- Pacman-dev - lists.archlinux.org

[pacman-dev] [PATCH] doc: add a couple missing manpages from the html index
by Eli Schwartz 13 Jan '19

13 Jan '19

The index embeds the names of all doc subpages, and these were never updated. Signed-off-by: Eli Schwartz <eschwartz(a)archlinux.org> --- doc/index.asciidoc | 2 ++ 1 file changed, 2 insertions(+) diff --git a/doc/index.asciidoc b/doc/index.asciidoc index c9e5bcc1..fa7444db 100644 --- a/doc/index.asciidoc +++ b/doc/index.asciidoc @@ -43,9 +43,11 @@ There are several man pages available for the programs, utilities, and configuration files dealing with pacman. * linkman:alpm-hooks[5] +* linkman:BUILDINFO[5] * linkman:PKGBUILD[5] * linkman:libalpm[3] * linkman:makepkg[8] +* linkman:makepkg-template[1] * linkman:makepkg.conf[5] * linkman:pacman[8] * linkman:pacman-key[8] -- 2.20.1

1 0

[pacman-dev] [PATCH] doc: keep manpages and html pages in sync
by Eli Schwartz 13 Jan '19

13 Jan '19

We forgot to add BUILDINFO to the list of html docs. Instead of always updating things in two places, just derive the one from the other. meson did not have this problem as it already derives both lists from one template. Signed-off-by: Eli Schwartz <eschwartz(a)archlinux.org> --- doc/Makefile.am | 14 +------------- 1 file changed, 1 insertion(+), 13 deletions(-) diff --git a/doc/Makefile.am b/doc/Makefile.am index 4bb0f01b..de87c409 100644 --- a/doc/Makefile.am +++ b/doc/Makefile.am @@ -20,19 +20,7 @@ MANPAGES = \ DOXYGEN_MANS = $(wildcard man3/*.3) -HTML_MANPAGES = \ - alpm-hooks.5.html \ - pacman.8.html \ - makepkg.8.html \ - makepkg-template.1.html \ - repo-add.8.html \ - vercmp.8.html \ - pkgdelta.8.html \ - pacman-key.8.html \ - PKGBUILD.5.html \ - makepkg.conf.5.html \ - pacman.conf.5.html \ - libalpm.3.html +HTML_MANPAGES = $(addsuffix .html,$(MANPAGES)) HTML_OTHER = \ index.html \ -- 2.20.1

1 0

[pacman-dev] [PATCH v2 0/3] makepkg: Improve lint_pkgbuild error messages for epoch/ver/rel
by Luke Shumaker 10 Jan '19

10 Jan '19

From: Luke Shumaker <lukeshu(a)parabola.nu> The error messages that makepkg prints when encountering a bad epoch/pkgver/pkgrel are often somewhat misleading. These patchset tries to improve that. The behavior is intended to be unchanged, other than what the messages say. v2: - Don't skip lint_pkgrel if (( PKGVERFUNC )), I should have never included that bit in v1. - check_fullpkgver: Don't let rel strip ver down to nothing Luke Shumaker (3): makepkg: Better error messages for versions in (check,make,opt)depends/provides/conflicts makepkg: check_pkgrel: Don't say "decimal" in the error message makepkg: check_pkgver: Report what the bad pkgver is scripts/Makefile.am | 1 + .../lint_pkgbuild/checkdepends.sh.in | 10 ++-- .../libmakepkg/lint_pkgbuild/conflicts.sh.in | 10 ++-- .../libmakepkg/lint_pkgbuild/depends.sh.in | 14 ++--- scripts/libmakepkg/lint_pkgbuild/epoch.sh.in | 10 +++- .../libmakepkg/lint_pkgbuild/fullpkgver.sh.in | 58 +++++++++++++++++++ .../lint_pkgbuild/makedepends.sh.in | 10 ++-- .../libmakepkg/lint_pkgbuild/optdepends.sh.in | 10 ++-- scripts/libmakepkg/lint_pkgbuild/pkgrel.sh.in | 15 +++-- scripts/libmakepkg/lint_pkgbuild/pkgver.sh.in | 2 +- .../libmakepkg/lint_pkgbuild/provides.sh.in | 10 ++-- 11 files changed, 106 insertions(+), 44 deletions(-) create mode 100644 scripts/libmakepkg/lint_pkgbuild/fullpkgver.sh.in -- Happy hacking, ~ Luke Shumaker 2.18.0

2 6

[pacman-dev] [PATCH] makepkg: fix unguarded use of printf
by Eli Schwartz 10 Jan '19

10 Jan '19

paths can contain printf-unsafe chars, and printf -v is not somehow immune to this Signed-off-by: Eli Schwartz <eschwartz(a)archlinux.org> --- scripts/makepkg.sh.in | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/makepkg.sh.in b/scripts/makepkg.sh.in index 56642b11..a040d384 100644 --- a/scripts/makepkg.sh.in +++ b/scripts/makepkg.sh.in @@ -1142,7 +1142,7 @@ fi # canonicalize paths and provide defaults if anything is still undefined for var in PKGDEST SRCDEST SRCPKGDEST LOGDEST BUILDDIR; do - printf -v "$var" "$(canonicalize_path "${!var:-$startdir}")" + printf -v "$var" '%s' "$(canonicalize_path "${!var:-$startdir}")" done unset var PACKAGER=${PACKAGER:-"Unknown Packager"} -- 2.20.1

1 0

[pacman-dev] [PATCH] libmakepkg/lint_pkgbuild: check for invalid variables even if they're empty
by Eli Schwartz 10 Jan '19

10 Jan '19

Checking the length of the variable to be non-zero before considering it an error is inconsistent; license=() and depends='' and `declare arch` should be considered just as wrong. In fact the current check detects depends='' as non-zero and returns an error, but happily considers the others to be perfectly okay. A more reliable check is to simply see if the name has been declared (whether it is set or not), and then enforce that it's been declared to the right type. As an added benefit, avoiding the creation of proxy-evaled variables to count the number of indexes results in simpler code. Signed-off-by: Eli Schwartz <eschwartz(a)archlinux.org> --- scripts/libmakepkg/lint_pkgbuild/variable.sh.in | 16 ++++++---------- 1 file changed, 6 insertions(+), 10 deletions(-) diff --git a/scripts/libmakepkg/lint_pkgbuild/variable.sh.in b/scripts/libmakepkg/lint_pkgbuild/variable.sh.in index 3266cb5e..52bb61a9 100644 --- a/scripts/libmakepkg/lint_pkgbuild/variable.sh.in +++ b/scripts/libmakepkg/lint_pkgbuild/variable.sh.in @@ -37,12 +37,11 @@ lint_variable() { replaces sha1sums sha256sums sha384sums sha512sums source) local string=(changelog epoch install pkgdesc pkgrel pkgver url) - local i a v pkg keys out bad ret=0 + local i a pkg out bad ret=0 # global variables for i in ${array[@]} ${arch_array[@]}; do - eval "keys=(\"\${!$i[@]}\")" - if (( ${#keys[*]} > 0 )); then + if declare -p $i > /dev/null 2>&1; then if ! is_array $i; then error "$(gettext "%s should be an array")" "$i" ret=1 @@ -54,11 +53,9 @@ lint_variable() { [[ $a == "any" ]] && continue for i in ${arch_array[@]}; do - v="${i}_${a}" - eval "keys=(\"\${!${v}[@]}\")" - if (( ${#keys[*]} > 0 )); then - if ! is_array $v; then - error "$(gettext "%s_%s should be an array")" "$i" "$a" + if declare -p "${i}_${a}" > /dev/null 2>&1; then + if ! is_array ${i}_${a}; then + error "$(gettext "%s should be an array")" "${i}_${a}" ret=1 fi fi @@ -66,8 +63,7 @@ lint_variable() { done for i in ${string[@]}; do - eval "keys=(\"\${!$i[@]}\")" - if (( ${#keys[*]} > 0 )); then + if declare -p "$i" > /dev/null 2>&1; then if is_array $i; then error "$(gettext "%s should not be an array")" "$i" ret=1 -- 2.17.1

2 1

[pacman-dev] [GIT] The official pacman repository branch, master, updated. v5.1.1-101-g9b637cc2
by Allan McRae 10 Jan '19

10 Jan '19

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 9b637cc27aba561e5270862a1a3eb515e4f3f744 (commit) via f9eb2aacb4655cce5faa8a2e7295a96cab91c68c (commit) from 238fa4af4507c573ae4d478d1d7e176429d2c5c5 (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 9b637cc27aba561e5270862a1a3eb515e4f3f744 Author: Allan McRae <allan(a)archlinux.org> Date: Thu Jan 10 14:10:47 2019 +1000 be_local.c: remove aligment in local_pkg_ops The alignment was not overly helpful and caused unnecessary churn when a new value with longer name was added. Signed-off-by: Allan McRae <allan(a)archlinux.org> commit f9eb2aacb4655cce5faa8a2e7295a96cab91c68c Author: morganamilo <morganamilo(a)gmail.com> Date: Tue Oct 9 02:05:15 2018 +0100 libalpm: parse {check, make}depends when reading database Commit 0994893b0e6b627d45a63884ac01af7d0967eff2 added the alpm_pkg_get_{make,check}depends functions but forgot to include logic for parsing these fields from the database. As a result these functions will always return an empty list. This commit adds the parsing logic. Signed-off-by: morganamilo <morganamilo(a)gmail.com> Signed-off-by: Allan McRae <allan(a)archlinux.org> ----------------------------------------------------------------------- Summary of changes: lib/libalpm/be_local.c | 70 +++++++++++++++++++++++++++++++------------------- lib/libalpm/be_sync.c | 12 ++------- 2 files changed, 46 insertions(+), 36 deletions(-) hooks/post-receive -- The official pacman repository

1 0

[pacman-dev] [GIT] The official pacman repository branch, master, updated. v5.1.1-99-g238fa4af
by Allan McRae 10 Jan '19

10 Jan '19

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 238fa4af4507c573ae4d478d1d7e176429d2c5c5 (commit) via 268b0255fa2dd6ccef2e4fa08859c7966002e15a (commit) via c5c7fc7f4a361a9a347845b930030077898cd83f (commit) via 2f15921230a424c57c13dc0ba0776d87c2f590c4 (commit) via 7c5958e0eb8f77feab743adcce05b8bfa600b69e (commit) from e1fc2f901ea7d70787d803ffb5c39fe7ad8bb92e (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 238fa4af4507c573ae4d478d1d7e176429d2c5c5 Author: Eli Schwartz <eschwartz(a)archlinux.org> Date: Fri Jan 4 00:26:28 2019 -0500 makepkg: fix broken check for the fakeroot binary In commit d8ee8d0c99c3820951e2e49dbdb71a5390bd1dc4 we made use of fakeroot absolutely mandatory, and disabled a lot of the code which checked to see if this now-defunct BUILDENV option was set, before setting up the environment to use fakeroot. Unfortunately, we missed one spot. The check_software routine still checked to see if fakeroot was enabled, but due to the option being removed, thought that it was in fact disabled, and as a result this check would never run. Fix by checking to see if we are trying to build either a package or a source package, and if so, checking for fakeroot. These are the only two situations where fakeroot is needed. Signed-off-by: Eli Schwartz <eschwartz(a)archlinux.org> Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 268b0255fa2dd6ccef2e4fa08859c7966002e15a Author: Eli Schwartz <eschwartz(a)archlinux.org> Date: Fri Jan 4 00:32:09 2019 -0500 makepkg: add internal variable to track when we're building a package There are state variables for everything else, and we use them to do conditional checks on things, but it's currently a bit difficult to test whether a package is being built, as it's the default action if *no* options are specified. Signed-off-by: Eli Schwartz <eschwartz(a)archlinux.org> Signed-off-by: Allan McRae <allan(a)archlinux.org> commit c5c7fc7f4a361a9a347845b930030077898cd83f Author: Jan Alexander Steffens (heftig) <jan.steffens(a)gmail.com> Date: Fri Jan 4 13:13:16 2019 +1000 makepkg: Make pkgdir a local Causes it to be reset (to $pkgdirbase/$pkgbase) between subpackages. This shouldn't be visible. Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 2f15921230a424c57c13dc0ba0776d87c2f590c4 Author: Jan Alexander Steffens (heftig) <jan.steffens(a)gmail.com> Date: Fri Jan 4 13:09:38 2019 +1000 makepkg: Create a single-package packaging function Merge the similar code handling unsplit PKGBUILDs and individual packages in a split PKGBUILD and make it a new function. Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 7c5958e0eb8f77feab743adcce05b8bfa600b69e Author: Jan Alexander Steffens (heftig) <jan.steffens(a)gmail.com> Date: Fri Jan 4 12:57:22 2019 +1000 makepkg: Only backup package variables once We don't need to re-backup the variables we restored on the previous iteration. Signed-off-by: Allan McRae <allan(a)archlinux.org> ----------------------------------------------------------------------- Summary of changes: scripts/libmakepkg/executable/fakeroot.sh.in | 2 +- scripts/makepkg.sh.in | 50 ++++++++++++++-------------- 2 files changed, 26 insertions(+), 26 deletions(-) hooks/post-receive -- The official pacman repository

1 0

[pacman-dev] [PATCH v2 1/2] libalpm/sync.c: change alpm_sync_newversion() to alpm_sync_get_new_version()
by Allan McRae 10 Jan '19

10 Jan '19

The behaviour of "pacman -Qu" was very strange... It would only consider packages from repos with Usage = Search (or All), and ignore those with Usage = Sync, Install or Upgrade. This is because the function alpm_sync_newversion() used ALPM_DB_USAGE_SEARCH for its filtering. Given this function is documented (at least in the source) to "Check for new version of pkg in sync repos", I would expect that to look at all repos. However, just changing this parameter, would result in a fairly silent change in behaviour of this function. Instead, rename the function and remove this filtering altogether. Users of this function can filter the dbs passed to this function to achieve their desired output. Signed-off-by: Allan McRae <allan(a)archlinux.org> --- This patch should address the comments Andrew made. I have not added the helper function for filtering databases by alpm_usage_type yet, but did file a bug to track it (FS#61333). lib/libalpm/alpm.h | 2 +- lib/libalpm/sync.c | 6 +----- src/pacman/query.c | 4 ++-- 3 files changed, 4 insertions(+), 8 deletions(-) diff --git a/lib/libalpm/alpm.h b/lib/libalpm/alpm.h index 597e11bd..bbce0971 100644 --- a/lib/libalpm/alpm.h +++ b/lib/libalpm/alpm.h @@ -1452,7 +1452,7 @@ alpm_list_t *alpm_find_group_pkgs(alpm_list_t *dbs, const char *name); * Sync */ -alpm_pkg_t *alpm_sync_newversion(alpm_pkg_t *pkg, alpm_list_t *dbs_sync); +alpm_pkg_t *alpm_sync_get_new_version(alpm_pkg_t *pkg, alpm_list_t *dbs_sync); /** @addtogroup alpm_api_trans Transaction Functions * Functions to manipulate libalpm transactions diff --git a/lib/libalpm/sync.c b/lib/libalpm/sync.c index 05f58fad..de579eac 100644 --- a/lib/libalpm/sync.c +++ b/lib/libalpm/sync.c @@ -51,7 +51,7 @@ /** Check for new version of pkg in sync repos * (only the first occurrence is considered in sync) */ -alpm_pkg_t SYMEXPORT *alpm_sync_newversion(alpm_pkg_t *pkg, alpm_list_t *dbs_sync) +alpm_pkg_t SYMEXPORT *alpm_sync_get_new_version(alpm_pkg_t *pkg, alpm_list_t *dbs_sync) { alpm_list_t *i; alpm_pkg_t *spkg = NULL; @@ -61,10 +61,6 @@ alpm_pkg_t SYMEXPORT *alpm_sync_newversion(alpm_pkg_t *pkg, alpm_list_t *dbs_syn for(i = dbs_sync; !spkg && i; i = i->next) { alpm_db_t *db = i->data; - if(!(db->usage & ALPM_DB_USAGE_SEARCH)) { - continue; - } - spkg = _alpm_db_get_pkgfromcache(db, pkg->name); } diff --git a/src/pacman/query.c b/src/pacman/query.c index 00c39638..9ac6e930 100644 --- a/src/pacman/query.c +++ b/src/pacman/query.c @@ -286,7 +286,7 @@ static int filter(alpm_pkg_t *pkg) return 0; } /* check if this pkg is outdated */ - if(config->op_q_upgrade && (alpm_sync_newversion(pkg, + if(config->op_q_upgrade && (alpm_sync_get_new_version(pkg, alpm_get_syncdbs(config->handle)) == NULL)) { return 0; } @@ -325,7 +325,7 @@ static int display(alpm_pkg_t *pkg) colstr->version, alpm_pkg_get_version(pkg), colstr->nocolor); if(config->op_q_upgrade) { - alpm_pkg_t *newpkg = alpm_sync_newversion(pkg, alpm_get_syncdbs(config->handle)); + alpm_pkg_t *newpkg = alpm_sync_get_new_version(pkg, alpm_get_syncdbs(config->handle)); printf(" -> %s%s%s", colstr->version, alpm_pkg_get_version(newpkg), colstr->nocolor); if(alpm_pkg_should_ignore(config->handle, pkg)) { -- 2.20.0

1 1

[pacman-dev] [PATCH] Add -fstack-clash-protection to CFLAGS in debug builds if available
by Allan McRae 10 Jan '19

10 Jan '19

Signed-off-by: Allan McRae <allan(a)archlinux.org> --- Tested with GCC (supports this flag) and clang (does not support). Did not add to meson builds, as it appears that all these flags are not added to the build anywhere in the meson files. configure.ac | 1 + m4/acinclude.m4 | 19 +++++++++++++++++++ 2 files changed, 20 insertions(+) diff --git a/configure.ac b/configure.ac index ba54c2dd..6f336022 100644 --- a/configure.ac +++ b/configure.ac @@ -443,6 +443,7 @@ if test "x$debug" = "xyes" ; then # Check for -fstack-protector availability GCC_STACK_PROTECT_LIB GCC_STACK_PROTECT_CC + GCC_STACK_CLASH_PROTECTION GCC_FORTIFY_SOURCE_CC WARNING_CFLAGS="-g -Wall -Werror" else diff --git a/m4/acinclude.m4 b/m4/acinclude.m4 index 15bb31e8..845c8286 100644 --- a/m4/acinclude.m4 +++ b/m4/acinclude.m4 @@ -38,6 +38,25 @@ AC_DEFUN([GCC_STACK_PROTECT_CC],[ fi ]) +dnl GCC_STACK_CLASH_PROTECTION +dnl check -fstack-clash-protection with the C compiler, if it exists then +dnl updates CFLAGS +AC_DEFUN([GCC_STACK_CLASH_PROTECTION],[ + AC_LANG_ASSERT(C) + if test "X$CC" != "X"; then + AC_CACHE_CHECK([whether ${CC} accepts -fstack-clash-protection], + scp_cv_cc, + [scp_old_cflags="$CFLAGS" + CFLAGS="$CFLAGS -fstack-clash-protection" + AC_TRY_COMPILE(,, scp_cv_cc=yes, scp_cv_cc=no) + CFLAGS="$scp_old_cflags" + ]) + if test $scp_cv_cc = yes; then + CFLAGS="$CFLAGS -fstack-clash-protection" + fi + fi +]) + dnl GCC_FORTIFY_SOURCE_CC dnl checks -D_FORTIFY_SOURCE with the C compiler, if it exists then updates dnl CPPFLAGS -- 2.20.0

1 0

[pacman-dev] RFC: Really running package_* functions in rbash during printsrcinfo
by Charles Duffy 06 Jan '19

06 Jan '19

Howdy -- I recently had need to dig into the implementation of makepkg --printsrcinfo, and ran into the "running regular expressions against source code" operations in the backend. Obviously, this is not ideal. Indeed, I've previously written packages (doing unusual and typically-undesirable things, granted) with conditional logic *assuming* that actual execution would be taking place. I fully appreciate the decision not to try to go with more expansive attempts at emulating bash parsing/execution in the future, but do folks have any thoughts on **really** executing PKGBUILDs in a restricted environment, including execution of the package_* functions? See a simple sandboxed parser for config files implemented as bash code in code I've written for NixOS at https://github.com/charles-dyfis-net/nixpkgs/blob/f50bfe267a312515d88e86c12…. We might need a little more complexity here -- using DEBUG traps to avoid "|| exit" logic from aborting, f/e -- but my initial impression is that "more accurate than the current implementation" (and maybe a fair bit faster, if we extract all variables in one subshell per function) is not a hard goal to achieve. Thoughts?

2 3