1.13 Apr 1, 2006
- Add another fix for filename handling in the tests. Tests were
giving false failures on Win32 because the tests needed to use
File::Spec->canonpath(), just like Devel::StackTrace does internally.
2.07 Wed 10 May 2006
- This release contains only build-time changes
- AutoInstall is only needed for options, so remove auto_install
2.06 Sun 23 Apr 2006
- No functional changes.
- Moved test.conf to the root dir, removing last use of File::Spec
- It also means we don't need FindBin, so removed that too
- Upgrading to Module::Install 0.62
2.05 Thu 23 Feb 2006
- No functional changes.
- Moved over from the old CVS repository to the new SVN one
- Updated tests for the new release system
- Upgrading to a newer Module::Install
1.42 - 3 July 2006
* Ticket #18986 - ExtUtils::Install 1.39 or better fix es the in-use
issue on win32/cygwin, so make the code that checks whether trying
to install via the cpan shell conditional on the version of
ExtUtils::Install.
1.16 Wed 10 May 2006
- This release contains only build-time changes
- AutoInstall is only needed for options, so remove auto_install
1.15 Sun 7 May 2006
- This release contains only build-time changes
- Upgrading to Module::Install 0.62
1.14 Sun 8 Apr 2006
- This release contains only build-time changes
- Moved from older CVS to newer SVN repository
- Upgraded to Module::Install 0.61
1.26 Wed Apr 12 2006
- Debugging change in 1.25 regressed an undocumented 5.6-related
bug first fixed in 1.20. Make sure $DEBUG is always defined
(Michael Schwern)
- Re-introduced the List::Util 1.18 dependency, but stay at 1.17
if on a Win32 (and cygwin, just in case) platform.
- Module::Install updated to pre-release of 0.62
1.25 Fri Apr 7 2006
- Expands the can hook to also hook isa (Michael Schwern)
- Minor change to debugging (Michael Schwern)
- Optimised out the warnings.pm hook change (Michael Schwern)
- Module::Install updated to 0.61
1.05 2006-05-26-08-22
- release version 1.05
- added test for auto_purge_on_get
- fixed infinite loop with auto_purge_on_get
- fixed directory paths on Windows partitions
- NullCache uses BaseCache
1.102 Sat 16 Apr 2006
- No functional changes, no need to upgrade
- Moving from old CVS to new SVN repository
- Doing some general cleaning up
- Updating to Module::Install 0.62
1.01 06/04/2006
- Re-release as non developer release.
No changes from 1.00_01.
1.00_01 05/29/2006
- As DBI has supported only perl 5.6.0 since 2003
v1.38 Apache::DBI now requires perl 5.6.0 as well.
- Fix a plethora of uninitialized variable warnings,
general code cleanup, don't import unneeded symbols
from Carp, Digest::SHA1, and Digest::MD5
- http://rt.cpan.org/Ticket/Display.html?id=17073
$sth->rows is inconsistent across DBD::* drivers
and sometimes always returns 0. We were using
this to distinguish between a blank password and
no passwd. Now we don't call this function.
- http://rt.cpan.org/Ticket/Display.html?id=17422
a fatal error involving mp1, mp2 constants co-existance
was fixed in AuthDBI.
- http://rt.cpan.org/Ticket/Display.html?id=17446
under mod_perl 2, the check to skip caching connections
at server startup was broken; thus, causing children
to incorrectly share dbh handles with the parent.
- http://rt.cpan.org/Ticket/Display.html?id=19491
a critical return was missing connect() under mod_perl2
- Moved module's repository to its new home in SVN from CVS
http://svn.perl.org/modules/Apache-DBI
Summary of changes from 0.11.8.1 to 0.11.8.2
============================================
[FT fea] CTCP requests and replies will now be stacked onto outgoing
messages and autoreplies, respectively. This feature applies to
both IRC and AIM.
[TOC fea] Preliminary support for having multiple clients connected
to the same screen name on AIM. naim is now able to keep buddy
lists synchronized between all clients. Eliminates warnings about
unknown INSERTED2 and DELETED2 messages.
[FT fea] Misc. code cleanups and simplifications backported from
ngnaim.
[TOC bug] Apparently, some users can sign on to AIM in such a way
as to mask their signon time. FireTalk used a signon time of 0 to
designate "offline", which led to some confusion when users signed
on with an actual reported signon time of 0. This led to assertion
failures in toc2.c. Fixed.
[IRC fea] A user's user@host is now displayed when the user joins
a channel on IRC.
[FT bug] In some cases, CTCP requests were compared to the list
of known types case insensitively, sometimes case sensitively. The
spec. requires sensitivity to case. Fixed.
[IRC fea] All messages sent without a source are treated as coming
from a pseudo-target of ":SERVER".
[IRC fea] All messages sent before client signon is completed that
are not handled by the signon parser are passed to the normal
parser. This allows forced joins, CTCP handling from bots during
signon, etc.
[FT fea] Some IRC servers send CTCP VERSION requests during signon,
before 001 (IRC's "you are now signed on and can send messages"
protocol message), and demand a reply. naim was naively replying,
but FireTalk was suppressing the reply (since it hadn't been told
it could send messages yet). A hack has been put in place to allow
CTCP replies to go through before signon has completed.
[IRC bug] The /SETPRIV command did not have a handler in irc.c.
This causes a reliable crash. Fixed.
[UI bug] In some situations, naim could report a negative number
of lines left to play back during scrollback. Workaround implemented.
[UI fea] /open WINDOW now jumps you to WINDOW if WINDOW already
exists.
[UI fea] Overusing nickname completion now only complains to you
a few times, then stops. Previously it complained forever unless
an obscure setting was toggled.
[UI fea] New $ctcpverbose. If set, unknown CTCP requests are
displayed in the UI.
[pkg bug] Most tail utilities accept either "tail +3 file" or "tail
-n +3 file" the same. Some only accept "tail -n +3 file". Others
only accept "tail +3 file". naim's packaging no longer uses the
tail utility, at all.
[UI fea] The /sync command has been removed.
[pkg fea] The --firetalk-lily-binary-link configure option has been
removed, along with all code paths that depended on it.
[UI fea] Improved error condition reporting backported from ngnaim.
[UI bug] In some situations, file transfers could report incorrect
completion statistics due to integer overflow. All calculations
are performed using floating point now, though the UI still only
displays whole numbers.
[UI bug] A change in how windows were looked up made in 0.11.8.1
caused all messages to both users and chats to be sent using the
send-to-user API. This did not cause problems on IRC or Lily, but
did cause chat room messages to be silently eaten on AIM. Fixed.
[UI fea] At some point since the 0.11.8.1 release, AOL's servers
began changing bare \n characters in profiles to the string <br/>\n,
which naim did not understand. naim now treats <br/> as a synonym
for <br>.
[UI fea] The dynamic library subsystem is initialized before
.naimrc is parsed, allowing your .naimrc to use the modload command.
had actually been ignoring LTCONFIG_OVERRIDE anyway and just using
the default LIBTOOL_OVERRIDE to replace libtool scripts in packages.
This just formalizes the fact that LTCONFIG_OVERRIDE is not used
meaningfully by pkgsrc.
perllink script. This was causing errors when building perl from
pkgsrc. The perllink script didn't create any symlinks because it
tried to invoke "perl", which silently fails with revision 1.20 of
mk/tools/perl.mk (after the pkgsrc-2006Q2 branch).
Bump the PKGREVISION to 3.
explicitly setting CC=cc, so override it on the bmake command line.
We use the CC taken from the environment first, and default to "gcc"
if CC is not set.
the old bsd.pkg.mk behavior (noted by adrianp in private email). Keep
"makedistinfo" as an alias for "distinfo".
While here, sprinkle some .PHONY declarations for correctness.
in do-fetch-file (formerly they were inserted by the old _FETCH_FILE
inserting some quotes that are no longer added by the fetch command list
macro). This makes fetching with MASTER_SORT* set work again.
Effectively, it's a packet sniffer which runs as a background process
on a cable/DSL router, gathers all sorts of useless but interesting
statistics, and serves them over HTTP.
Features:
* Traffic graphs.
* Tracks traffic per host.
* Tracks traffic per TCP and UDP port for each host.
* Embedded web-server with deflate compression.
* Asynchronous reverse DNS resolution using a child process.
* Small. Portable. Single-threaded. Efficient.
==[ MAJOR BUGS FIXED ]==
- Spamfilter was not always working properly
- MS Visual studio 2005 (8.x) was unable to compile Unreal and/or caused crashes
- Certain IPv6 listen blocks could crash the ircd on-boot/on-rehash
==[ MINOR BUGS FIXED ]==
- "Looking up your hostname" message was missing if
set::options::show-connect-notice was enabled (other messages, like "looking
up ident" were shown, however)
- It was sometimes impossible to update a link { } block: all old settings
would still be used, this happened if connfreq was low. This might also have
caused crashes.
- Netsynch problem, which could cause the wrong modes to be applied to a
channel in some rare cases.
- Setting set::maxdccallow to 0 (or lower) still allowed one entry to be added
- Spamfilter oversized-checking is no longer done when removing a spamfilter
- Operator count bug (there might still be others...)
- Some chinese-* charsets could not be selected individually
- No longer requiring a C++ compiler (was caused by resolver in 3.2.4)
- Added workaround for "make: Permission denied" bug in some FreeBSD's
* All the smarts is now encapsulated in the "fetch" script. The fetch
script understands how to use the distinfo file (if specified) to
look up the size and checksums of the file to fetch and will use
that information to verify checksums of the fetched files or resume
transfers of interrupted fetches.
* Move the default settings for FETCH_RESUME_ARGS and FETCH_OUTPUT_ARGS
for "ftp" from mk/defaults/mk.conf into mk/fetch/fetch.mk. We rewrite
it to avoid needing conditional statements.
* Avoid spawning a new make(1) process just to mirror a distfile.
* Split out fetch-list targets into a separate file fetch-list.mk.
These targets should probably be moved into a standalone script.
* Fix distclean target to properly remove partial downloads.
