fixes for GTK+ 2.10)
2.2.9 changes:
* The crash that occurred if the summary was sorted by a key
which doesn't have visible column was fixed.
* The URI of the about dialog was updated.
2.2.8 changes:
* The bug that possibly lost messages when there was no disk space
has been fixed, so please upgrade if possible.
* The PLAIN authentication method for IMAP4 was supported.
* The verification of encrypted and signed messages with combined
method was supported.
* Oniguruma regex library can be used instead of system's regex API.
* The LDIF import of the addressbook was improved.
* Full PGP signature information is displayed on the main text view.
* All trailing spaces are removed when signing with PGP/MIME instead
of using quoted-printable to improve compatibility with other MUAs.
* A warning is displayed before sending when Bcc is specified at PGP
encryption.
* The bug that possibly lost relatively small messages on receiving
when there was no disk space was fixed.
* The behavior of manual junk filtering was fixed.
* UTF-8 text can be inserted correctly in the compose window.
* The error messages of the failure of displaying message body was
improved.
* Some workarounds for GTK+ 2.10 were made.
* Minor UI fixes were made.
* Other minor bugs were fixed.
* The "-t" archive integrity checking option has been added. This was
requested by several users. cabextract can unpack cabinet files and give
you MD5 checksums of the files inside, without writing the unpacked
files to disk.
* Large files (more than 2 gigabytes) are now correctly searched for
cabinet files.
* A security vulnerability has been fixed. Files compressed with the
Quantum method, using a window size less than 32768 bytes, could cause
cabextract to write beyond the end of the window and cause a
segmentation fault. This fix also permits cabextract to unpack this type
of cabinet file (of which only one has been found in the wild) correctly.
* The unnecessary GNU source mempcpy.c, which caused compilation
failures on several systems, was removed.
* An off-by-one error introduced in 1.1's UTF-8 decoder was fixed. Files
with UTF-8 filenames can now be extracted. The UTF-8 decoder was also
upgraded to support the latest Unicode characer maps.
as tight as possible. Files we don't handle shouldn't be skipped.
- fonts.alias is not created automatically, so don't remove it.
- create fonts.encoding with mkfontdir using -e X11_ENCODINGSDIR.
On platforms not following the X11R6 loayout this might need to
be overriden.
- Fix type1inst calls.
- Modify packages which installed fonts.alias before to actually
include it in the PLIST and bump the revisions accordingly.
- Modify xorg-fonts* packages to use FONTS_DIRS.* to build indices
at run time.
Discussed with wiz and jlam.
the file (~/.face) is on a local filesystem which didn't work because
it didn't recognize NetBSD's way to use statvfs()
(I had to make the check for statvfs a bit weaker, hope that doesn't
break on another platform.)
struct uvmexp -- use uvmexp_sysctl on NetBSD instead
-correct filesystem size calculation: the unit is vfsstat.f_frsize
-bump PKGREVISION
with jmmv's blessing
libgnomedb_entry_builtin_plugins.so.4.0.0 links with libssl, and if
PREFER.openssl=pkgsrc then it should link with Pkgsrc libssl. Without
including security/openssl, it links with /usr/lib/libssl.so instead.
on ssh anyways, but better to make sure). Fixed some
more ssh_*_{en,de}code calls missing necessary casts.
Disabled x11-security extension on x86_64 as it does not
work there (uses xauth instead). Updated pkgrevision.
1. Using ${LIBOSSAUDIO} instead of assuming -lossaudio
2. Disabling sound entirely if ${OSS_TYPE} == none
This should fix the build for every platform but NetBSD (tested on Solaris),
and no change for NetBSD builds.
handling large repositories like NetBSD src. Problem found by riz@,
fix from mainline a:njs/d:2006-09-22T07:01:26, tested by me.
bump rev to nb1
ChangeLog:
2006-09-21 Nathaniel Smith <njs@pobox.com>
* lru_writeback_cache.hh (clean_insert): Correct the cache
overflow logic so as to _actually_ never remove the last element.
Add an invariant to verify this.
(insert_dirty): Add an invariant here too.
