----- Forwarded message from [EMAIL PROTECTED] -----
Date: Fri, 18 Feb 2000 21:38:09 +0100
Subject: CPAN Upload: JPRIT/Event-0.69.tar.gz
From: [EMAIL PROTECTED]
To: [EMAIL PROTECTED],
[EMAIL PROTECTED]
The uploaded file
Event-0.69.tar.gz
has entered CPAN as
file: $CPAN/authors/id/JPRIT/Event-0.69.tar.gz
size: 66867 bytes
md5: d3aa10f9141d4404b5570148f791be4e
No action is required on your part
Request entered by: JPRIT (Joshua N. Pritikin)
Request entered on: Fri, 18 Feb 2000 20:38:08 GMT
Request completed: Fri, 18 Feb 2000 20:38:09 GMT
Virtually Yours,
Id: paused,v 1.68 1999/10/22 14:39:12 k Exp k
# This is a patch for Event-0.68 to update it to Event-0.69
#
# To apply this patch:
# STEP 1: Chdir to the source directory.
# STEP 2: Run the 'applypatch' program with this patch file as input.
#
# If you do not have 'applypatch', it is part of the 'makepatch' package
# that you can fetch from the Comprehensive Perl Archive Network:
# http://www.perl.com/CPAN/authors/Johan_Vromans/makepatch-x.y.tar.gz
# In the above URL, 'x' should be 2 or higher.
#
# To apply this patch without the use of 'applypatch':
# STEP 1: Chdir to the source directory.
# STEP 2: Run the 'patch' program with this file as input.
#
#### End of Preamble ####
#### Patch data follows ####
gdiff -up '/usr/tmp/mp8092.d/old/Event-0.68/ChangeLog'
'/usr/tmp/mp8092.d/new/Event-0.69/ChangeLog'
Index: ./ChangeLog
--- ./ChangeLog Tue Feb 15 15:19:11 2000
+++ ./ChangeLog Fri Feb 18 15:30:29 2000
@@ -1,3 +1,11 @@
+2000-02-18 Joshua Pritikin <[EMAIL PROTECTED]>
+
+ * Release 0.69.
+
+2000-02-17 Joshua Pritikin <[EMAIL PROTECTED]>
+
+ * Tweak documentation.
+
2000-02-15 Joshua Pritikin <[EMAIL PROTECTED]>
* Release 0.68.
@@ -5,20 +13,20 @@
2000-02-14 Joshua Pritikin <[EMAIL PROTECTED]>
* More careful checking of
- intervals. (suggested by [EMAIL PROTECTED])
+ intervals. (suggested by [EMAIL PROTECTED])
2000-02-10 Joshua Pritikin <[EMAIL PROTECTED]>
- * Fix yet another typemap bug. ([EMAIL PROTECTED])
+ * Fix yet another typemap bug. ([EMAIL PROTECTED])
* Make io default to poll=r; more stringent start test;
- fix demo/stdin.t (suggested by [EMAIL PROTECTED])
+ fix demo/stdin.t (suggested by [EMAIL PROTECTED])
2000-02-09 Joshua Pritikin <[EMAIL PROTECTED]>
* Release 0.67. Hopefully this release will prove stable.
- * Fix bomb in group watchers. ([EMAIL PROTECTED])
+ * Fix bomb in group watchers. ([EMAIL PROTECTED])
* Fixed another bug introduced in 0.61. (Insure++)
@@ -35,7 +43,7 @@
* Eliminate (hopefully all) silent failures.
* Restrict var watchers to
- scalars. ([EMAIL PROTECTED] is too creative)
+ scalars. ([EMAIL PROTECTED] is too creative)
2000-02-07 Joshua Pritikin <[EMAIL PROTECTED]>
@@ -44,7 +52,7 @@
* Group watchers for watching watchers. See demo/group.t
* Fix got() documentation, spotted by
- [EMAIL PROTECTED]
+ [EMAIL PROTECTED]
* Make callback a prerequisite for starting watchers.
@@ -71,7 +79,7 @@
* Deprecated AUTOLOAD of Event->$type. (Nick & Graham)
* Reworked priority documentation based on comments from
- [EMAIL PROTECTED]
+ [EMAIL PROTECTED]
2000-02-01 Joshua Pritikin <[EMAIL PROTECTED]>
@@ -85,17 +93,17 @@
* Release 0.61.
* is_cancelled() method (requested by
- [EMAIL PROTECTED])
+ [EMAIL PROTECTED])
* Doc updates.
* Prevent starting unconfigured or badly configured
- watchers. Added tests. ([EMAIL PROTECTED])
+ watchers. Added tests. ([EMAIL PROTECTED])
XXX DANGER! THIS CHANGE CAN BREAK EXISTING CODE !DANGER XXX
* Refuse to start cancelled watchers. Added test & updated doc.
- ([EMAIL PROTECTED]).
+ ([EMAIL PROTECTED]).
2000-01-26 Joshua Pritikin <[EMAIL PROTECTED]>
@@ -115,7 +123,7 @@
* Release 0.59.
* Fixed serious memory leak (isolated by
- [EMAIL PROTECTED]). Added t/leak2.t.
+ [EMAIL PROTECTED]). Added t/leak2.t.
2000-01-19 Joshua Pritikin <[EMAIL PROTECTED]>
@@ -159,7 +167,7 @@
1999-09-23 Joshua Pritikin <[EMAIL PROTECTED]>
- * Document is_active/is_suspended. ([EMAIL PROTECTED])
+ * Document is_active/is_suspended. ([EMAIL PROTECTED])
1999-09-22 Joshua Pritikin <[EMAIL PROTECTED]>
@@ -323,7 +331,7 @@
1999-05-04 Joshua Pritikin <[EMAIL PROTECTED]>
- * Doc fixes catalyzed by [EMAIL PROTECTED]
+ * Doc fixes catalyzed by [EMAIL PROTECTED]
1999-04-27 Joshua Pritikin <[EMAIL PROTECTED]>
gdiff -up '/usr/tmp/mp8092.d/old/Event-0.68/TODO'
'/usr/tmp/mp8092.d/new/Event-0.69/TODO'
Index: ./TODO
--- ./TODO Tue Feb 15 15:09:02 2000
+++ ./TODO Thu Feb 17 11:04:19 2000
@@ -1,14 +1,19 @@
finish up lexical warnings
-deprecate inactivity completely?
-
-omitting parked=>1 should be a warning, not an error
+omitting parked=>1 should be a warning, not an error?
attribute set methods should return the new value
add timeout for var watchers?
+remove PE_QUEUED (unused?)
+
find owner for Win32 port
+
+I am planning to change $watcher->pending to return the T/F in a
+scalar context and the actual events in a list context.
+
+discontinue inactivity watchers
documentation
how to design new event types (in perl)
gdiff -up '/usr/tmp/mp8092.d/old/Event-0.68/lib/Event.pm'
'/usr/tmp/mp8092.d/new/Event-0.69/lib/Event.pm'
Index: ./lib/Event.pm
--- ./lib/Event.pm Tue Feb 15 15:09:11 2000
+++ ./lib/Event.pm Fri Feb 18 15:30:19 2000
@@ -13,7 +13,7 @@ use Carp;
eval { require Carp::Heavy; }; # work around perl_call_pv bug XXX
use vars qw($VERSION @EXPORT_OK
$API $DebugLevel $Eval $DIED $Now);
-$VERSION = '0.68';
+$VERSION = '0.69';
# If we inherit DynaLoader then we inherit AutoLoader; Bletch!
require DynaLoader;
gdiff -up '/usr/tmp/mp8092.d/old/Event-0.68/lib/Event.pod'
'/usr/tmp/mp8092.d/new/Event-0.69/lib/Event.pod'
Index: ./lib/Event.pod
--- ./lib/Event.pod Tue Feb 8 11:42:40 2000
+++ ./lib/Event.pod Thu Feb 17 11:46:13 2000
@@ -7,7 +7,7 @@ Event - Event loop processing
use Event qw(loop unloop);
# initialize application
- Event->type(attribute => value, ...);
+ Event->flavor(attribute => value, ...);
my $ret = loop();
@@ -54,14 +54,46 @@ Queue all pending events and dispatch an
than $max_prio (the highest priority is 0). The default is to process
all events except idle events. (While idle B<events> are ignored by
sweep, idle watchers are B<not> ignored. If you want to avoid
-triggering an idle watcher then set C<max> to C<undef> or suspend it.)
+triggering an idle watcher then set C<max> to C<undef> or C<stop()> it.)
=item one_event([$timeout])
If any events are outstanding then invoke the corresponding callback
of the highest priority event. If there are no events available,
block forever or until $timeout. Use of this API is not recommended
-because it is not efficient and does not trap exceptions.
+because it is not efficient and does not trap exceptions. However,
+you might wish to understand how it works:
+
+=over 4
+
+=item 1
+
+Queue asyncronous events (signals, etc).
+
+=item 2
+
+If there are any events with priority 5 or less (see StarvePrio) then
+service the next one and return.
+
+=item 3
+
+Calculate the maximum wait time (minimum time till the next timer
+expiration) and pass control to the poll/select system call. Upon
+return, queue all pending events.
+
+=item 4
+
+Queue asyncronous events again.
+
+=item 5
+
+If there are any events then service the next one and return.
+
+=item 6
+
+Service the next idle watcher.
+
+=back
=item all_watchers()
@@ -85,15 +117,14 @@ its C<max> wait time is exceeded.
All watchers are constructed in one of the following ways:
- $w = Event->type( [attr1 => $value,]... );
+ $w = Event->flavor( [attr1 => $value,]... );
-or
-
- use Event::type;
- $w = Event::type( [attr1 => $value,]...);
+ $w = Event::flavor($Class, [attr1 => $value,]...);
+
+ $w = Event::flavor->new([attr1 => $value,]...);
-Where I<type> is substituted with the kind of watcher. Built-in types
-include idle, io, signal, timer, and var.
+Where I<flavor> is substituted with the kind of watcher. Built-in
+types include idle, io, signal, timer, and var.
New watchers (hopefully) have reasonable defaults and can also be
customized by passing extra attributes to the constructor. When
@@ -114,10 +145,10 @@ For example:
warn $event->w->desc.": ".$event->hits." events happened; Wow!";
All watchers support the following attributes: cb, cbtime, debug,
-desc, prio, max_cb_tm, reentrant, repeat, and running. Watcher
-constructors accept the preceding and additionally: async and nice.
-All events support: hits, prio, and w. Moreover, watchers also offer
-extra attributes according to their specialty.
+desc, prio, max_cb_tm, reentrant, and repeat. Watcher constructors
+accept the preceding and additionally: async and nice. All events
+support: hits, prio, and w. Moreover, watchers also offer extra
+attributes according to their specialty.
=head2 Shared Watcher Methods
@@ -166,7 +197,7 @@ them of the ability to generate events.
Stop and destroy C<$watcher>. Running events are allowed to complete
but pending events are cancelled. Cancelled watchers are no longer
-valid except for read-only type operations. For example, prio() can
+valid except for read-only operations. For example, prio() can
return the watcher's priority, but start() will fail.
=item $watcher->is_cancelled
@@ -178,6 +209,13 @@ Reports whether the C<$watcher> has been
Reports whether the C<$watcher> has been started. The return value is
not affected by suspend.
+=item $watcher->is_running
+
+Zero if the callback is not running. Otherwise, the number of levels
+that the callback has been entered. This can be greater than one if a
+C<reentrant> callback invokes C<loop> (or C<sweep>, with lesser
+probability).
+
=item $watcher->is_suspended
Reports whether the C<$watcher> is suspended.
@@ -201,17 +239,6 @@ The callback is invoked only when no eve
never a chance to idle, an event will be generated at least every
C<max> seconds and not more often than C<min> seconds.
-=item inactivity (EXPERIMENTAL)
-
-Extra attributes: level => $level, timeout => $seconds
-
-Inactivity watchers are similar to idle watchers. The difference is
-that they judge idleness by whether there have been any events with
-priority higher than the threshold C<$level>.
-
-If you find these useful, then send email to the mailing list.
-Otherwise, they might be deprecated.
-
=item var
Extra attributes: var => \$var, poll => 'rw'
@@ -335,6 +362,9 @@ C<parked=>1>.
=item at => $time
+The expiration time in the same units as the system clock. For a
+timer, C<at> will usually be in the future.
+
=item cb => \&code
=item cb => [$class_or_object, $method_name]
@@ -366,10 +396,8 @@ constructed.
=item fd => $filehandle
-=item flags => $bits
-
-The C<flag> attribute encodes information about the state of an
-event. [XXX Fill in exact bits] [DEPRECATED?]
+This attribute can accept either a perl-esque filehandle or a system
+call derived file descriptor number.
=item hard => $bool
@@ -383,10 +411,6 @@ this can make a significant difference.
How long between repeating timeouts. The C<at> attribute is
recalculated using C<interval> upon callback return.
-=item level => $priority
-
-For inactivity - EXPERIMENTAL.
-
=item max => $seconds
The maximum number of seconds to wait before triggering the callback.
@@ -445,12 +469,6 @@ one-shot or continue waiting for new eve
depends on the type of watcher. I<io>, I<signal>, and I<var> default
to true.
-=item running => $int
-
-Zero if the callback is not running. Otherwise, the number of levels
-that the callback has been entered. This can be greater than one if
-a C<reentrant> callback invokes C<loop> (or to a lesser extent C<sweep>).
-
=item signal => $str
The callback will be invoked when the specified signal is received.
@@ -459,10 +477,13 @@ the documentation for C<%SIG>.
=item suspend => $bool
-Don't look for events any more. Running events are allowed to
-complete but queued events are cancelled and subsequent pending events
-are ignored. Suspend should be considered a debugging tool. If you
-actually want to stop a watcher then use the C<stop> method.
+Stop looking for events. Running events are allowed to complete, but
+queued events are cancelled.
+
+Suspend is for debugging. If you suspend all watchers in an
+application then you can examine the complete state unchanged for as
+long as you like without worrying about timer expirations. If you
+actually wish to stop a watcher then use the C<stop()> method.
=item timeout => $seconds
@@ -513,10 +534,6 @@ This method return the event's watcher.
=head2 Customization and Exceptions
-The bulk of Event's implementation is in C for B<maximum>
-performance. However, a broad range of customization hooks are
-available.
-
=over 4
=item * $Event::DebugLevel
@@ -524,31 +541,37 @@ available.
Enables progressively more debugging output. Meaningful levels range
from 1 (least output) to 5 (most output). Also see C<debug>.
-=item * $Event::Eval
-
-Strictly for debugging. Do not enable this unless you know what you
-are doing.
-
=item * $Event::DIED
When C<loop> or C<sweep> is called, an exception context is
established for the duration of event processing. If an exception is
-detected then C<$Event::DIED> is invoked:
-
- $Event::DIED->($event, $@);
-
-The default hook uses C<warn> to print the exception. After the DIED
-handler returns, event processing continues as if nothing happened.
-If you'd like to see more detailed output you can use the verbose
-handler (below) or write your own.
+detected then C<$Event::DIED> is invoked. The default hook uses
+C<warn> to output the exception. After the DIED handler completes,
+event processing continues as if nothing happened.
+
+If you'd like more detailed output you can install the verbose
+handler:
$Event::DIED = \&Event::verbose_exception_handler;
+Or you can write your own. The handler is invoked like this:
+
+ $Event::DIED->($event, $@);
+
+If you do not want to continue looping after an error, you can do
+something like this:
+
+ $Event::DIED = sub { die &Event::verbose_exception_handler(@_) };
+
=item * Event->add_hooks(key => sub { ... }, ...);
-The C<add_hooks> method allows customization at key points in the
-optimized event processing core. Currently support hooks are detailed
-below:
+The bulk of Event's implementation is in C for B<maximum> performance.
+The C<add_hooks> method allows insertion of perl code at key points in
+the optimized event processing core. While flexible, this can hurt
+performance *significantly*. If you want customization *and*
+performance, please see the C API.
+
+Currently support hooks are detailed as follows:
hook purpose
------------- ----------------------------------------------
@@ -591,8 +614,8 @@ rendered and to process subsequent incom
Threads can either substitute for an event loop or complement it.
Threads are similar to processes in that the operating system manages
task switching for you. However, the difference is that all threads
-share the same address space. This is good and bad. Much higher
-performance can be acheived but since data is shared between threads
+share the same address space. This is good and bad. Higher
+performance can be acheived but since data is shared between threads,
extreme care must be taken to access or modify global data. The
operating system can switch threads at any moment or can execute
multiple threads simultaineously. I hope this sounds dangerous! It
@@ -620,9 +643,9 @@ features you need. It is likely that al
implementation soon but at this point in history it is best to double
check.
-=head1 WHAT ABOUT NON-PREEMPTIVE THREADS?
+Many suggestions by Mark Mielke <[EMAIL PROTECTED]>
-(Contributed by [EMAIL PROTECTED] 12 Jul 1999.)
+=head1 WHAT ABOUT NON-PREEMPTIVE THREADS?
The Java language is oriented to use non-preemptive threads, yet even
Java uses an event-loop for Swing (AFAIK). That is one of the reasons
@@ -635,19 +658,47 @@ like working with a state machine. I fi
better code. It also has the benefit of abstracting away how
concurrency is achieved.
+Contributed by [EMAIL PROTECTED], 12 Jul 1999.
+
=head1 BUGS
The meaning of $io->timeout(0) might change. Use C<undef> to unset
the timeout.
+There seems to be some sort of bug in the global destruction phase:
+
+ Attempt to free unreferenced scalar during global destruction.
+ Use of uninitialized value during global destruction.
+ Explicit blessing to '' (assuming package main) during global
+ destruction.
+
=head2 WHY MICROSOFT WINDOWS MIGHT BE FASTER THAN UNIX
+B<A scalable and explicit event delivery mechanism for UNIX>
+
+I<Abstract:> UNIX applications not wishing to block when doing I/O often
+use the select() system call, to wait for events on multiple file
+descriptors. The select() mechanism works well for small-scale
+applications, but scales poorly as the number of file descriptors
+increases. Many modern applications, such as Internet servers, use
+hundreds or thousands of file descriptors, and suffer greatly from the
+poor scalability of select(). Previous work has shown that while the
+traditional implementation of select() can be improved, the poor
+scalability is inherent in the design. We present a new event-delivery
+mechanism, which allows the application to register interest in one or
+more sources of events, and to efficiently dequeue new events. We show
+that this mechanism, which requires only minor changes to
+applications, performs independently of the number of file
+descriptors.
+
http://www.usenix.org/events/usenix99/full_papers/banga/banga_html/index.html
=head1 ALSO SEE
L<Time::HiRes>, L<NetServer::ProcessTop>, and L<Time::Warp>.
+While Tk does not yet support Event, L<PerlQt> does.
+
=head1 SUPPORT
If you have insights or complaints then please subscribe to the
@@ -667,20 +718,17 @@ Thanks!
=head1 AUTHORS
-Joshua N. Pritikin E<lt>F<[EMAIL PROTECTED]>E<gt>
+Joshua N. Pritikin E<lt>F<[EMAIL PROTECTED]>E<gt>
Initial 0.01 implementation by Graham Barr
E<lt>F<[EMAIL PROTECTED]>E<gt>. Other contributors include at least
those lists below and folks mentioned in the ChangeLog.
- Gisle Aas E<lt>F<[EMAIL PROTECTED]>E<gt>
- E<lt>F<[EMAIL PROTECTED]>E<gt> (Win32)
- E<lt>F<[EMAIL PROTECTED]>E<gt> (File::Tail)
- Uri Guttman E<lt>F<[EMAIL PROTECTED]>E<gt>
- Nick Ing-Simmons E<lt>F<[EMAIL PROTECTED]>E<gt> (Tk)
- Mark Mielke E<lt>F<[EMAIL PROTECTED]>E<gt>
- Sarathy E<lt>F<[EMAIL PROTECTED]>E<gt>
- E<lt>F<[EMAIL PROTECTED]>E<gt>
+ Gisle Aas <[EMAIL PROTECTED]>
+ Uri Guttman <[EMAIL PROTECTED]>
+ Nick Ing-Simmons <[EMAIL PROTECTED]> (Tk)
+ Sarathy <[EMAIL PROTECTED]>
+ Jochen Stenzel <[EMAIL PROTECTED]>
=head1 COPYRIGHT
gdiff -up '/usr/tmp/mp8092.d/old/Event-0.68/lib/Event/inactivity.pm'
'/usr/tmp/mp8092.d/new/Event-0.69/lib/Event/inactivity.pm'
Index: ./lib/Event/inactivity.pm
--- ./lib/Event/inactivity.pm Wed Jan 26 15:55:44 2000
+++ ./lib/Event/inactivity.pm Thu Feb 17 11:04:40 2000
@@ -6,6 +6,8 @@ use base 'Event::Watcher::Tied';
use vars qw($DefaultPriority);
$DefaultPriority = PRIO_NORMAL + 1;
+warn "Event::inactivity is deprecated";
+
'Event::Watcher'->register();
sub new {
#### End of Patch data ####
#### ApplyPatch data follows ####
# Data version : 1.0
# Date generated : Fri Feb 18 15:36:23 2000
# Generated by : makepatch 2.00 (2.0BETA)
# Recurse directories : Yes
# p 'ChangeLog' 23116 950905829 0100444
# p 'TODO' 1753 950803459 0100444
# p 'lib/Event.pm' 4254 950905819 0100444
# p 'lib/Event.pod' 22308 950805973 0100444
# p 'lib/Event/inactivity.pm' 1292 950803480 0100444
#### End of ApplyPatch data ####
#### End of Patch kit [created: Fri Feb 18 15:36:23 2000] ####
#### Checksum: 570 19716 15454 ####
