Patches by Adrian Portelli <adrianp@stindustries.net> (PR#20142)
Changes :
- Steve Oliphant pointed out that test case 105 did not work anymore
and this was due to a missing fix for the password prompting.
- Bryan Kemp pointed out that curl -u could not provide a blank password
without prompting the user. It can now. -u username: makes the password
empty, while -u username makes curl prompt the user for a password.
- Kjetil Jacobsen found a remaining connect problem in the multi interface
on ipv4 systems (Linux only?), that I fixed and Kjetil verified that it
fixed his problems.
- memanalyze.pl now reads a file name from the command line, and no longer
takes the data on stdin as before.
- Fixed tests/memanalyze.pl to work with file names that contain colons
(as on Windows).
- Kjetil Jacobsen quickly pointed out that lib/share.h was missing...
* For more, see CHANGES.
Changes from 0.45
+ Added patch from Pavel Hlavnicka for freeing memory leaks
from SSL_CTX_use_pkcs12_file() whose functionality is triggered
by the $ENV{HTTPS_PKCS12_*} settings
+ Set timeout to 15 seconds for ./net_ssl_test and lwp-ssl-test
sample scripts for better testing of timeout behavior
+ Added alarm() during Net::SSL->read() to honor socket timeout setting
for more robust applications. read() will die_with_error() which
in consistent with previous semantics used during SSL read() failure
Thanks to Pavel Hlavnicka for prompting this change.
+ Removed code that supported versions of SSLeay before version 0.8
I believe SSLeay v.8 was released back in 1998
+ Added patch from Devin Heitmueller so that initial random seed
would be taken from /dev/urandom if available via RAND_load_file API
from stable branch of SA CVS repository.
On other operating systems 'spamc' was reported to cause a core dump if
'spamd' was not running. At least NetBSD/i386 1.5.3 seems not to be as
severely affected, I only got 'spamc in free(): warning: junk
pointer, too high to make sense.'.
from stable branch of SA CVS repository.
On other operating systems 'spamc' was reported to cause a core dump if
'spamd' was not running. At least NetBSD/i386 1.5.3 seems not to be as
severely affected, I only got 'spamc in free(): warning: junk
pointer, too high to make sense.'.
gthread.la file doesn't note that pthread is a dependancy, so linking with
libtool against libgthread breaks. It seems the workaround for libtool not
understanding -pthread doesn't work, so we use the same thing as linux, and
convert -pthread to -lpthread. gtk2 now builds to completion on -current
Note that this has no impact on 1.6.1 as it doesn't use -pthread.
* Use genius2-1.1.
* Fix distinfo to match Makefile distfile list.
Changes since 3.2.2 include:
* eb/subbook.c (eb_set_subbook_epwing): Open `honmon2' for
sound or graphic file if `honmong' or `honmons' doesn't exist.
* eb/binary.c (eb_set_binary_wave): Fix a bug that the function
composes broken WAVE header.
* eb/text.h, eb/readtext.c: Define `EB_HOOK_BEGIN_DECORATION' and
`EB_HOOK_END_DECORATION', requested from Kenichi SUTO.
to deal with a bunch of broken packages on ftp.netbsd.org and on the
CD-ROMS, as one of the dependencies, faad2, gets automatically removed
by virtue of the properly noted patent restriction. That'll make any
mplayer binaries on the ftp server or on the 1.6.1 CD's impossible for
a user to install.
- - Actually set NO_SRC_ON_FTP to something, rather than the undefined
variable ${RESTRICTED}.
bulk builds, rather than choking up on the unfillfulled dependency
on realplayer, which package is truly interactive.
We have to make the INTERACTIVE_STAGE=fetch, because "configure" is
too late. It's really the "dependency installing phase" that's
interactive, but is there is no such phase -- dependencies are
installed pursuant to "extract". This whole thing needs to revisited
after the 1.6.1 freeze.
works around a bug with optimization in gcc-2.95.3 which prevents building
this pkg.
Optimization is still turned on during the build of the final libraries
and compilers.
Before using native pthreads per default, we should make sure it is
at least stable enough for the Python selftests and pkgsrc applications.
(in response to PR pkg/20214)
