summaryrefslogtreecommitdiffstats
path: root/lib/libalpm
AgeCommit message (Collapse)AuthorFilesLines
2007-11-15Move alpm_splitdep usage to db_readDan McGee3-41/+22
Holy inefficient batman! For a pacman -Qt operation (when we are using compute_requiredby and not database entries), splitdep was being called ~1.3 million times on my local database. By splitting when we read the DB, we drop this number to around 1700 and save a LOT of time in doing so (a 5x increase in pacman -Qt speed here). Note that the depends alpm_list_t in the package struct is no longer a string list, but a list of pmdepent_t objects. Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-15Remove REQUIREDBY usage from libalpmDan McGee11-174/+18
Instead of using the often-busted REQUIREDBY entries in the pacman database, compute them each time they are required. This should help many things: 1. Simplify the codebase 2. Prevent future database corruption 3. Ensure when we do use requiredby, it is always correct 4. Shrink the pmpkg_t memory overhead Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-15alpm_list_add == alpm_list_add_lastNagy Gabor4-33/+24
It's time to define that alpm_list_add(list, foo) adds 'foo' to the end of 'list' and returns with 'list', because: 1. list is a list, not a set. 2. sortbydeps _needs_ an alpm_list_add definition to work properly. As a first step, I used this definition in recursedeps. Signed-off-by: Nagy Gabor <ngaba@bibl.u-szeged.hu> [Dan: punctuation cleanup in commit message and code comments, added comment to alpm_list_add] Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-13Enforce const correctness on dep functions and rewrite alpm_dep_get_stringDan McGee2-34/+36
Add some const specifiers to the dep functions that can have them. In addition, rewrite alpm_dep_get_string to use snprintf and cover all of the bases (operators). Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-13Fix display of -Qip output when a package file is givenDan McGee1-0/+4
Too many fields were being shown on -Qip output, and sizes were not always correct (-Qi and -Qip output on the same package did not agree). Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-11Fix memleak with new alpm_list_reverse usageDan McGee1-2/+4
Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-11Add a horrible little hack to get symlink001.py to pass againDan McGee1-5/+14
This really doesn't give us any regressions in behavior, so it is safe to do although quite ugly. Tell the conflict checking code to ignore symlinks to dirs so that they are not seen as conflicts. Hopefully this entire commit will get factored out soon enough. Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-11Ensure list tail pointer is updated when we remove tail nodeDan McGee1-2/+10
Commit 2ee90ddae23dd86c68223c0d6c49f0b92d62429d did a special check to see if we were removing the head node, but not the tail node. Add a special case for the tail node to ensure all relevant pointers get updated. Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-11Incorrect usage of alpm_db_whatprovides in sync.cNagy Gabor1-4/+4
The old code thought that alpm_db_whatprovides returns with a list of strings (package names). Signed-off-by: Nagy Gabor <ngaba@bibl.u-szeged.hu> Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-11libalpm: fix lstat wrapper to actually use newpathDan McGee1-1/+1
Commit b55abdce7aebb142ce79da3aa3645afe7693a3c4 introduced an lstat wrapper function that never dereferences paths with a trailing slash, but still called lstat on path instead of newpath. Oops! Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-11Remove unused and broken alpm_list_remove_node functionDan McGee2-30/+0
Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-11Implement TotalDownload option.Nathan Jones5-16/+55
Setting this option will change the download progress to show the amount downloaded, download rate, ETA, and download percent of the entire download list rather than per each individual file. The progress bar is still based on the completion of the current file regardless if the TotalDownload option is set. This closes FS#7205. Signed-off-by: Nathan Jones <nathanj@insightbb.com> Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-11Implement IgnoreGroup.Nathan Jones5-7/+37
This option acts as if IgnorePkg was set on each package in the group. This closes FS#1592. Signed-off-by: Nathan Jones <nathanj@insightbb.com> Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-11Add IgnoreGroup and --ignoregroup option.Nathan Jones3-0/+26
This will be used in the next commit. Signed-off-by: Nathan Jones <nathanj@insightbb.com> Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-09Fix some issues with localized dates/epoch usageDan McGee2-0/+6
Commit 47622eef4dd8fd86a0aa0e3ebdb7b33f7c9d6804 introduced localized times in the metadata by way of storing the UNIX epoch value instead of a hard coded date string. However, it missed a few things: * If we weren't in the C/POSIX/en_US locale, the date parsing would fail as it tried to use the abbreviations of the locale being used. Fix this by switching the LC_TIME value before we parse a date. * We used ctime to print the date value, which is always the C locale string. Instead, use strftime to print a localized date string. Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-09Update Italian TranslationGiovanni Scafora1-334/+329
Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-08Update en_GB translationJeff Bailes1-329/+368
Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-07Add STRDUP macro to mirror MALLOC/CALLOCAaron Griffin1-0/+2
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
2007-11-07Maintain list tail pointers in the head nodeAaron Griffin2-16/+46
List head nodes contain null 'prev' pointer, which we can (ab)use to maintain a back reference to the tail pointer of the list. While list additions are not _significantly_ improved, they are still sped up. Original $ time pacman -Qo /usr/bin/wtpt /usr/bin/wtpt is owned by lcms 1.17-2 real 0m3.623s user 0m1.883s sys 0m1.473s New $ time pacman -Qo /usr/bin/wtpt /usr/bin/wtpt is owned by lcms 1.17-2 real 0m2.006s user 0m0.263s sys 0m1.627s Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
2007-11-06libalpm/db.c: change two warnings to errorsDan McGee1-2/+2
These two warnings really indicate failure, so the message they print should do so as well. Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-05Remove the newline automatically added by alpm_logaction.Chantry Xavier3-17/+16
This way, _alpm_logaction behaves like _alpm_log, and gives more control. Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
2007-11-05libalpm: use an lstat wrapper so we never dereference dir symlinksDan McGee5-6/+32
Linux lstat follows POSIX standards and dereferences a symlink pointing to a directory if there is a trailing slash. For purposes of libalpm, we don't want this so make a lstat wrapper that suppresses this behavior. Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-04libalpm: open the logstream on demandDan McGee2-23/+25
Don't open a stream to the logfile until necessary. This will allow us to catch any errors in opening the logfile instead of ignorning them. Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-04Clean up usage of extern variablesDan McGee4-3/+4
Instead of declaring the extern variable in every *.c file, include it in the header file that makes sense. This means handle.h for the handle, and conf.h for the pacman side config object. Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-04Readd scriptlet logging that got lost in an earlier commitDan McGee5-9/+25
I broke scriptlet logging with ad691001e20272b794d2ed574b556f520e3555c0. Readd more or less what was there before, although it still needs a lot of work including hopefully rewriting it to a new event subsystem and having it log to a seperate file. Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-04Add some more autoconf macros to filter our CFLAGS usageDan McGee1-1/+8
Hopefully these new autoconf macros, with a little magic, will allow us to compile with any compiler and still choose the options we have available to us. Tested locally with gcc 4.2.2 and gcc 3.4.6; the latter doesn't support two of the items we previously had hardcoded in our CFLAGS. Signed-off-by: Dan McGee <dan@archlinux.org>
2007-11-04Add a little const correctness fix to alpm_listDan McGee1-3/+4
Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-30When printing size_t, use %zdDan McGee3-5/+5
%d was used, which worked for Linux and FreeBSD. Not so for Darwin. The warning was probably spat out when compiling on x68_64 as well, but no developers use this architecture as their primary one. Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-30libalpm/handle.c: make realpath() call portableDan McGee1-4/+3
BSD didn't support the NULL second argument GNU extension, so do it the old fashioned way. Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-30Copy pmdelta_t objects in _alpm_pkg_dup()Dan McGee1-0/+2
This was forgotten in the original set of patches. CC: Nathan Jones <nathanj@insightbb.com> Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-30Make general list copy functionDan McGee2-2/+25
Package dup needs to copy all members. Nathan had his implementation, but I generalized it to this new alpm_list function (and will use it in the next commit). CC: Nathan Jones <nathanj@insightbb.com> Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-29libalpm/delta: add const to most methodsDan McGee3-9/+9
Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-29libalpm: introduce MALLOC and CALLOC macrosDan McGee11-78/+30
These macros take the place of the common 4 or 5 line blocks of code we had in most places that called malloc or calloc. This should reduce some code duplication and make memory allocation more standard in libalpm. Highlights: * Note that the MALLOC macro actually uses calloc, this is just for safety so that memory is initialized to 0. This can be easily changed in one place. * One malloc call was completely eliminated- it made more sense to do it on the stack. * The use of RET_ERR in public functions (mainly the alpm_*_new functions) was standardized, this makes sense so pm_errno is set. Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-29Add some more debugging output to dep checking codeDan McGee2-6/+11
Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-27Clean up LDADD usageDan McGee1-1/+0
We had a lot of unnecessary overstatements of libraries to include on linking, and autoconf/automake takes care of this for us. This also helps some compilation issues on other platforms. Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-27Explicitly cast void* to char* during printf callDan McGee1-4/+4
If we don't have an explicit cast, make fails during -Wall -Werror. Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-27dependency error message fix in libalpm/remove.cNagy Gabor1-1/+1
Normally you must never see that error message. Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-27Dependency error/log messages fixNagy Gabor2-8/+14
The old code used only the depend.name in messages, which might have not been informative. The new code uses the whole dependency string in %DEPENDS% format. (Dan: slight English clarification in one of the messages) Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-25Add alpm_dep_get_string methodNagy Gabor2-0/+35
Public alpm_dep_get_string function is introduced, which converts a pmdepend_t structure to printable string in %DEPENDS% format. This function is now used in pacman to print dependency error messages. Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-24_alpm_depmiss_isin fixNagy Gabor1-2/+5
The old code used memcmp, which is not good for comparing strings: "pkgname"'\0''\0' should be equal to "pkgname"'\0''a' for example. The new code uses strcmp. Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-23Print and typecast time_t correctlyDan McGee1-4/+3
The recommended C99 way to print the value of a time_t is to cast it to uintmax_t or intmax_t. Do this to ensure compatability with all platforms. Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-23Ensure all localization stuff is correctly guardedDan McGee2-0/+5
Anything dealing with libintl and localization should be correctly guarded inside an ENABLE_NLS block on both the pacman and libalpm sides. Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-23Fix broken or missing includesDan McGee2-8/+9
Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-23Fix invalid static scoping of strverscmpDan McGee2-2/+2
Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-20Add download size to target list.Nathan Jones2-0/+36
This displays the download size, taking into account delta files and cached files. This closes FS#4182. Signed-off-by: Nathan Jones <nathanj@insightbb.com> Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-20Download delta files if UseDelta is set.Nathan Jones6-3/+271
Delta files will be used if the size is smaller than a percent (MAX_DELTA_RATIO) of the package size. Signed-off-by: Nathan Jones <nathanj@insightbb.com> Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-20Refactor md5sum checking.Nathan Jones1-44/+89
This will allow deltas and packages to share the md5sum checking code. Signed-off-by: Nathan Jones <nathanj@insightbb.com> Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-20Add pmdelta_t structure and functions to libalpm.Nathan Jones11-1/+380
Signed-off-by: Nathan Jones <nathanj@insightbb.com> Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-18libalpm/package.c : fix requiredby with multiple providers.Chantry Xavier1-5/+2
The code didn't match the following comment : "A depends on B through n depends <=> A listed in B's requiredby n times" It stopped at n=1 with a break. I was surprised to see this case happens in real, that's how I noticed the bug: wine depends on both freeglut and glut, while freeglut provides glut. So when installing wine, the update_depends function listed wine twice in freeglut's requiredby. But the compute_requiredby function (used when installing freeglut, and used by testdb) listed wine only once in freeglut's requiredby. That made testdb unhappy. Signed-off-by: Chantry Xavier <shiningxc@gmail.com> Signed-off-by: Dan McGee <dan@archlinux.org>
2007-10-15libalpm/trans.c : fix a recently introduced breakage in scriptlets handling.Chantry Xavier1-6/+10
Commit 4853a4aad97fe36f9237ffb7356201adab507a1c used the tmpdir variable for checking the existence of /bin/sh, without resetting it. This caused /bin/sh to be deleted during the cleanup part, as soon as a scriptlet other than pre_upgrade or pre_install was executed. For example, on the first post_upgrade during a -Su. I introduced two variables : clean_tmpdir and restore_cwd, for deciding what should be done in the cleanup part. Signed-off-by: Chantry Xavier <shiningxc@gmail.com>