following, when the documentation refers to of "bytes" then this |
55 | means characters. As sysread and syswrite are used for all I/O, their |
58 | means characters. As sysread and syswrite are used for all I/O, their |
56 | treatment of characters applies to this module as well. |
59 | treatment of characters applies to this module as well. |
57 | |
60 | |
58 | All callbacks will be invoked with the handle object as their first |
61 | All callbacks will be invoked with the handle object as their first |
59 | argument. |
62 | argument. |
60 | |
63 | |
|
|
64 | =head2 SIGPIPE is not handled by this module |
|
|
65 | |
|
|
66 | SIGPIPE is not handled by this module, so one of the practical |
|
|
67 | requirements of using it is to ignore SIGPIPE (C<$SIG{PIPE} = |
|
|
68 | 'IGNORE'>). At least, this is highly recommend in a networked program: If |
|
|
69 | you use AnyEvent::Handle in a filter program (like sort), exiting on |
|
|
70 | SIGPIPE is probably the right thing to do. |
|
|
71 | |
61 | =head1 METHODS |
72 | =head1 METHODS |
62 | |
73 | |
63 | =over 4 |
74 | =over 4 |
64 | |
75 | |
65 | =item B<new (%args)> |
76 | =item B<new (%args)> |
… | |
… | |
70 | |
81 | |
71 | =item fh => $filehandle [MANDATORY] |
82 | =item fh => $filehandle [MANDATORY] |
72 | |
83 | |
73 | The filehandle this L<AnyEvent::Handle> object will operate on. |
84 | The filehandle this L<AnyEvent::Handle> object will operate on. |
74 | |
85 | |
75 | NOTE: The filehandle will be set to non-blocking (using |
86 | NOTE: The filehandle will be set to non-blocking mode (using |
76 | AnyEvent::Util::fh_nonblocking). |
87 | C<AnyEvent::Util::fh_nonblocking>) by the constructor and needs to stay in |
|
|
88 | that mode. |
77 | |
89 | |
78 | =item on_eof => $cb->($handle) |
90 | =item on_eof => $cb->($handle) |
79 | |
91 | |
80 | Set the callback to be called when an end-of-file condition is detcted, |
92 | Set the callback to be called when an end-of-file condition is detected, |
81 | i.e. in the case of a socket, when the other side has closed the |
93 | i.e. in the case of a socket, when the other side has closed the |
82 | connection cleanly. |
94 | connection cleanly. |
83 | |
95 | |
|
|
96 | For sockets, this just means that the other side has stopped sending data, |
|
|
97 | you can still try to write data, and, in fact, one can return from the eof |
|
|
98 | callback and continue writing data, as only the read part has been shut |
|
|
99 | down. |
|
|
100 | |
84 | While not mandatory, it is highly recommended to set an eof callback, |
101 | While not mandatory, it is I<highly> recommended to set an eof callback, |
85 | otherwise you might end up with a closed socket while you are still |
102 | otherwise you might end up with a closed socket while you are still |
86 | waiting for data. |
103 | waiting for data. |
|
|
104 | |
|
|
105 | If an EOF condition has been detected but no C<on_eof> callback has been |
|
|
106 | set, then a fatal error will be raised with C<$!> set to <0>. |
87 | |
107 | |
88 | =item on_error => $cb->($handle, $fatal) |
108 | =item on_error => $cb->($handle, $fatal) |
89 | |
109 | |
90 | This is the error callback, which is called when, well, some error |
110 | This is the error callback, which is called when, well, some error |
91 | occured, such as not being able to resolve the hostname, failure to |
111 | occured, such as not being able to resolve the hostname, failure to |
92 | connect or a read error. |
112 | connect or a read error. |
93 | |
113 | |
94 | Some errors are fatal (which is indicated by C<$fatal> being true). On |
114 | Some errors are fatal (which is indicated by C<$fatal> being true). On |
95 | fatal errors the handle object will be shut down and will not be |
115 | fatal errors the handle object will be shut down and will not be usable |
|
|
116 | (but you are free to look at the current C<< ->rbuf >>). Examples of fatal |
|
|
117 | errors are an EOF condition with active (but unsatisifable) read watchers |
|
|
118 | (C<EPIPE>) or I/O errors. |
|
|
119 | |
96 | usable. Non-fatal errors can be retried by simply returning, but it is |
120 | Non-fatal errors can be retried by simply returning, but it is recommended |
97 | recommended to simply ignore this parameter and instead abondon the handle |
121 | to simply ignore this parameter and instead abondon the handle object |
98 | object when this callback is invoked. |
122 | when this callback is invoked. Examples of non-fatal errors are timeouts |
|
|
123 | C<ETIMEDOUT>) or badly-formatted data (C<EBADMSG>). |
99 | |
124 | |
100 | On callback entrance, the value of C<$!> contains the operating system |
125 | On callback entrance, the value of C<$!> contains the operating system |
101 | error (or C<ENOSPC>, C<EPIPE>, C<ETIMEDOUT> or C<EBADMSG>). |
126 | error (or C<ENOSPC>, C<EPIPE>, C<ETIMEDOUT> or C<EBADMSG>). |
102 | |
127 | |
103 | While not mandatory, it is I<highly> recommended to set this callback, as |
128 | While not mandatory, it is I<highly> recommended to set this callback, as |
… | |
… | |
135 | =item timeout => $fractional_seconds |
160 | =item timeout => $fractional_seconds |
136 | |
161 | |
137 | If non-zero, then this enables an "inactivity" timeout: whenever this many |
162 | If non-zero, then this enables an "inactivity" timeout: whenever this many |
138 | seconds pass without a successful read or write on the underlying file |
163 | seconds pass without a successful read or write on the underlying file |
139 | handle, the C<on_timeout> callback will be invoked (and if that one is |
164 | handle, the C<on_timeout> callback will be invoked (and if that one is |
140 | missing, an C<ETIMEDOUT> error will be raised). |
165 | missing, a non-fatal C<ETIMEDOUT> error will be raised). |
141 | |
166 | |
142 | Note that timeout processing is also active when you currently do not have |
167 | Note that timeout processing is also active when you currently do not have |
143 | any outstanding read or write requests: If you plan to keep the connection |
168 | any outstanding read or write requests: If you plan to keep the connection |
144 | idle then you should disable the timout temporarily or ignore the timeout |
169 | idle then you should disable the timout temporarily or ignore the timeout |
145 | in the C<on_timeout> callback. |
170 | in the C<on_timeout> callback, in which case AnyEvent::Handle will simply |
|
|
171 | restart the timeout. |
146 | |
172 | |
147 | Zero (the default) disables this timeout. |
173 | Zero (the default) disables this timeout. |
148 | |
174 | |
149 | =item on_timeout => $cb->($handle) |
175 | =item on_timeout => $cb->($handle) |
150 | |
176 | |
… | |
… | |
154 | |
180 | |
155 | =item rbuf_max => <bytes> |
181 | =item rbuf_max => <bytes> |
156 | |
182 | |
157 | If defined, then a fatal error will be raised (with C<$!> set to C<ENOSPC>) |
183 | If defined, then a fatal error will be raised (with C<$!> set to C<ENOSPC>) |
158 | when the read buffer ever (strictly) exceeds this size. This is useful to |
184 | when the read buffer ever (strictly) exceeds this size. This is useful to |
159 | avoid denial-of-service attacks. |
185 | avoid some forms of denial-of-service attacks. |
160 | |
186 | |
161 | For example, a server accepting connections from untrusted sources should |
187 | For example, a server accepting connections from untrusted sources should |
162 | be configured to accept only so-and-so much data that it cannot act on |
188 | be configured to accept only so-and-so much data that it cannot act on |
163 | (for example, when expecting a line, an attacker could send an unlimited |
189 | (for example, when expecting a line, an attacker could send an unlimited |
164 | amount of data without a callback ever being called as long as the line |
190 | amount of data without a callback ever being called as long as the line |
165 | isn't finished). |
191 | isn't finished). |
166 | |
192 | |
167 | =item autocork => <boolean> |
193 | =item autocork => <boolean> |
168 | |
194 | |
169 | When disabled (the default), then C<push_write> will try to immediately |
195 | When disabled (the default), then C<push_write> will try to immediately |
170 | write the data to the handle if possible. This avoids having to register |
196 | write the data to the handle, if possible. This avoids having to register |
171 | a write watcher and wait for the next event loop iteration, but can be |
197 | a write watcher and wait for the next event loop iteration, but can |
172 | inefficient if you write multiple small chunks (this disadvantage is |
198 | be inefficient if you write multiple small chunks (on the wire, this |
173 | usually avoided by your kernel's nagle algorithm, see C<low_delay>). |
199 | disadvantage is usually avoided by your kernel's nagle algorithm, see |
|
|
200 | C<no_delay>, but this option can save costly syscalls). |
174 | |
201 | |
175 | When enabled, then writes will always be queued till the next event loop |
202 | When enabled, then writes will always be queued till the next event loop |
176 | iteration. This is efficient when you do many small writes per iteration, |
203 | iteration. This is efficient when you do many small writes per iteration, |
177 | but less efficient when you do a single write only. |
204 | but less efficient when you do a single write only per iteration (or when |
|
|
205 | the write buffer often is full). It also increases write latency. |
178 | |
206 | |
179 | =item no_delay => <boolean> |
207 | =item no_delay => <boolean> |
180 | |
208 | |
181 | When doing small writes on sockets, your operating system kernel might |
209 | When doing small writes on sockets, your operating system kernel might |
182 | wait a bit for more data before actually sending it out. This is called |
210 | wait a bit for more data before actually sending it out. This is called |
183 | the Nagle algorithm, and usually it is beneficial. |
211 | the Nagle algorithm, and usually it is beneficial. |
184 | |
212 | |
185 | In some situations you want as low a delay as possible, which cna be |
213 | In some situations you want as low a delay as possible, which can be |
186 | accomplishd by setting this option to true. |
214 | accomplishd by setting this option to a true value. |
187 | |
215 | |
188 | The default is your opertaing system's default behaviour, this option |
216 | The default is your opertaing system's default behaviour (most likely |
189 | explicitly enables or disables it, if possible. |
217 | enabled), this option explicitly enables or disables it, if possible. |
190 | |
218 | |
191 | =item read_size => <bytes> |
219 | =item read_size => <bytes> |
192 | |
220 | |
193 | The default read block size (the amount of bytes this module will try to read |
221 | The default read block size (the amount of bytes this module will |
194 | during each (loop iteration). Default: C<8192>. |
222 | try to read during each loop iteration, which affects memory |
|
|
223 | requirements). Default: C<8192>. |
195 | |
224 | |
196 | =item low_water_mark => <bytes> |
225 | =item low_water_mark => <bytes> |
197 | |
226 | |
198 | Sets the amount of bytes (default: C<0>) that make up an "empty" write |
227 | Sets the amount of bytes (default: C<0>) that make up an "empty" write |
199 | buffer: If the write reaches this size or gets even samller it is |
228 | buffer: If the write reaches this size or gets even samller it is |
200 | considered empty. |
229 | considered empty. |
201 | |
230 | |
|
|
231 | Sometimes it can be beneficial (for performance reasons) to add data to |
|
|
232 | the write buffer before it is fully drained, but this is a rare case, as |
|
|
233 | the operating system kernel usually buffers data as well, so the default |
|
|
234 | is good in almost all cases. |
|
|
235 | |
202 | =item linger => <seconds> |
236 | =item linger => <seconds> |
203 | |
237 | |
204 | If non-zero (default: C<3600>), then the destructor of the |
238 | If non-zero (default: C<3600>), then the destructor of the |
205 | AnyEvent::Handle object will check wether there is still outstanding write |
239 | AnyEvent::Handle object will check whether there is still outstanding |
206 | data and will install a watcher that will write out this data. No errors |
240 | write data and will install a watcher that will write this data to the |
207 | will be reported (this mostly matches how the operating system treats |
241 | socket. No errors will be reported (this mostly matches how the operating |
208 | outstanding data at socket close time). |
242 | system treats outstanding data at socket close time). |
209 | |
243 | |
210 | This will not work for partial TLS data that could not yet been |
244 | This will not work for partial TLS data that could not be encoded |
211 | encoded. This data will be lost. |
245 | yet. This data will be lost. |
212 | |
246 | |
213 | =item tls => "accept" | "connect" | Net::SSLeay::SSL object |
247 | =item tls => "accept" | "connect" | Net::SSLeay::SSL object |
214 | |
248 | |
215 | When this parameter is given, it enables TLS (SSL) mode, that means it |
249 | When this parameter is given, it enables TLS (SSL) mode, that means |
216 | will start making tls handshake and will transparently encrypt/decrypt |
250 | AnyEvent will start a TLS handshake as soon as the conenction has been |
217 | data. |
251 | established and will transparently encrypt/decrypt data afterwards. |
218 | |
252 | |
219 | TLS mode requires Net::SSLeay to be installed (it will be loaded |
253 | TLS mode requires Net::SSLeay to be installed (it will be loaded |
220 | automatically when you try to create a TLS handle). |
254 | automatically when you try to create a TLS handle): this module doesn't |
|
|
255 | have a dependency on that module, so if your module requires it, you have |
|
|
256 | to add the dependency yourself. |
221 | |
257 | |
222 | For the TLS server side, use C<accept>, and for the TLS client side of a |
258 | Unlike TCP, TLS has a server and client side: for the TLS server side, use |
223 | connection, use C<connect> mode. |
259 | C<accept>, and for the TLS client side of a connection, use C<connect> |
|
|
260 | mode. |
224 | |
261 | |
225 | You can also provide your own TLS connection object, but you have |
262 | You can also provide your own TLS connection object, but you have |
226 | to make sure that you call either C<Net::SSLeay::set_connect_state> |
263 | to make sure that you call either C<Net::SSLeay::set_connect_state> |
227 | or C<Net::SSLeay::set_accept_state> on it before you pass it to |
264 | or C<Net::SSLeay::set_accept_state> on it before you pass it to |
228 | AnyEvent::Handle. |
265 | AnyEvent::Handle. |
229 | |
266 | |
230 | See the C<starttls> method if you need to start TLs negotiation later. |
267 | See the C<< ->starttls >> method for when need to start TLS negotiation later. |
231 | |
268 | |
232 | =item tls_ctx => $ssl_ctx |
269 | =item tls_ctx => $ssl_ctx |
233 | |
270 | |
234 | Use the given Net::SSLeay::CTX object to create the new TLS connection |
271 | Use the given C<Net::SSLeay::CTX> object to create the new TLS connection |
235 | (unless a connection object was specified directly). If this parameter is |
272 | (unless a connection object was specified directly). If this parameter is |
236 | missing, then AnyEvent::Handle will use C<AnyEvent::Handle::TLS_CTX>. |
273 | missing, then AnyEvent::Handle will use C<AnyEvent::Handle::TLS_CTX>. |
237 | |
274 | |
238 | =item json => JSON or JSON::XS object |
275 | =item json => JSON or JSON::XS object |
239 | |
276 | |
240 | This is the json coder object used by the C<json> read and write types. |
277 | This is the json coder object used by the C<json> read and write types. |
241 | |
278 | |
242 | If you don't supply it, then AnyEvent::Handle will create and use a |
279 | If you don't supply it, then AnyEvent::Handle will create and use a |
243 | suitable one, which will write and expect UTF-8 encoded JSON texts. |
280 | suitable one (on demand), which will write and expect UTF-8 encoded JSON |
|
|
281 | texts. |
244 | |
282 | |
245 | Note that you are responsible to depend on the JSON module if you want to |
283 | Note that you are responsible to depend on the JSON module if you want to |
246 | use this functionality, as AnyEvent does not have a dependency itself. |
284 | use this functionality, as AnyEvent does not have a dependency itself. |
247 | |
285 | |
248 | =item filter_r => $cb |
286 | =item filter_r => $cb |
249 | |
287 | |
250 | =item filter_w => $cb |
288 | =item filter_w => $cb |
251 | |
289 | |
252 | These exist, but are undocumented at this time. |
290 | These exist, but are undocumented at this time. (They are used internally |
|
|
291 | by the TLS code). |
253 | |
292 | |
254 | =back |
293 | =back |
255 | |
294 | |
256 | =cut |
295 | =cut |
257 | |
296 | |
… | |
… | |
288 | delete $self->{_rw}; |
327 | delete $self->{_rw}; |
289 | delete $self->{_ww}; |
328 | delete $self->{_ww}; |
290 | delete $self->{fh}; |
329 | delete $self->{fh}; |
291 | |
330 | |
292 | $self->stoptls; |
331 | $self->stoptls; |
|
|
332 | |
|
|
333 | delete $self->{on_read}; |
|
|
334 | delete $self->{_queue}; |
293 | } |
335 | } |
294 | |
336 | |
295 | sub _error { |
337 | sub _error { |
296 | my ($self, $errno, $fatal) = @_; |
338 | my ($self, $errno, $fatal) = @_; |
297 | |
339 | |
… | |
… | |
307 | } |
349 | } |
308 | } |
350 | } |
309 | |
351 | |
310 | =item $fh = $handle->fh |
352 | =item $fh = $handle->fh |
311 | |
353 | |
312 | This method returns the file handle of the L<AnyEvent::Handle> object. |
354 | This method returns the file handle used to create the L<AnyEvent::Handle> object. |
313 | |
355 | |
314 | =cut |
356 | =cut |
315 | |
357 | |
316 | sub fh { $_[0]{fh} } |
358 | sub fh { $_[0]{fh} } |
317 | |
359 | |
… | |
… | |
335 | $_[0]{on_eof} = $_[1]; |
377 | $_[0]{on_eof} = $_[1]; |
336 | } |
378 | } |
337 | |
379 | |
338 | =item $handle->on_timeout ($cb) |
380 | =item $handle->on_timeout ($cb) |
339 | |
381 | |
340 | Replace the current C<on_timeout> callback, or disables the callback |
382 | Replace the current C<on_timeout> callback, or disables the callback (but |
341 | (but not the timeout) if C<$cb> = C<undef>. See C<timeout> constructor |
383 | not the timeout) if C<$cb> = C<undef>. See the C<timeout> constructor |
342 | argument. |
384 | argument and method. |
343 | |
385 | |
344 | =cut |
386 | =cut |
345 | |
387 | |
346 | sub on_timeout { |
388 | sub on_timeout { |
347 | $_[0]{on_timeout} = $_[1]; |
389 | $_[0]{on_timeout} = $_[1]; |
… | |
… | |
726 | |
768 | |
727 | if ( |
769 | if ( |
728 | defined $self->{rbuf_max} |
770 | defined $self->{rbuf_max} |
729 | && $self->{rbuf_max} < length $self->{rbuf} |
771 | && $self->{rbuf_max} < length $self->{rbuf} |
730 | ) { |
772 | ) { |
731 | return $self->_error (&Errno::ENOSPC, 1); |
773 | $self->_error (&Errno::ENOSPC, 1), return; |
732 | } |
774 | } |
733 | |
775 | |
734 | while () { |
776 | while () { |
735 | no strict 'refs'; |
|
|
736 | |
|
|
737 | my $len = length $self->{rbuf}; |
777 | my $len = length $self->{rbuf}; |
738 | |
778 | |
739 | if (my $cb = shift @{ $self->{_queue} }) { |
779 | if (my $cb = shift @{ $self->{_queue} }) { |
740 | unless ($cb->($self)) { |
780 | unless ($cb->($self)) { |
741 | if ($self->{_eof}) { |
781 | if ($self->{_eof}) { |
742 | # no progress can be made (not enough data and no data forthcoming) |
782 | # no progress can be made (not enough data and no data forthcoming) |
743 | $self->_error (&Errno::EPIPE, 1), last; |
783 | $self->_error (&Errno::EPIPE, 1), return; |
744 | } |
784 | } |
745 | |
785 | |
746 | unshift @{ $self->{_queue} }, $cb; |
786 | unshift @{ $self->{_queue} }, $cb; |
747 | last; |
787 | last; |
748 | } |
788 | } |
… | |
… | |
756 | && !@{ $self->{_queue} } # and the queue is still empty |
796 | && !@{ $self->{_queue} } # and the queue is still empty |
757 | && $self->{on_read} # but we still have on_read |
797 | && $self->{on_read} # but we still have on_read |
758 | ) { |
798 | ) { |
759 | # no further data will arrive |
799 | # no further data will arrive |
760 | # so no progress can be made |
800 | # so no progress can be made |
761 | $self->_error (&Errno::EPIPE, 1), last |
801 | $self->_error (&Errno::EPIPE, 1), return |
762 | if $self->{_eof}; |
802 | if $self->{_eof}; |
763 | |
803 | |
764 | last; # more data might arrive |
804 | last; # more data might arrive |
765 | } |
805 | } |
766 | } else { |
806 | } else { |
… | |
… | |
768 | delete $self->{_rw}; |
808 | delete $self->{_rw}; |
769 | last; |
809 | last; |
770 | } |
810 | } |
771 | } |
811 | } |
772 | |
812 | |
|
|
813 | if ($self->{_eof}) { |
|
|
814 | if ($self->{on_eof}) { |
773 | $self->{on_eof}($self) |
815 | $self->{on_eof}($self) |
774 | if $self->{_eof} && $self->{on_eof}; |
816 | } else { |
|
|
817 | $self->_error (0, 1); |
|
|
818 | } |
|
|
819 | } |
775 | |
820 | |
776 | # may need to restart read watcher |
821 | # may need to restart read watcher |
777 | unless ($self->{_rw}) { |
822 | unless ($self->{_rw}) { |
778 | $self->start_read |
823 | $self->start_read |
779 | if $self->{on_read} || @{ $self->{_queue} }; |
824 | if $self->{on_read} || @{ $self->{_queue} }; |
… | |
… | |
905 | $cb->($_[0], substr $_[0]{rbuf}, 0, $len, ""); |
950 | $cb->($_[0], substr $_[0]{rbuf}, 0, $len, ""); |
906 | 1 |
951 | 1 |
907 | } |
952 | } |
908 | }; |
953 | }; |
909 | |
954 | |
910 | # compatibility with older API |
|
|
911 | sub push_read_chunk { |
|
|
912 | $_[0]->push_read (chunk => $_[1], $_[2]); |
|
|
913 | } |
|
|
914 | |
|
|
915 | sub unshift_read_chunk { |
|
|
916 | $_[0]->unshift_read (chunk => $_[1], $_[2]); |
|
|
917 | } |
|
|
918 | |
|
|
919 | =item line => [$eol, ]$cb->($handle, $line, $eol) |
955 | =item line => [$eol, ]$cb->($handle, $line, $eol) |
920 | |
956 | |
921 | The callback will be called only once a full line (including the end of |
957 | The callback will be called only once a full line (including the end of |
922 | line marker, C<$eol>) has been read. This line (excluding the end of line |
958 | line marker, C<$eol>) has been read. This line (excluding the end of line |
923 | marker) will be passed to the callback as second argument (C<$line>), and |
959 | marker) will be passed to the callback as second argument (C<$line>), and |
… | |
… | |
938 | =cut |
974 | =cut |
939 | |
975 | |
940 | register_read_type line => sub { |
976 | register_read_type line => sub { |
941 | my ($self, $cb, $eol) = @_; |
977 | my ($self, $cb, $eol) = @_; |
942 | |
978 | |
943 | $eol = qr|(\015?\012)| if @_ < 3; |
979 | if (@_ < 3) { |
|
|
980 | # this is more than twice as fast as the generic code below |
|
|
981 | sub { |
|
|
982 | $_[0]{rbuf} =~ s/^([^\015\012]*)(\015?\012)// or return; |
|
|
983 | |
|
|
984 | $cb->($_[0], $1, $2); |
|
|
985 | 1 |
|
|
986 | } |
|
|
987 | } else { |
944 | $eol = quotemeta $eol unless ref $eol; |
988 | $eol = quotemeta $eol unless ref $eol; |
945 | $eol = qr|^(.*?)($eol)|s; |
989 | $eol = qr|^(.*?)($eol)|s; |
946 | |
990 | |
947 | sub { |
991 | sub { |
948 | $_[0]{rbuf} =~ s/$eol// or return; |
992 | $_[0]{rbuf} =~ s/$eol// or return; |
949 | |
993 | |
950 | $cb->($_[0], $1, $2); |
994 | $cb->($_[0], $1, $2); |
|
|
995 | 1 |
951 | 1 |
996 | } |
952 | } |
997 | } |
953 | }; |
998 | }; |
954 | |
|
|
955 | # compatibility with older API |
|
|
956 | sub push_read_line { |
|
|
957 | my $self = shift; |
|
|
958 | $self->push_read (line => @_); |
|
|
959 | } |
|
|
960 | |
|
|
961 | sub unshift_read_line { |
|
|
962 | my $self = shift; |
|
|
963 | $self->unshift_read (line => @_); |
|
|
964 | } |
|
|
965 | |
999 | |
966 | =item regex => $accept[, $reject[, $skip], $cb->($handle, $data) |
1000 | =item regex => $accept[, $reject[, $skip], $cb->($handle, $data) |
967 | |
1001 | |
968 | Makes a regex match against the regex object C<$accept> and returns |
1002 | Makes a regex match against the regex object C<$accept> and returns |
969 | everything up to and including the match. |
1003 | everything up to and including the match. |
… | |
… | |
1090 | register_read_type packstring => sub { |
1124 | register_read_type packstring => sub { |
1091 | my ($self, $cb, $format) = @_; |
1125 | my ($self, $cb, $format) = @_; |
1092 | |
1126 | |
1093 | sub { |
1127 | sub { |
1094 | # when we can use 5.10 we can use ".", but for 5.8 we use the re-pack method |
1128 | # when we can use 5.10 we can use ".", but for 5.8 we use the re-pack method |
1095 | defined (my $len = eval { unpack $format, $_[0]->{rbuf} }) |
1129 | defined (my $len = eval { unpack $format, $_[0]{rbuf} }) |
1096 | or return; |
1130 | or return; |
1097 | |
1131 | |
|
|
1132 | $format = length pack $format, $len; |
|
|
1133 | |
|
|
1134 | # bypass unshift if we already have the remaining chunk |
|
|
1135 | if ($format + $len <= length $_[0]{rbuf}) { |
|
|
1136 | my $data = substr $_[0]{rbuf}, $format, $len; |
|
|
1137 | substr $_[0]{rbuf}, 0, $format + $len, ""; |
|
|
1138 | $cb->($_[0], $data); |
|
|
1139 | } else { |
1098 | # remove prefix |
1140 | # remove prefix |
1099 | substr $_[0]->{rbuf}, 0, (length pack $format, $len), ""; |
1141 | substr $_[0]{rbuf}, 0, $format, ""; |
1100 | |
1142 | |
1101 | # read rest |
1143 | # read remaining chunk |
1102 | $_[0]->unshift_read (chunk => $len, $cb); |
1144 | $_[0]->unshift_read (chunk => $len, $cb); |
|
|
1145 | } |
1103 | |
1146 | |
1104 | 1 |
1147 | 1 |
1105 | } |
1148 | } |
1106 | }; |
1149 | }; |
1107 | |
1150 | |
… | |
… | |
1164 | |
1207 | |
1165 | require Storable; |
1208 | require Storable; |
1166 | |
1209 | |
1167 | sub { |
1210 | sub { |
1168 | # when we can use 5.10 we can use ".", but for 5.8 we use the re-pack method |
1211 | # when we can use 5.10 we can use ".", but for 5.8 we use the re-pack method |
1169 | defined (my $len = eval { unpack "w", $_[0]->{rbuf} }) |
1212 | defined (my $len = eval { unpack "w", $_[0]{rbuf} }) |
1170 | or return; |
1213 | or return; |
1171 | |
1214 | |
|
|
1215 | my $format = length pack "w", $len; |
|
|
1216 | |
|
|
1217 | # bypass unshift if we already have the remaining chunk |
|
|
1218 | if ($format + $len <= length $_[0]{rbuf}) { |
|
|
1219 | my $data = substr $_[0]{rbuf}, $format, $len; |
|
|
1220 | substr $_[0]{rbuf}, 0, $format + $len, ""; |
|
|
1221 | $cb->($_[0], Storable::thaw ($data)); |
|
|
1222 | } else { |
1172 | # remove prefix |
1223 | # remove prefix |
1173 | substr $_[0]->{rbuf}, 0, (length pack "w", $len), ""; |
1224 | substr $_[0]{rbuf}, 0, $format, ""; |
1174 | |
1225 | |
1175 | # read rest |
1226 | # read remaining chunk |
1176 | $_[0]->unshift_read (chunk => $len, sub { |
1227 | $_[0]->unshift_read (chunk => $len, sub { |
1177 | if (my $ref = eval { Storable::thaw ($_[1]) }) { |
1228 | if (my $ref = eval { Storable::thaw ($_[1]) }) { |
1178 | $cb->($_[0], $ref); |
1229 | $cb->($_[0], $ref); |
1179 | } else { |
1230 | } else { |
1180 | $self->_error (&Errno::EBADMSG); |
1231 | $self->_error (&Errno::EBADMSG); |
|
|
1232 | } |
1181 | } |
1233 | }); |
1182 | }); |
1234 | } |
|
|
1235 | |
|
|
1236 | 1 |
1183 | } |
1237 | } |
1184 | }; |
1238 | }; |
1185 | |
1239 | |
1186 | =back |
1240 | =back |
1187 | |
1241 | |
… | |
… | |
1265 | while ((my $len = Net::SSLeay::write ($self->{tls}, $self->{_tls_wbuf})) > 0) { |
1319 | while ((my $len = Net::SSLeay::write ($self->{tls}, $self->{_tls_wbuf})) > 0) { |
1266 | substr $self->{_tls_wbuf}, 0, $len, ""; |
1320 | substr $self->{_tls_wbuf}, 0, $len, ""; |
1267 | } |
1321 | } |
1268 | } |
1322 | } |
1269 | |
1323 | |
1270 | if (length ($buf = Net::SSLeay::BIO_read ($self->{_wbio}))) { |
|
|
1271 | $self->{wbuf} .= $buf; |
|
|
1272 | $self->_drain_wbuf; |
|
|
1273 | } |
|
|
1274 | |
|
|
1275 | while (defined ($buf = Net::SSLeay::read ($self->{tls}))) { |
1324 | while (defined ($buf = Net::SSLeay::read ($self->{tls}))) { |
1276 | if (length $buf) { |
1325 | unless (length $buf) { |
1277 | $self->{rbuf} .= $buf; |
|
|
1278 | $self->_drain_rbuf unless $self->{_in_drain}; |
|
|
1279 | } else { |
|
|
1280 | # let's treat SSL-eof as we treat normal EOF |
1326 | # let's treat SSL-eof as we treat normal EOF |
|
|
1327 | delete $self->{_rw}; |
1281 | $self->{_eof} = 1; |
1328 | $self->{_eof} = 1; |
1282 | $self->_shutdown; |
|
|
1283 | return; |
|
|
1284 | } |
1329 | } |
|
|
1330 | |
|
|
1331 | $self->{rbuf} .= $buf; |
|
|
1332 | $self->_drain_rbuf unless $self->{_in_drain}; |
|
|
1333 | |
|
|
1334 | $self->{tls} or return; # tls could have gone away |
1285 | } |
1335 | } |
1286 | |
1336 | |
1287 | my $err = Net::SSLeay::get_error ($self->{tls}, -1); |
1337 | my $err = Net::SSLeay::get_error ($self->{tls}, -1); |
1288 | |
1338 | |
1289 | if ($err!= Net::SSLeay::ERROR_WANT_READ ()) { |
1339 | if ($err!= Net::SSLeay::ERROR_WANT_READ ()) { |
… | |
… | |
1293 | return $self->_error (&Errno::EIO, 1); |
1343 | return $self->_error (&Errno::EIO, 1); |
1294 | } |
1344 | } |
1295 | |
1345 | |
1296 | # all others are fine for our purposes |
1346 | # all others are fine for our purposes |
1297 | } |
1347 | } |
|
|
1348 | |
|
|
1349 | if (length ($buf = Net::SSLeay::BIO_read ($self->{_wbio}))) { |
|
|
1350 | $self->{wbuf} .= $buf; |
|
|
1351 | $self->_drain_wbuf; |
|
|
1352 | } |
1298 | } |
1353 | } |
1299 | |
1354 | |
1300 | =item $handle->starttls ($tls[, $tls_ctx]) |
1355 | =item $handle->starttls ($tls[, $tls_ctx]) |
1301 | |
1356 | |
1302 | Instead of starting TLS negotiation immediately when the AnyEvent::Handle |
1357 | Instead of starting TLS negotiation immediately when the AnyEvent::Handle |
… | |
… | |
1333 | # basically, this is deep magic (because SSL_read should have the same issues) |
1388 | # basically, this is deep magic (because SSL_read should have the same issues) |
1334 | # but the openssl maintainers basically said: "trust us, it just works". |
1389 | # but the openssl maintainers basically said: "trust us, it just works". |
1335 | # (unfortunately, we have to hardcode constants because the abysmally misdesigned |
1390 | # (unfortunately, we have to hardcode constants because the abysmally misdesigned |
1336 | # and mismaintained ssleay-module doesn't even offer them). |
1391 | # and mismaintained ssleay-module doesn't even offer them). |
1337 | # http://www.mail-archive.com/openssl-dev@openssl.org/msg22420.html |
1392 | # http://www.mail-archive.com/openssl-dev@openssl.org/msg22420.html |
|
|
1393 | # |
|
|
1394 | # in short: this is a mess. |
|
|
1395 | # |
|
|
1396 | # note that we do not try to kepe the length constant between writes as we are required to do. |
|
|
1397 | # we assume that most (but not all) of this insanity only applies to non-blocking cases, |
|
|
1398 | # and we drive openssl fully in blocking mode here. |
1338 | Net::SSLeay::CTX_set_mode ($self->{tls}, |
1399 | Net::SSLeay::CTX_set_mode ($self->{tls}, |
1339 | (eval { local $SIG{__DIE__}; Net::SSLeay::MODE_ENABLE_PARTIAL_WRITE () } || 1) |
1400 | (eval { local $SIG{__DIE__}; Net::SSLeay::MODE_ENABLE_PARTIAL_WRITE () } || 1) |
1340 | | (eval { local $SIG{__DIE__}; Net::SSLeay::MODE_ACCEPT_MOVING_WRITE_BUFFER () } || 2)); |
1401 | | (eval { local $SIG{__DIE__}; Net::SSLeay::MODE_ACCEPT_MOVING_WRITE_BUFFER () } || 2)); |
1341 | |
1402 | |
1342 | $self->{_rbio} = Net::SSLeay::BIO_new (Net::SSLeay::BIO_s_mem ()); |
1403 | $self->{_rbio} = Net::SSLeay::BIO_new (Net::SSLeay::BIO_s_mem ()); |
… | |
… | |
1350 | }; |
1411 | }; |
1351 | $self->{filter_r} = sub { |
1412 | $self->{filter_r} = sub { |
1352 | Net::SSLeay::BIO_write ($_[0]{_rbio}, ${$_[1]}); |
1413 | Net::SSLeay::BIO_write ($_[0]{_rbio}, ${$_[1]}); |
1353 | &_dotls; |
1414 | &_dotls; |
1354 | }; |
1415 | }; |
|
|
1416 | |
|
|
1417 | &_dotls; # need to trigger the initial negotiation exchange |
1355 | } |
1418 | } |
1356 | |
1419 | |
1357 | =item $handle->stoptls |
1420 | =item $handle->stoptls |
1358 | |
1421 | |
1359 | Destroys the SSL connection, if any. Partial read or write data will be |
1422 | Destroys the SSL connection, if any. Partial read or write data will be |
… | |
… | |
1448 | =over 4 |
1511 | =over 4 |
1449 | |
1512 | |
1450 | =item * all constructor arguments become object members. |
1513 | =item * all constructor arguments become object members. |
1451 | |
1514 | |
1452 | At least initially, when you pass a C<tls>-argument to the constructor it |
1515 | At least initially, when you pass a C<tls>-argument to the constructor it |
1453 | will end up in C<< $handle->{tls} >>. Those members might be changes or |
1516 | will end up in C<< $handle->{tls} >>. Those members might be changed or |
1454 | mutated later on (for example C<tls> will hold the TLS connection object). |
1517 | mutated later on (for example C<tls> will hold the TLS connection object). |
1455 | |
1518 | |
1456 | =item * other object member names are prefixed with an C<_>. |
1519 | =item * other object member names are prefixed with an C<_>. |
1457 | |
1520 | |
1458 | All object members not explicitly documented (internal use) are prefixed |
1521 | All object members not explicitly documented (internal use) are prefixed |