]> git.saurik.com Git - apt.git/log
apt.git
13 years agomerged from lp:~donkult/apt/sid/
Michael Vogt [Tue, 20 Dec 2011 13:56:17 +0000 (14:56 +0100)] 
merged from lp:~donkult/apt/sid/

13 years agomerged from lp:~donkult/apt/sid/
Michael Vogt [Tue, 20 Dec 2011 13:55:19 +0000 (14:55 +0100)] 
merged from lp:~donkult/apt/sid/

13 years agomerged from debian-sid
Michael Vogt [Mon, 19 Dec 2011 13:40:28 +0000 (14:40 +0100)] 
merged from debian-sid

13 years agomerged from lp:~donkult/apt/experimental
Michael Vogt [Mon, 19 Dec 2011 13:22:05 +0000 (14:22 +0100)] 
merged from lp:~donkult/apt/experimental

13 years agomerged from lp:~donkult/apt/experimental
Michael Vogt [Mon, 19 Dec 2011 13:07:51 +0000 (14:07 +0100)] 
merged from lp:~donkult/apt/experimental

13 years agoimplement the fallback method of rred by using the FileFd and the included
David Kalnischkies [Sun, 18 Dec 2011 00:22:04 +0000 (01:22 +0100)] 
implement the fallback method of rred by using the FileFd and the included
ReadLine instead of accessing the files directly with fgets()

13 years agousage of Skipping in pipes can't work, so we ignore-read instead
David Kalnischkies [Sun, 18 Dec 2011 00:21:20 +0000 (01:21 +0100)] 
usage of Skipping in pipes can't work, so we ignore-read instead
Also, read only one char in each step of ReadLine instead of back-"seeking"

13 years agotry to avoid direct usage of .Fd() if possible and do read()s and co
David Kalnischkies [Sat, 17 Dec 2011 22:53:31 +0000 (23:53 +0100)] 
try to avoid direct usage of .Fd() if possible and do read()s and co
on the FileFd instead

13 years agokeep track of where we are in a filedescriptor so we can use it as Tell()
David Kalnischkies [Sat, 17 Dec 2011 16:31:47 +0000 (17:31 +0100)] 
keep track of where we are in a filedescriptor so we can use it as Tell()
information if we are working on a pipe which can't seek

13 years agotry seeking on fds opened with OpenDescriptor before giving up
David Kalnischkies [Thu, 15 Dec 2011 23:04:52 +0000 (00:04 +0100)] 
try seeking on fds opened with OpenDescriptor before giving up

13 years agorefactor compressor calling so that we don't (need to) export ExecCompressor
David Kalnischkies [Thu, 15 Dec 2011 22:38:38 +0000 (23:38 +0100)] 
refactor compressor calling so that we don't (need to) export ExecCompressor
anymore and therefore are also able to drop quiet a bit of duplicated code

13 years agoDutch (Jeroen Schot). Closes: #652230
bubulle@debian.org [Thu, 15 Dec 2011 19:34:20 +0000 (20:34 +0100)] 
Dutch (Jeroen Schot). Closes: #652230

13 years agoatleast libapt should announce to itself that it is clean…
David Kalnischkies [Thu, 15 Dec 2011 08:13:21 +0000 (09:13 +0100)] 
atleast libapt should announce to itself that it is clean…
(and be it if it tries to announce that…)

13 years agoremove the second usage instance of ExecCompressor in ftparchive
David Kalnischkies [Wed, 14 Dec 2011 21:35:03 +0000 (22:35 +0100)] 
remove the second usage instance of ExecCompressor in ftparchive
by again using the FileFd directly

13 years agouse FileFd instead of forking the compression childs by hand
David Kalnischkies [Wed, 14 Dec 2011 21:11:43 +0000 (22:11 +0100)] 
use FileFd instead of forking the compression childs by hand

13 years agoallow Open() and OpenDescriptor() to be called with a Compressor
David Kalnischkies [Tue, 13 Dec 2011 22:48:14 +0000 (23:48 +0100)] 
allow Open() and OpenDescriptor() to be called with a Compressor

13 years agonote to myself: In case you rename something, make sure that you have
David Kalnischkies [Tue, 13 Dec 2011 00:43:28 +0000 (01:43 +0100)] 
note to myself: In case you rename something, make sure that you have
renamed it everywhere as otherwise stuff "magically" starts to fail…

Fixes commit 2209 as the mixture of #define names generates a lovely
compilable but non-functional mixture of gzip usage…

13 years agorevert 2184.1.3: forward declaration instead of headers
David Kalnischkies [Tue, 13 Dec 2011 00:22:38 +0000 (01:22 +0100)] 
revert 2184.1.3: forward declaration instead of headers

The breakage is just to big for now, so guard the change with
#ifndef APT_8_CLEANER_HEADERS and be nice to library users

13 years agorevert 2184.1.2: do not pollute namespace in headers
David Kalnischkies [Mon, 12 Dec 2011 23:54:37 +0000 (00:54 +0100)] 
revert 2184.1.2: do not pollute namespace in headers

The breakage is just to big for now, so guard the change with
#ifndef APT_8_CLEANER_HEADERS and be nice to library users

13 years agoAllow the FileFd to use an external Compressor to uncompress a given file
David Kalnischkies [Mon, 12 Dec 2011 23:17:30 +0000 (00:17 +0100)] 
Allow the FileFd to use an external Compressor to uncompress a given file
internally so that it is exported and can be used like a "normal"
uncompressed file with FileFd

This allows us to hide th zlib usage in the implementation and use gzip
instead if we don't have zlib builtin (the same for other compressors).

The code includes quiet a few FIXME's so while all tests are working
it shouldn't be used just yet outside of libapt as it might break.

13 years ago - add a ReadLine method
David Kalnischkies [Sun, 11 Dec 2011 18:46:59 +0000 (19:46 +0100)] 
  - add a ReadLine method
  - drop the explicit export of gz-compression handling

13 years agouse different compression types in the test now that we support them
David Kalnischkies [Sun, 11 Dec 2011 12:08:21 +0000 (13:08 +0100)] 
use different compression types in the test now that we support them

13 years agoadd a testcase for FindPackages() to better validate that cdrom should work.
David Kalnischkies [Sun, 11 Dec 2011 01:55:20 +0000 (02:55 +0100)] 
add a testcase for FindPackages() to better validate that cdrom should work.
Unfortunately it's hard to do an automated integration test with cd, so we
test this method in isolation which tries to find Indexes and dropping
of duplications with DropRepeats()

13 years ago* apt-pkg/cdrom.cc:
David Kalnischkies [Sun, 11 Dec 2011 00:30:45 +0000 (01:30 +0100)] 
* apt-pkg/cdrom.cc:
  - support InRelease files on cdrom

13 years agouse fileutl exists-functions instead of doing the stat'ing by hand
David Kalnischkies [Sat, 10 Dec 2011 23:58:35 +0000 (00:58 +0100)] 
use fileutl exists-functions instead of doing the stat'ing by hand

13 years agostrip the extension of the translation file before storing it in the list
David Kalnischkies [Sat, 10 Dec 2011 23:41:50 +0000 (00:41 +0100)] 
strip the extension of the translation file before storing it in the list
(regression from compression rewrite; found by Steve McIntyre, thanks!)

13 years ago* apt-pkg/contrib/fileutl.{h,cc}:
David Kalnischkies [Sat, 10 Dec 2011 19:03:49 +0000 (20:03 +0100)] 
* apt-pkg/contrib/fileutl.{h,cc}:
  - implement a ModificationTime method for FileFd

13 years agoenable FileFd to guess the compressor based on the filename if requested or
David Kalnischkies [Sat, 10 Dec 2011 18:31:36 +0000 (19:31 +0100)] 
enable FileFd to guess the compressor based on the filename if requested or
to search for compressed silbings of the given filename and use this guessing
instead of hardcoding Gzip compression

13 years ago* cmdline/apt-config.cc:
David Kalnischkies [Sat, 10 Dec 2011 16:35:54 +0000 (17:35 +0100)] 
* cmdline/apt-config.cc:
  - dump the APT::Compressor settings correctly and completely

13 years ago* doc/po/de.po:
Michael Basse (michael-alpha-unix) [Tue, 6 Dec 2011 15:56:10 +0000 (16:56 +0100)] 
* doc/po/de.po:
  - apply typo-fix from Michael Basse, thanks! (LP: #900770)

13 years agoensure that we exit instead of proceed in execution after the trap
David Kalnischkies [Wed, 30 Nov 2011 15:37:09 +0000 (16:37 +0100)] 
ensure that we exit instead of proceed in execution after the trap
(the execution leads to hard failures anyway)

13 years agosplit up the OpenMode into OpenMode and CompressionMode and
David Kalnischkies [Tue, 29 Nov 2011 11:14:31 +0000 (12:14 +0100)] 
split up the OpenMode into OpenMode and CompressionMode and
provide ReadOnly, WriteOnly and ReadWrite as flags alongside
the additional flags as decompression will be one-way later,
but certain parts really depend on Write* openmodes being
ReadWrite opens, so we will have to fail for those.

13 years agorevert libapt-common addition because of http://lists.debian.org/debian-devel/2011...
Michael Vogt [Mon, 28 Nov 2011 11:07:55 +0000 (12:07 +0100)] 
revert libapt-common addition because of http://lists.debian.org/debian-devel/2011/11/msg00742.html

13 years agouse getCompressors() instead of getCompressorTypes() and use it everywhere
David Kalnischkies [Wed, 23 Nov 2011 23:53:47 +0000 (00:53 +0100)] 
use getCompressors() instead of getCompressorTypes() and use it everywhere
to replace hardcoding of compressiontypes and compressors

13 years agofix another hickup in the changelog
David Kalnischkies [Wed, 23 Nov 2011 18:48:07 +0000 (19:48 +0100)] 
fix another hickup in the changelog

13 years agofactored out the decompressor code in IndexCopy::CopyPackages() and
Steve McIntyre [Wed, 23 Nov 2011 18:34:58 +0000 (19:34 +0100)] 
factored out the decompressor code in IndexCopy::CopyPackages() and
TranslationsCopy::CopyTranslations() into a single common function

13 years agofixup changelog
Michael Vogt [Wed, 23 Nov 2011 10:04:24 +0000 (11:04 +0100)] 
fixup changelog

13 years agomerged from lp:~donkult/apt/experimental
Michael Vogt [Wed, 23 Nov 2011 09:58:44 +0000 (10:58 +0100)] 
merged from lp:~donkult/apt/experimental

13 years agomerged lp:~mvo/apt/convert-to-multiarch-lib-thanks-colin
Michael Vogt [Wed, 23 Nov 2011 09:46:56 +0000 (10:46 +0100)] 
merged lp:~mvo/apt/convert-to-multiarch-lib-thanks-colin

13 years ago* cmdline/apt-get.cc:
David Kalnischkies [Wed, 23 Nov 2011 08:54:17 +0000 (09:54 +0100)] 
* cmdline/apt-get.cc:
  - ignore foreign architectures if we check if a provides has only one
    resolver as it's basically the same for the user, so no need to choose

13 years ago* apt-pkg/depcache.cc:
David Kalnischkies [Tue, 22 Nov 2011 23:49:45 +0000 (00:49 +0100)] 
* apt-pkg/depcache.cc:
  - prefer native providers over foreigns even if the chain is foreign

The code preferred real over virtual packages and based on priorities.
This is changed in so far that a real package from any arch is preferred
over any virtual provider and if priorities doesn't help in choosing the
best provider we choose it based on architectures

13 years agofix the operator++ implementations in the cachesets
David Kalnischkies [Tue, 22 Nov 2011 20:54:32 +0000 (21:54 +0100)] 
fix the operator++ implementations in the cachesets

13 years ago* apt-pkg/cdrom.cc:
Michael Vogt [Tue, 22 Nov 2011 17:59:28 +0000 (18:59 +0100)] 
* apt-pkg/cdrom.cc:
  - use aptconfiguration to get the supported compression types

13 years agomerged from lp:~donkult/apt/sid
Michael Vogt [Tue, 22 Nov 2011 12:25:27 +0000 (13:25 +0100)] 
merged from lp:~donkult/apt/sid

13 years agofix remaining rename
Michael Vogt [Tue, 22 Nov 2011 10:34:01 +0000 (11:34 +0100)] 
fix remaining rename

13 years agorename .install to .install.in when needed
Michael Vogt [Tue, 22 Nov 2011 10:20:33 +0000 (11:20 +0100)] 
rename .install to .install.in when needed

13 years agomerged libapt multiarch enable patch from Colin Watson, many thanks!
Michael Vogt [Tue, 22 Nov 2011 09:41:25 +0000 (10:41 +0100)] 
merged libapt multiarch enable patch from Colin Watson, many thanks!

13 years agoapt-pkg/cdrom.cc: Accept .bz2, .xz files in addition to .gz files (Closes: #649451)
Julian Andres Klode [Mon, 21 Nov 2011 17:53:29 +0000 (18:53 +0100)] 
apt-pkg/cdrom.cc: Accept .bz2, .xz files in addition to .gz files (Closes: #649451)

13 years ago* German manpage translation update
Chris Leick [Sun, 20 Nov 2011 17:11:56 +0000 (18:11 +0100)] 
* German manpage translation update
* doc/*.xml:
  - find and fix a bunch of misspellings

13 years ago* doc/manpage-style.xml:
David Kalnischkies [Sun, 20 Nov 2011 16:12:15 +0000 (17:12 +0100)] 
* doc/manpage-style.xml:
  - put <brackets> around email addresses

13 years ago - provide a {Package,Version}List similar to {Package,Version}Set
David Kalnischkies [Fri, 11 Nov 2011 15:03:40 +0000 (16:03 +0100)] 
  - provide a {Package,Version}List similar to {Package,Version}Set
* cmdline/apt-{get,cache,mark}.cc:
  - use Lists instead of Sets if input order should be preserved for
    commands accepting lists of packages, e.g. policy (Closes: #625960)

13 years agomerged from lp:~donkult/apt/experimental
Michael Vogt [Thu, 10 Nov 2011 15:10:55 +0000 (16:10 +0100)] 
merged from lp:~donkult/apt/experimental

13 years ago* apt-pkg/cacheset.cc:
David Kalnischkies [Wed, 9 Nov 2011 16:22:57 +0000 (17:22 +0100)] 
* apt-pkg/cacheset.cc:
  - make the cachesets real containers which can embedding any container
    to be able to use the same interface regardless of set or list usage

13 years ago* apt-pkg/deb/deblistparser.cc:
David Kalnischkies [Thu, 3 Nov 2011 14:41:14 +0000 (09:41 -0500)] 
* apt-pkg/deb/deblistparser.cc:
  - M-A: foreign packages provide for other archs, too

13 years agodo not enter an endless loop for (essential) pre-dependency loops
David Kalnischkies [Mon, 31 Oct 2011 19:36:05 +0000 (14:36 -0500)] 
do not enter an endless loop for (essential) pre-dependency loops

13 years ago* apt-pkg/contrib/sha2_internal.cc:
David Kalnischkies [Sun, 30 Oct 2011 19:17:09 +0000 (14:17 -0500)] 
* apt-pkg/contrib/sha2_internal.cc:
  - use a pointer-union to peace gcc strict-aliasing warning

13 years agomerge with my debian-sid branch
David Kalnischkies [Sun, 30 Oct 2011 18:48:05 +0000 (13:48 -0500)] 
merge with my debian-sid branch

13 years agoFrench translation update
bubulle@debian.org [Mon, 24 Oct 2011 23:50:03 +0000 (05:20 +0530)] 
French translation update

13 years agomerged from donkult
Michael Vogt [Thu, 20 Oct 2011 09:45:44 +0000 (11:45 +0200)] 
merged from donkult

13 years agomerged from donkult
Michael Vogt [Thu, 20 Oct 2011 09:44:44 +0000 (11:44 +0200)] 
merged from donkult

13 years ago* algorithms.cc:
David Kalnischkies [Mon, 17 Oct 2011 14:41:54 +0000 (16:41 +0200)] 
* algorithms.cc:
  - show a debug why a package was kept by ResolveByKeep()

13 years ago* apt-pkg/packagemanager.cc:
David Kalnischkies [Mon, 17 Oct 2011 09:22:45 +0000 (11:22 +0200)] 
* apt-pkg/packagemanager.cc:
  - do not fail on unpacked packages in SmartUnPack, just don't
    shedule them for unpack, but do all checks and configure them

13 years agocreate "real" testcases out of the tests for chris
David Kalnischkies [Mon, 17 Oct 2011 09:21:12 +0000 (11:21 +0200)] 
create "real" testcases out of the tests for chris

13 years agoreleasing version 0.8.16~exp7
Michael Vogt [Fri, 14 Oct 2011 12:05:03 +0000 (14:05 +0200)] 
releasing version 0.8.16~exp7

13 years agomerged lp:~donkult/apt/experimental
Michael Vogt [Fri, 14 Oct 2011 11:55:50 +0000 (13:55 +0200)] 
merged lp:~donkult/apt/experimental

13 years agoreleasing version 0.8.15.9 0.8.15.9
Michael Vogt [Fri, 14 Oct 2011 11:27:49 +0000 (13:27 +0200)] 
releasing version 0.8.15.9

13 years agomerged from lp:~donkult/apt/sid
Michael Vogt [Fri, 14 Oct 2011 09:54:12 +0000 (11:54 +0200)] 
merged from lp:~donkult/apt/sid

13 years ago* apt-pkg/pkgcache.cc:
David Kalnischkies [Wed, 12 Oct 2011 20:28:46 +0000 (22:28 +0200)] 
* apt-pkg/pkgcache.cc:
  - always prefer "en" over "" for "en"-language regardless of cache-order

13 years agoadd implicit dependencies needed for Multi-Arch at the time a Version
David Kalnischkies [Wed, 12 Oct 2011 18:16:02 +0000 (20:16 +0200)] 
add implicit dependencies needed for Multi-Arch at the time a Version
struct is created and not at the end of the cache generation

This allows us to be independent from the configured architectures for
these kind of conflicts, we get natural progress for free and
only the needed dependencies are in th respective binary cache.

13 years agouse one string to construct the error message instead of using multiple
David Kalnischkies [Wed, 12 Oct 2011 17:34:06 +0000 (19:34 +0200)] 
use one string to construct the error message instead of using multiple
just with different debugging information at the end

13 years agoa version can have only a single md5 for descriptions, so we can optimize
David Kalnischkies [Wed, 12 Oct 2011 13:47:56 +0000 (15:47 +0200)] 
a version can have only a single md5 for descriptions, so we can optimize
the merging with this knowledge a bit and by correctly sharing the lists
we only need to have a single description list for possibly many different
versions. This also means that description translations are shared between
different sources

13 years agoshare description list between "same" versions (LP: #868977)
David Kalnischkies [Tue, 11 Oct 2011 19:10:31 +0000 (21:10 +0200)] 
share description list between "same" versions (LP: #868977)

13 years ago* apt-pkg/pkgcachegen.cc:
David Kalnischkies [Tue, 11 Oct 2011 16:34:21 +0000 (18:34 +0200)] 
* apt-pkg/pkgcachegen.cc:
  - refactor MergeList by creating -Group, -Package and -Version specialist

13 years agorun apt-get download in quiet mode as it messes with output otherwise
David Kalnischkies [Tue, 11 Oct 2011 12:44:05 +0000 (14:44 +0200)] 
run apt-get download in quiet mode as it messes with output otherwise

13 years agorun integration tests on my new amd64 box without failures
David Kalnischkies [Tue, 11 Oct 2011 12:37:03 +0000 (14:37 +0200)] 
run integration tests on my new amd64 box without failures

13 years agofix compiler warning about signed comparision on amd64, too
David Kalnischkies [Tue, 11 Oct 2011 11:05:40 +0000 (13:05 +0200)] 
fix compiler warning about signed comparision on amd64, too

13 years ago* doc/apt.conf.5.xml:
David Kalnischkies [Wed, 5 Oct 2011 21:09:09 +0000 (23:09 +0200)] 
* doc/apt.conf.5.xml:
  - apply spelling fix by Kevin Lyda, thanks! (Closes: #644104)

13 years agomerge with current debian apt/sid
David Kalnischkies [Wed, 5 Oct 2011 21:06:26 +0000 (23:06 +0200)] 
merge with current debian apt/sid

13 years agocherrypick from my apt/experimental branch
David Kalnischkies [Wed, 5 Oct 2011 21:00:47 +0000 (23:00 +0200)] 
cherrypick from my apt/experimental branch

* apt-pkg/deb/debmetaindex.cc:
  - none is a separator, not a language: no need for Index (Closes: #624218)
* apt-pkg/aptconfiguration.cc:
  - do not builtin languages only if none is forced (Closes: #643787)

13 years ago* apt-pkg/deb/debmetaindex.cc:
David Kalnischkies [Wed, 5 Oct 2011 20:45:22 +0000 (22:45 +0200)] 
* apt-pkg/deb/debmetaindex.cc:
  - none is a separator, not a language: no need for Index (Closes: #624218)
* apt-pkg/aptconfiguration.cc:
  - do not builtin languages only if none is forced (Closes: #643787)

13 years agomerge with current debian apt/experimental
David Kalnischkies [Wed, 5 Oct 2011 16:43:00 +0000 (18:43 +0200)] 
merge with current debian apt/experimental

13 years ago* apt-pkg/contrib/configuration.cc:
Michael Vogt [Wed, 5 Oct 2011 16:14:38 +0000 (18:14 +0200)] 
* apt-pkg/contrib/configuration.cc:
  - fix double delete (LP: #848907)
  - ignore only the invalid regexp instead of all options
* apt-pkg/acquire-item.h, apt-pkg/deb/debmetaindex.cc:
  - fix fetching language information by adding OptionalSubIndexTarget
* methods/https.cc:
  - cleanup broken downloads properly

* ftparchive/cachedb.cc:
  - fix buffersize in bytes2hex
* apt-pkg/deb/deblistparser.cc:
  - fix crash when the dynamic mmap needs to be grown in
    LoadReleaseInfo (LP: #854090)

13 years ago* apt-pkg/policy.cc:
David Kalnischkies [Mon, 3 Oct 2011 22:14:38 +0000 (00:14 +0200)] 
* apt-pkg/policy.cc:
  - accept generic release pin expressions again in -t (Closes: #644166)

13 years agomerged fix from donkult
Michael Vogt [Mon, 26 Sep 2011 11:30:19 +0000 (13:30 +0200)] 
merged fix from donkult

13 years ago* ftparchive/cachedb.cc:
Michael Vogt [Sat, 24 Sep 2011 08:49:49 +0000 (10:49 +0200)] 
* ftparchive/cachedb.cc:
  - fix buffersize in bytes2hex

* ftparchive/cachedb.cc:
  - make buffer fully dynamic (thanks to Colin Watson)

13 years agoSymbol file update
Michael Vogt [Sat, 24 Sep 2011 08:46:03 +0000 (10:46 +0200)] 
Symbol file update

13 years agoftparchive/cachedb.cc: rewrite to fix the fixed length field
Michael Vogt [Thu, 22 Sep 2011 13:15:30 +0000 (15:15 +0200)] 
ftparchive/cachedb.cc: rewrite to fix the fixed length field

13 years ago* ftparchive/cachedb.cc:
Michael Vogt [Thu, 22 Sep 2011 12:17:13 +0000 (14:17 +0200)] 
* ftparchive/cachedb.cc:
  - fix buffersize in bytes2hex

13 years ago* apt-pkg/init.cc:
David Kalnischkies [Wed, 21 Sep 2011 17:37:31 +0000 (19:37 +0200)] 
* apt-pkg/init.cc:
  - silently ignore *.orig and *.save files by default

13 years ago* apt-pkg/deb/debsrcrecords.cc:
David Kalnischkies [Wed, 21 Sep 2011 17:31:03 +0000 (19:31 +0200)] 
* apt-pkg/deb/debsrcrecords.cc:
  - remove the limit of 400 Binaries for a source package (Closes: #622110)

13 years agoconvert a few for-loop char finds to proper strchr and memchr
David Kalnischkies [Wed, 21 Sep 2011 16:42:08 +0000 (18:42 +0200)] 
convert a few for-loop char finds to proper strchr and memchr

13 years ago* methods/https.cc:
Michael Vogt [Tue, 20 Sep 2011 16:21:21 +0000 (18:21 +0200)] 
* methods/https.cc:
  - cleanup broken downloads properly

13 years ago* apt-pkg/deb/dpkgpm.cc:
David Kalnischkies [Tue, 20 Sep 2011 12:30:31 +0000 (14:30 +0200)] 
* apt-pkg/deb/dpkgpm.cc:
  - use std::vector instead of fixed size arrays to store args and
    multiarch-packagename strings
  - load the dpkg base arguments only one time and reuse them later
* cmdline/apt-get.cc:
  - follow Provides in the evaluation of saving candidates, too, for
    statisfying garbage package dependencies (Closes: #640590)
* apt-pkg/algorithms.cc:
  - if a package is garbage, don't try to save it with FixByInstall

13 years ago* apt-pkg/algorithms.cc:
David Kalnischkies [Tue, 20 Sep 2011 12:21:23 +0000 (14:21 +0200)] 
* apt-pkg/algorithms.cc:
  - if a package is garbage, don't try to save it with FixByInstall

13 years ago* cmdline/apt-get.cc:
David Kalnischkies [Tue, 20 Sep 2011 11:27:53 +0000 (13:27 +0200)] 
* cmdline/apt-get.cc:
  - follow Provides in the evaluation of saving candidates, too, for
    statisfying garbage package dependencies (Closes: #640590)

13 years agoload the dpkg base arguments only one time and reuse them later
David Kalnischkies [Tue, 20 Sep 2011 09:54:15 +0000 (11:54 +0200)] 
load the dpkg base arguments only one time and reuse them later

13 years ago* apt-pkg/deb/dpkgpm.cc:
David Kalnischkies [Tue, 20 Sep 2011 09:34:37 +0000 (11:34 +0200)] 
* apt-pkg/deb/dpkgpm.cc:
  - use std::vector instead of fixed size arrays to store args and
    multiarch-packagename strings

13 years agofix foldmarker in algorithms.h
David Kalnischkies [Mon, 19 Sep 2011 19:05:00 +0000 (21:05 +0200)] 
fix foldmarker in algorithms.h

13 years agoremove old APT_COMPATIBILITY ifdef's
David Kalnischkies [Mon, 19 Sep 2011 17:27:07 +0000 (19:27 +0200)] 
remove old APT_COMPATIBILITY ifdef's

13 years agouse forward declaration in headers if possible instead of includes
David Kalnischkies [Mon, 19 Sep 2011 17:14:19 +0000 (19:14 +0200)] 
use forward declaration in headers if possible instead of includes