summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2007-02-06fix makefiles for PKGBUILD.8 -> PKGBUILD.5Dan McGee
2007-02-06Minor fixups for the moment, still partially completeAaron Griffin
2007-02-06Swapped out PKGBUILD.8 for PKGBUILD.5 from Dan McGee - original PKGBUILD.8 ↵Aaron Griffin
had too many Frugal specific changes.
2007-02-06Copy-pasta errors. 8)Aaron Griffin
2007-02-06Attempt to unify the -S and -Su version comparissons for when a package is newerAaron Griffin
See FS#5388
2007-02-06* Added some backwards compatability stuff to makepkg, along with helpfulDan McGee
warnings so people aren't caught off guard in the future. * Added a temp fix for a bug in file 4.19 where bziped tarballs were returning type "application/empty".
2007-02-04oops, ENVIRONMENT -> BUILDENV in commentsDan McGee
2007-02-04* Updated makepkg to use a new form of options. These two arrays now replaceDan McGee
most of the former options in makepkg.conf, and are described in detail there: BUILDENV=(fakeroot !distcc color !ccache) OPTIONS=(strip !docs !libtool emptydirs) Note that PKGBUILD specified options will always have final say. They will now be specified as they are above, except using lowercase 'options' array. * makepkg was fixed so options actually work as expected.
2007-02-04Added the --upgrades option for -Qu (--query --upgrades)Aaron Griffin
2007-02-04Implemented a crappy version of -Qu (query upgrades). This simply outputs theAaron Griffin
packages to be upgraded in a -Su operation. Much of the code is duplicated from sync.c. TODO: move the implementation to upgrades.c, and reimplement the sync_sysupgrade function in terms of this: trans->packages = alpm_get_upgrades();
2007-02-04* Added archive verification when loading package metadata for -u and -AAaron Griffin
operations (now aborts on a corrupt archive) * Fixed the pm_fprintf newline error that was plaguing us. It seems a line resetting 'neednl' was removed a while back (by me). This causes all the output errors we've been seeing
2007-02-04* unified the progress bars (fill_progress function)Aaron Griffin
* fixed progress output (needs an fflush to move cursor properly) * broke display_targets function out, to display a list of syncpkgs in preparation for a -Qu option * added get_update_time function to deal with progress functions that shouldn't update too fast due to output redraw speeds
2007-02-03Add newlines to error messagesDan McGee
2007-02-03No longer necessary to explicitly specify SRCDEST, leave it up to user toDan McGee
decide.
2007-02-03Fix problem with PKGDEST variable not being used correctly.Dan McGee
2007-02-03Updated news file to be more relevant with previous release of pacman;Dan McGee
ommitted many of the bugfixes in 'unreleased' versions, at least on the Arch Linux end.
2007-02-01* fixed clean target: no error if repository files do not existJürgen Hötzel
2007-02-01* removal of swig generated bindingsJürgen Hötzel
2007-02-01* removal of swig generated bindingsJürgen Hötzel
2007-02-01* Tried to clean up newline display a bit in the frontend.Dan McGee
* Removed useless buildstring function from util.h; replaced all calls of it with list_display. * Made list_display output 2 spaces instead of 1 between each item.
2007-02-01Added mirror in Finland to mirrorlistDan McGee
2007-02-01* Minor output fix (move colon so it's not part of the URL)Aaron Griffin
* Added ldconfig on --sync operations too
2007-02-01Loading package data from a file was ALWAYS generating the filelist, instead ofAaron Griffin
using the in-package one. This is now fixed, and an autogenerated one is used as a last resort. This fixes the bug where /.CHANGELOG showed up in -Ql.
2007-02-01It appears I misunderstood what this bash-ism didAaron Griffin
2007-02-01Cleaned up some of the metadata controlAaron Griffin
2007-02-01-Ql needs the 'root' prepended to each fileAaron Griffin
2007-02-01This removes the warning about skipping dependency checking when the user isDan McGee
either generating the md5sums, repackaging, or just downloading and extracting. The patch is below. ~ Jamie / yankees26 Signed-off-by: James Rosten <seinfeld90@gmail.com>
2007-02-01install scriptlet is _always_ .INSTALLAaron Griffin
2007-02-01Move filelist creation ABOVE the other metafiles to keep them out of the listAaron Griffin
2007-02-01* !strcmp reads funny (as "not string compare" - switch those to == 0, which hasAaron Griffin
a positive connotation. * Added a nice TODO
2007-02-01removal of unnecessary '\' in integrity check generationDan McGee
2007-01-31Removed the extra newline based on general opinionAaron Griffin
2007-01-31* Forgot the int->enum header checkinAaron Griffin
* Removed the dep-sorting TODO item - switch the sorting to a topological sort later (I'll let Dan do the CS stuff, heh)
2007-01-31* Make -Qo work on multiple files againAaron Griffin
* no real sense in query_fileowner returning an int
2007-01-31Switched an int to an enumAaron Griffin
2007-01-31just an updateDan McGee
2007-01-31TODO changes 8)Aaron Griffin
2007-01-31Fix corrupt DB entry handling when loading package entries.Aaron Griffin
* scan loops no longer abort on corrupt entries * reloading a cache package (and discovering it corrupt) no longer prints 2 messages and uses the pre-build "remove from package cache" function NOTE: The TODOs in there are important for later w.r.t. cleaning up corrupt DB entries but there are some logical complexities with doing so, so I'm holding off for now.
2007-01-31Debug logging changes:Aaron Griffin
* The --debug params were goofy. New setup allows --debug without params, --debug=<level> where level 1=debug output, 2=debug and download output, 3=debug, download, and function tracing output. This seems more sane to me. * Removed PM_LOG_FLOW1 and PM_LOG_FLOW2. They were just confusing. When adding new functions, it is near impossible to determin if your output should be "flow1" or "flow2" without tracking all the way up the call chain. Rarely would one ever say "ok, lets just show "flow2" output. These have both been replaced with PM_LOG_DEBUG * Removed the need for the root parameter on alpm_initialize. it is now defaulted to PM_ROOT just like dbpath and cachedir. This allows alpm to be initialized BEFORE option parsing in the front end, saving us some duplicate variables in the frontend. * Cleaned up front end variables due to early alpm_initialize call.
2007-01-31More uninitialized stuff... I believe these were removed at some point. ShouldAaron Griffin
be fixed.
2007-01-31slight line spacing/indent updatesDan McGee
2007-01-31* Refactored some functions to clean up variable declaration.Dan McGee
* Reduced magic number usage by fill_progress function (new). * Some switch indent fixing. * Remove use of log10 call.
2007-01-31* Updated repo-add script to remove same package, different version whenDan McGee
adding a package to a database. Also added commenting. :)
2007-01-30* doxygen check cleanupJürgen Hötzel
2007-01-30Fixing line spacings of download/install size messages.Dan McGee
2007-01-30Well that silly memory clobber was plain old "uninitialized data" - fixedAaron Griffin
2007-01-30TODO changesAaron Griffin
2007-01-30Turn debug off by defaultAaron Griffin
2007-01-30K. Piche <kevin.piche@cgi.com>Aaron Griffin
* ALPM_LOG_FUNCTION macro and all the great work to add this macro everywhere
2007-01-30K. Piche <kpiche@rogers.com>Aaron Griffin
* gcc visiblity changes Also modified _alpm_versioncmp -> alpm_versioncmp (public function) as per K. Piche's suggestions