| … | |
… | |
| 340 | =head2 CHILD PROCESS WATCHERS |
340 | =head2 CHILD PROCESS WATCHERS |
| 341 | |
341 | |
| 342 | You can also watch on a child process exit and catch its exit status. |
342 | You can also watch on a child process exit and catch its exit status. |
| 343 | |
343 | |
| 344 | The child process is specified by the C<pid> argument (if set to C<0>, it |
344 | The child process is specified by the C<pid> argument (if set to C<0>, it |
| 345 | watches for any child process exit). The watcher will trigger as often |
345 | watches for any child process exit). The watcher will triggered only when |
| 346 | as status change for the child are received. This works by installing a |
346 | the child process has finished and an exit status is available, not on |
| 347 | signal handler for C<SIGCHLD>. The callback will be called with the pid |
347 | any trace events (stopped/continued). |
| 348 | and exit status (as returned by waitpid), so unlike other watcher types, |
348 | |
| 349 | you I<can> rely on child watcher callback arguments. |
349 | The callback will be called with the pid and exit status (as returned by |
|
|
350 | waitpid), so unlike other watcher types, you I<can> rely on child watcher |
|
|
351 | callback arguments. |
|
|
352 | |
|
|
353 | This watcher type works by installing a signal handler for C<SIGCHLD>, |
|
|
354 | and since it cannot be shared, nothing else should use SIGCHLD or reap |
|
|
355 | random child processes (waiting for specific child processes, e.g. inside |
|
|
356 | C<system>, is just fine). |
| 350 | |
357 | |
| 351 | There is a slight catch to child watchers, however: you usually start them |
358 | There is a slight catch to child watchers, however: you usually start them |
| 352 | I<after> the child process was created, and this means the process could |
359 | I<after> the child process was created, and this means the process could |
| 353 | have exited already (and no SIGCHLD will be sent anymore). |
360 | have exited already (and no SIGCHLD will be sent anymore). |
| 354 | |
361 | |
| … | |
… | |
| 818 | =item L<AnyEvent::IGS> |
825 | =item L<AnyEvent::IGS> |
| 819 | |
826 | |
| 820 | A non-blocking interface to the Internet Go Server protocol (used by |
827 | A non-blocking interface to the Internet Go Server protocol (used by |
| 821 | L<App::IGS>). |
828 | L<App::IGS>). |
| 822 | |
829 | |
| 823 | =item L<Net::IRC3> |
830 | =item L<AnyEvent::IRC> |
| 824 | |
831 | |
| 825 | AnyEvent based IRC client module family. |
832 | AnyEvent based IRC client module family (replacing the older Net::IRC3). |
| 826 | |
833 | |
| 827 | =item L<Net::XMPP2> |
834 | =item L<Net::XMPP2> |
| 828 | |
835 | |
| 829 | AnyEvent based XMPP (Jabber protocol) module family. |
836 | AnyEvent based XMPP (Jabber protocol) module family. |
| 830 | |
837 | |
| … | |
… | |
| 854 | no warnings; |
861 | no warnings; |
| 855 | use strict qw(vars subs); |
862 | use strict qw(vars subs); |
| 856 | |
863 | |
| 857 | use Carp; |
864 | use Carp; |
| 858 | |
865 | |
| 859 | our $VERSION = 4.233; |
866 | our $VERSION = 4.3; |
| 860 | our $MODEL; |
867 | our $MODEL; |
| 861 | |
868 | |
| 862 | our $AUTOLOAD; |
869 | our $AUTOLOAD; |
| 863 | our @ISA; |
870 | our @ISA; |
| 864 | |
871 | |
| … | |
… | |
| 1021 | BEGIN { |
1028 | BEGIN { |
| 1022 | if (eval "use Time::HiRes (); time (); 1") { |
1029 | if (eval "use Time::HiRes (); time (); 1") { |
| 1023 | *_time = \&Time::HiRes::time; |
1030 | *_time = \&Time::HiRes::time; |
| 1024 | # if (eval "use POSIX (); (POSIX::times())... |
1031 | # if (eval "use POSIX (); (POSIX::times())... |
| 1025 | } else { |
1032 | } else { |
| 1026 | *_time = \&CORE::time; # epic fail |
1033 | *_time = sub { time }; # epic fail |
| 1027 | } |
1034 | } |
| 1028 | } |
1035 | } |
| 1029 | |
1036 | |
| 1030 | sub time { _time } |
1037 | sub time { _time } |
| 1031 | sub now { _time } |
1038 | sub now { _time } |
| … | |
… | |
| 1747 | watchers, as the management overhead dominates. |
1754 | watchers, as the management overhead dominates. |
| 1748 | |
1755 | |
| 1749 | =back |
1756 | =back |
| 1750 | |
1757 | |
| 1751 | |
1758 | |
|
|
1759 | =head1 SIGNALS |
|
|
1760 | |
|
|
1761 | AnyEvent currently installs handlers for these signals: |
|
|
1762 | |
|
|
1763 | =over 4 |
|
|
1764 | |
|
|
1765 | =item SIGCHLD |
|
|
1766 | |
|
|
1767 | A handler for C<SIGCHLD> is installed by AnyEvent's child watcher |
|
|
1768 | emulation for event loops that do not support them natively. Also, some |
|
|
1769 | event loops install a similar handler. |
|
|
1770 | |
|
|
1771 | =item SIGPIPE |
|
|
1772 | |
|
|
1773 | A no-op handler is installed for C<SIGPIPE> when C<$SIG{PIPE}> is C<undef> |
|
|
1774 | when AnyEvent gets loaded. |
|
|
1775 | |
|
|
1776 | The rationale for this is that AnyEvent users usually do not really depend |
|
|
1777 | on SIGPIPE delivery (which is purely an optimisation for shell use, or |
|
|
1778 | badly-written programs), but C<SIGPIPE> can cause spurious and rare |
|
|
1779 | program exits as a lot of people do not expect C<SIGPIPE> when writing to |
|
|
1780 | some random socket. |
|
|
1781 | |
|
|
1782 | The rationale for installing a no-op handler as opposed to ignoring it is |
|
|
1783 | that this way, the handler will be restored to defaults on exec. |
|
|
1784 | |
|
|
1785 | Feel free to install your own handler, or reset it to defaults. |
|
|
1786 | |
|
|
1787 | =back |
|
|
1788 | |
|
|
1789 | =cut |
|
|
1790 | |
|
|
1791 | $SIG{PIPE} = sub { } |
|
|
1792 | unless defined $SIG{PIPE}; |
|
|
1793 | |
|
|
1794 | |
| 1752 | =head1 FORK |
1795 | =head1 FORK |
| 1753 | |
1796 | |
| 1754 | Most event libraries are not fork-safe. The ones who are usually are |
1797 | Most event libraries are not fork-safe. The ones who are usually are |
| 1755 | because they rely on inefficient but fork-safe C<select> or C<poll> |
1798 | because they rely on inefficient but fork-safe C<select> or C<poll> |
| 1756 | calls. Only L<EV> is fully fork-aware. |
1799 | calls. Only L<EV> is fully fork-aware. |
