- Pacman-dev - lists.archlinux.org

[pacman-dev] Check variables for array
by Alexpux 05 Nov '15

05 Nov '15

Hi all! Checking variables in PKGBUILD with grep cause troubles. For example, if we have configure option like «—with-arch=…» grep give us false positive result and makepkg exit with error that «arch should be an array». This is due next code in pacman/scripts/libmakepkg/lint_pkgbuild/variable.sh.in: local array=(arch backup checkdepends groups license noextract options validpgpkeys) local arch_array=(conflicts depends makedepends md5sums optdepends provides replaces sha1sums sha256sums sha384sums sha512sums source) local string=(changelog epoch install pkgdesc pkgrel pkgver url) local i for i in ${array[@]}; do if grep -w -q -e "$i=[^(]" -e "$i+=[^(]" "$BUILDSCRIPT"; then error "$(gettext "%s should be an array")" "$i" ret=1 fi done Regards, Alexey.

1 0

[pacman-dev] Signing the database again
by Remi Gacogne 05 Nov '15

05 Nov '15

Hi, I have been thinking about the DB signing feature recently, and I have dug up old emails regarding this topic. If I understand correctly: - we would like database signing to prevent an attacker from messing with the information contained in the database, being at rest on a mirror or in flight. Packages being signed protects us from most issues, but it is still possible for example to prevent a specific package from being upgraded by altering its entry in the database ; - requiring TU and devs to sign the database when publishing a package is not easy and ; - we don't want to have a package-signing key online if we can prevent it, so ideally the key used to sign the database should not be able to sign packages ; - in addition to that, it would be nice if package-signing keys would not be able to sign the database. I believe a simple solution would be to use a separate WoT for database signing, distinct from the one use for package-signing. This would require the generation of new, separate master keys for verifying db-signing only keys. The master keys would be kept offline while a db-signing key would be kept online on an official Arch server and used by repo-add to sign the database upon modifications. As I am not a TU nor a developer, I am not familiar with the exact process used to publish packages. I have discussed this a bit with Levente, but this is clearly the fuzziest part for me, so please let me know if this is non-sense. This may require the use of a set-uid or set-gid binary so that different users are allowed to use repo-add without giving them access to the db-signing key. Setting up database-signing this way, a compromised mirror couldn't alter the database or the packages without being detected. If the db-signing key were to be compromised, this would not be enough to be able to alter the content of packages, and the key could be revoked and replaced quite easily, so we would not be in a worse position than without db signing. Most of the code required for this to work is already in pacman. We would still need to patch pacman to be able to use a separate gnupg directory for the database WoT, distinct from the packages WoT. I have attached a PoC patch to illustrate this idea, but it clearly needs more work. Any thoughts?

3 2

[pacman-dev] [PATCH] Refactor lint_source to work with earlier versions of Bash
by Aaron Campbell 04 Nov '15

04 Nov '15

Negative subscripts to indexed arrays are not supported before 4.2. However, since substring expansion works on arrays, we can specify an offset of -1 to be taken relative to one greater than the maximum index of the specified array (see Parameter Expansion section of the bash man page). This works with both Bash 4.1 and 4.2, and 4.1 is already the oldest supported by pacman. Signed-off-by: Aaron Campbell <aaron(a)monkey.org> --- scripts/libmakepkg/lint_pkgbuild/source.sh.in | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/libmakepkg/lint_pkgbuild/source.sh.in b/scripts/libmakepkg/lint_pkgbuild/source.sh.in index c327299..b5f4586 100644 --- a/scripts/libmakepkg/lint_pkgbuild/source.sh.in +++ b/scripts/libmakepkg/lint_pkgbuild/source.sh.in @@ -32,7 +32,7 @@ lint_pkgbuild_functions+=('lint_source') lint_source() { local idx=("${!source[@]}") - if (( ${#source[*]} > 0 && (idx[-1] + 1) != ${#source[*]} )); then + if (( ${#source[*]} > 0 && (${idx[@]: -1} + 1) != ${#source[*]} )); then error "$(gettext "Sparse arrays are not allowed for source")" return 1 fi -- 2.5.0

1 0

[pacman-dev] [PATCH] Bump minimum required bash version to 4.2
by Aaron Campbell 04 Nov '15

04 Nov '15

4.1 does not have support for negative subscripts to indexed arrays (a feature that makepkg depends on). >From the bash CHANGES log list of changes between 4.1-release and 4.2-alpha: Negative subscripts to indexed arrays, previously errors, now are treated as offsets from the maximum assigned index + 1. Fixes "/usr/bin/makepkg: line 2774: idx: bad array subscript" seen w/ 4.1. Signed-off-by: Aaron Campbell <aaron(a)monkey.org> --- configure.ac | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/configure.ac b/configure.ac index 359d61b..a8d339c 100644 --- a/configure.ac +++ b/configure.ac @@ -183,18 +183,18 @@ AC_DEFUN([AX_PROG_PERL_VERSION], AX_PROG_PERL_VERSION([5.10.1], [], [AC_MSG_ERROR([perl is too old])]) AS_IF([test "x$BASH_SHELL" = "xfalse"], - AC_MSG_WARN([*** bash >= 4.1.0 is required for pacman scripts]), + AC_MSG_WARN([*** bash >= 4.2.0 is required for pacman scripts]), [bash_version_major=`$BASH_SHELL -c 'echo "${BASH_VERSINFO[[0]]}"'` bash_version_minor=`$BASH_SHELL -c 'echo "${BASH_VERSINFO[[1]]}"'` ok=yes if test "$bash_version_major" -lt 4; then ok=no fi - if test "$bash_version_major" -eq 4 && test "$bash_version_minor" -lt 1; then + if test "$bash_version_major" -eq 4 && test "$bash_version_minor" -lt 2; then ok=no fi if test "$ok" = "no"; then - AC_MSG_ERROR([*** bash >= 4.1.0 is required for pacman scripts]) + AC_MSG_ERROR([*** bash >= 4.2.0 is required for pacman scripts]) fi unset bash_version_major bash_version_minor ok]) -- 2.5.0

2 2

[pacman-dev] [GIT] The official pacman repository branch, master, updated. v4.2.1-253-g8fa0203
by allan＠archlinux.org 04 Nov '15

04 Nov '15

This is an automated email from the git hooks/post-receive script. It was generated because a ref change was pushed to the repository containing the project "The official pacman repository". The branch, master has been updated via 8fa02036c3175f4cb1e837b6da5281520524ccba (commit) via 726712aa08bb9699b5cea71a4a7a997551e9d7c8 (commit) via 978b197120f3a942cad36726e9678fbbabe3e660 (commit) via 0d877ec4296fdde75946301f9a44723f7cbce5bd (commit) via 08b0b6de96243a67b6db095f760e0c2b22ef4536 (commit) via 556c56d4d573655023bb26c47c57fddbb1a36477 (commit) via 6eac7258cd3ae1cb63e59fceb68cc5d144a9dd4f (commit) via 0fd6d354a6b8de976ec9b6a146c4cb8cdcdd5ebb (commit) via d5b0f0c26f5dc139c5c8e534cafc017956229f66 (commit) via 906dc0ce24d62284b6a3caf612948553e7ffceb3 (commit) via 697377aae3df2ed9743f20c92e807526105c60c0 (commit) from d8621b981e100ecf46d260e1a85cc13d202cafda (commit) Those revisions listed above that are new to this repository have not appeared on any other notification email; so we list those revisions in full, below. - Log ----------------------------------------------------------------- commit 8fa02036c3175f4cb1e837b6da5281520524ccba Author: Andrew Gregory <andrew.gregory.8(a)gmail.com> Date: Sun Nov 1 13:56:38 2015 -0500 files_search: reset found for each target Otherwise any invalid targets following a successful match will not get an error message. Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 726712aa08bb9699b5cea71a4a7a997551e9d7c8 Author: Andrew Gregory <andrew.gregory.8(a)gmail.com> Date: Sun Nov 1 13:56:06 2015 -0500 files_search: free compiled regex Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 978b197120f3a942cad36726e9678fbbabe3e660 Author: Andrew Gregory <andrew.gregory.8(a)gmail.com> Date: Sun Nov 1 13:55:10 2015 -0500 files: do not unnecessarily strdup targets Targets are never modified so we can just use the original copy. Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 0d877ec4296fdde75946301f9a44723f7cbce5bd Author: Andrew Gregory <andrew.gregory.8(a)gmail.com> Date: Sun Nov 1 20:41:56 2015 -0500 files.c: add vim modeline Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 08b0b6de96243a67b6db095f760e0c2b22ef4536 Author: Andrew Gregory <andrew.gregory.8(a)gmail.com> Date: Sun Nov 1 20:42:01 2015 -0500 add detail to broken dependency errors The difference between a sync target having an unmet dependency and breaking a dependency for an installed package is a common source of confusion. Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 556c56d4d573655023bb26c47c57fddbb1a36477 Author: Andrew Gregory <andrew.gregory.8(a)gmail.com> Date: Sun Nov 1 20:42:00 2015 -0500 add --quiet to -Fh Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 6eac7258cd3ae1cb63e59fceb68cc5d144a9dd4f Author: Andrew Gregory <andrew.gregory.8(a)gmail.com> Date: Sun Nov 1 20:41:55 2015 -0500 ensure realloc has a positive size If given size 0 POSIX allows realloc to return a pointer that is not suitable for use. Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 0fd6d354a6b8de976ec9b6a146c4cb8cdcdd5ebb Author: Mohammad Alsaleh <ce.mohammad.alsaleh(a)gmail.com> Date: Fri Oct 23 01:38:50 2015 +0300 zsh_completion: Handle --ignore/--ignoregroup correctly Completion now works as expected with a comma-separated sequence. Signed-off-by: Mohammad Alsaleh <CE.Mohammad.AlSaleh(a)gmail.com> Signed-off-by: Allan McRae <allan(a)archlinux.org> commit d5b0f0c26f5dc139c5c8e534cafc017956229f66 Author: Pierre Neidhardt <ambrevar(a)gmail.com> Date: Mon Nov 2 14:00:32 2015 +0100 package.c: Fix incorrect buffersize and constant-folding Signed-off-by: Rikard Falkeborn <rikard.falkeborn(a)gmail.com> Signed-off-by: Pierre Neidhardt <ambrevar(a)gmail.com> Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 906dc0ce24d62284b6a3caf612948553e7ffceb3 Author: Florian Pritz <bluewind(a)xinu.at> Date: Sat Oct 17 14:37:07 2015 +0200 Add -F --machinereadable option Signed-off-by: Florian Pritz <bluewind(a)xinu.at> Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 697377aae3df2ed9743f20c92e807526105c60c0 Author: Florian Pritz <bluewind(a)xinu.at> Date: Sun Oct 25 16:19:54 2015 +0100 Document pacman -F Signed-off-by: Florian Pritz <bluewind(a)xinu.at> Signed-off-by: Allan McRae <allan(a)archlinux.org> ----------------------------------------------------------------------- Summary of changes: contrib/zsh_completion.in | 21 ++++++++--- doc/pacman.8.txt | 34 ++++++++++++++++++ lib/libalpm/be_local.c | 10 ++++-- lib/libalpm/be_sync.c | 10 ++++-- src/pacman/conf.h | 2 ++ src/pacman/files.c | 88 ++++++++++++++++++++++++++++++----------------- src/pacman/package.c | 7 ++-- src/pacman/pacman.c | 7 ++++ src/pacman/remove.c | 3 +- src/pacman/sync.c | 27 ++++++++++++--- 10 files changed, 158 insertions(+), 51 deletions(-) hooks/post-receive -- The official pacman repository

1 0

[pacman-dev] [PATCH] ensure realloc has a positive size
by Andrew Gregory 03 Nov '15

03 Nov '15

If given size 0 POSIX allows realloc to return a pointer that is not suitable for use. Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> --- lib/libalpm/be_local.c | 10 +++++++--- lib/libalpm/be_sync.c | 10 +++++++--- 2 files changed, 14 insertions(+), 6 deletions(-) diff --git a/lib/libalpm/be_local.c b/lib/libalpm/be_local.c index 556157d..e35fc0e 100644 --- a/lib/libalpm/be_local.c +++ b/lib/libalpm/be_local.c @@ -810,9 +810,13 @@ static int local_db_read(alpm_pkg_t *info, alpm_dbinfrq_t inforeq) files_count++; } /* attempt to hand back any memory we don't need */ - files = realloc(files, sizeof(alpm_file_t) * files_count); - /* make sure the list is sorted */ - qsort(files, files_count, sizeof(alpm_file_t), _alpm_files_cmp); + if(files_count > 0) { + files = realloc(files, sizeof(alpm_file_t) * files_count); + /* make sure the list is sorted */ + qsort(files, files_count, sizeof(alpm_file_t), _alpm_files_cmp); + } else { + FREE(files); + } info->files.count = files_count; info->files.files = files; } else if(strcmp(line, "%BACKUP%") == 0) { diff --git a/lib/libalpm/be_sync.c b/lib/libalpm/be_sync.c index 25c8cbe..b09b060 100644 --- a/lib/libalpm/be_sync.c +++ b/lib/libalpm/be_sync.c @@ -721,9 +721,13 @@ static int sync_db_read(alpm_db_t *db, struct archive *archive, files_count++; } /* attempt to hand back any memory we don't need */ - files = realloc(files, sizeof(alpm_file_t) * files_count); - /* make sure the list is sorted */ - qsort(files, files_count, sizeof(alpm_file_t), _alpm_files_cmp); + if(files_count > 0) { + files = realloc(files, sizeof(alpm_file_t) * files_count); + /* make sure the list is sorted */ + qsort(files, files_count, sizeof(alpm_file_t), _alpm_files_cmp); + } else { + FREE(files); + } pkg->files.count = files_count; pkg->files.files = files; } -- 2.6.2

2 2

[pacman-dev] [PATCH 1/2] zsh_completion: Add --asdeps/--asexplicit to -U options
by Mohammad Alsaleh 03 Nov '15

03 Nov '15

Signed-off-by: Mohammad Alsaleh <CE.Mohammad.AlSaleh(a)gmail.com> --- contrib/zsh_completion.in | 2 ++ 1 file changed, 2 insertions(+) diff --git a/contrib/zsh_completion.in b/contrib/zsh_completion.in index 1cae1de..69485f5 100644 --- a/contrib/zsh_completion.in +++ b/contrib/zsh_completion.in @@ -41,6 +41,8 @@ _pacman_opts_pkgfile=( '--dbonly[Only remove database entry, do not remove files]' '--force[Overwrite conflicting files]' '--needed[Do not reinstall up to date packages]' + '--asdeps[mark packages as non-explicitly installed]' + '--asexplicit[mark packages as explicitly installed]' '*:package file:_files -g "*.pkg.tar*~*.sig(.,@)"' ) -- 2.6.1

3 6

[pacman-dev] [PATCH] be_package: validate package file paths
by Andrew Gregory 03 Nov '15

03 Nov '15

Overly long paths cannot be extracted and paths with newlines cannot be represented in our database format. Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> --- lib/libalpm/be_package.c | 15 ++++++++++++--- test/pacman/tests/TESTS | 3 +++ test/pacman/tests/filename-basename-too-long.py | 15 +++++++++++++++ test/pacman/tests/filename-path-too-long.py | 20 ++++++++++++++++++++ test/pacman/tests/filename-with-newline.py | 11 +++++++++++ 5 files changed, 61 insertions(+), 3 deletions(-) create mode 100644 test/pacman/tests/filename-basename-too-long.py create mode 100644 test/pacman/tests/filename-path-too-long.py create mode 100644 test/pacman/tests/filename-with-newline.py diff --git a/lib/libalpm/be_package.c b/lib/libalpm/be_package.c index 53399a3..52db319 100644 --- a/lib/libalpm/be_package.c +++ b/lib/libalpm/be_package.c @@ -18,6 +18,7 @@ * along with this program. If not, see <http://www.gnu.org/licenses/>. */ +#include <limits.h> #include <stdlib.h> #include <string.h> #include <errno.h> @@ -390,7 +391,17 @@ static int add_entry_to_files_list(alpm_filelist_t *filelist, const size_t files_count = filelist->count; alpm_file_t *current_file; mode_t type; - size_t pathlen; + size_t pathlen = strlen(path); + + /* +2 to leave space for prepending minimal possible root and appending + * trailing slash if a directory */ + if(pathlen + 2 >= PATH_MAX || strlen(mbasename(path)) >= NAME_MAX) { + return -1; + } + /* our database format cannot represent paths with newlines */ + if(memchr(path, '\n', pathlen)) { + return -1; + } if(!_alpm_greedy_grow((void **)&filelist->files, files_size, (files_count + 1) * sizeof(alpm_file_t))) { @@ -399,8 +410,6 @@ static int add_entry_to_files_list(alpm_filelist_t *filelist, type = archive_entry_filetype(entry); - pathlen = strlen(path); - current_file = filelist->files + files_count; /* mtree paths don't contain a tailing slash, those we get from diff --git a/test/pacman/tests/TESTS b/test/pacman/tests/TESTS index 8ad1b9c..210134f 100644 --- a/test/pacman/tests/TESTS +++ b/test/pacman/tests/TESTS @@ -50,6 +50,9 @@ TESTS += test/pacman/tests/fileconflict025.py TESTS += test/pacman/tests/fileconflict030.py TESTS += test/pacman/tests/fileconflict031.py TESTS += test/pacman/tests/fileconflict032.py +TESTS += test/pacman/tests/filename-basename-too-long.py +TESTS += test/pacman/tests/filename-path-too-long.py +TESTS += test/pacman/tests/filename-with-newline.py TESTS += test/pacman/tests/hook-abortonfail.py TESTS += test/pacman/tests/hook-file-change-packages.py TESTS += test/pacman/tests/hook-file-remove-trigger-match.py diff --git a/test/pacman/tests/filename-basename-too-long.py b/test/pacman/tests/filename-basename-too-long.py new file mode 100644 index 0000000..93a14a5 --- /dev/null +++ b/test/pacman/tests/filename-basename-too-long.py @@ -0,0 +1,15 @@ +import os + +self.description = "Install packages with overly long file names" + +filename = 'A' * (os.pathconf(self.rootdir(), 'PC_NAME_MAX') + 1) + +p = pmpkg("foo") +p.files = [filename] +self.addpkg(p) + +self.args = "-U %s" % p.filename() + +self.addrule("!PACMAN_RETCODE=0") +self.addrule("!PKG_EXIST=foo") +self.addrule("!FILE_EXIST=%s" % filename) diff --git a/test/pacman/tests/filename-path-too-long.py b/test/pacman/tests/filename-path-too-long.py new file mode 100644 index 0000000..c4278b5 --- /dev/null +++ b/test/pacman/tests/filename-path-too-long.py @@ -0,0 +1,20 @@ +import os + +self.description = "Install packages with overly long file paths" + +# construct a path > PATH_MAX with the fewest possible components to avoid +# having to build giant filelists and deep recursion in rmtree() +name_max = os.pathconf(self.rootdir(), 'PC_NAME_MAX') - 1 +path_max = os.pathconf(self.rootdir(), 'PC_PATH_MAX') +basename = 'A' * name_max + '/' +filename = basename * ((path_max / name_max) + 1) + +p = pmpkg("foo") +p.files = [filename] +self.addpkg(p) + +self.args = "-U %s" % p.filename() + +self.addrule("!PACMAN_RETCODE=0") +self.addrule("!PKG_EXIST=foo") +self.addrule("!FILE_EXIST=%s" % filename) diff --git a/test/pacman/tests/filename-with-newline.py b/test/pacman/tests/filename-with-newline.py new file mode 100644 index 0000000..df412db --- /dev/null +++ b/test/pacman/tests/filename-with-newline.py @@ -0,0 +1,11 @@ +self.description = "Install a package including a file with a newline in its name" + +p = pmpkg("foo") +p.files = ["foo\nbar"] +self.addpkg(p) + +self.args = "-U %s" % p.filename() + +self.addrule("!PACMAN_RETCODE=0") +self.addrule("!PKG_EXIST=foo") +self.addrule("!FILE_EXIST=foo\nbar") -- 2.6.2

2 1

[pacman-dev] [PATCH 1/2] Document pacman -F
by Florian Pritz 02 Nov '15

02 Nov '15

Signed-off-by: Florian Pritz <bluewind(a)xinu.at> --- doc/pacman.8.txt | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) diff --git a/doc/pacman.8.txt b/doc/pacman.8.txt index 5c57450..a4dd1f0 100644 --- a/doc/pacman.8.txt +++ b/doc/pacman.8.txt @@ -110,6 +110,11 @@ greater than `1:3.6-1`. Options>> below; also see <<HCF,Handling Config Files>> for an explanation on how pacman takes care of configuration files. +*-F, \--files*:: + Query the files database. This operation allows you to look for package + owning certain files or display files owned by certain packages. See + <<FO,File Options>> below. + *-V, \--version*:: Display version and exit. @@ -451,6 +456,23 @@ Database Options[[QO]] a check on the sync databases to ensure all specified dependencies are available. +File Options[[FO]] +------------------ + +The '--list', '--search' and '--owns' options are exclusive and can not be combined. + +*-l <package>, \--list <package>*:: + List the files owned by the queried package. + +*-s <pattern>, \--search <pattern>*:: + Search package file names for matching strings. + +*-x, --regex*:: + Treat arguments to '--search' as regular expressions. + +*-o <file>, \--owns<file>*:: + Search for packages that own a particular file. + Handling Config Files[[HCF]] ---------------------------- -- 2.6.2

5 19

[pacman-dev] [PATCH v2] package.c: Fix incorrect buffersize and constant-folding
by Pierre Neidhardt 02 Nov '15

02 Nov '15

Signed-off-by: Rikard Falkeborn <rikard.falkeborn(a)gmail.com> Signed-off-by: Pierre Neidhardt <ambrevar(a)gmail.com> --- src/pacman/package.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/src/pacman/package.c b/src/pacman/package.c index dbd23f5..c9301ac 100644 --- a/src/pacman/package.c +++ b/src/pacman/package.c @@ -90,9 +90,8 @@ static void make_aligned_titles(void) { unsigned int i; size_t max = 0; - static const wchar_t *title_suffix = L" :"; - static const size_t title_suffix_len = sizeof(title_suffix); - wchar_t wbuf[ARRAYSIZE(titles)][TITLE_MAXLEN + title_suffix_len]; + static const wchar_t title_suffix[] = L" :"; + wchar_t wbuf[ARRAYSIZE(titles)][TITLE_MAXLEN + ARRAYSIZE(title_suffix)]; size_t wlen[ARRAYSIZE(wbuf)]; char *buf[ARRAYSIZE(wbuf)]; buf[T_ARCHITECTURE] = _("Architecture"); @@ -133,7 +132,7 @@ static void make_aligned_titles(void) for(i = 0; i < ARRAYSIZE(wbuf); i++) { wmemset(wbuf[i] + wlen[i], L' ', max - wlen[i]); - wmemcpy(wbuf[i] + max, title_suffix, title_suffix_len); + wmemcpy(wbuf[i] + max, title_suffix, ARRAYSIZE(title_suffix)); wcstombs(titles[i], wbuf[i], sizeof(wbuf[i])); } } -- 2.6.2

1 0