- Pacman-dev - lists.archlinux.org

Runtime dependencies not required for building
by Hugo Osvaldo Barrera 02 Aug '22

02 Aug '22

Some packages I maintain have a lot of runtime dependencies that are not required at build time (this is common for packages which only ship scripts and have no tests). When building on a clean chroot, I have to wait for hundreds of megabytes in packages to be copied and installed. This is rather inefficient, since these are not required. However, the `makedepends` ARE required in this case, but there is no overlap. I'd like some way to indicate that a PKGBUILD does not require its runtime dependencies for building the package. Perhaps something like `options=(!depends)` I'd like to hear feedback on this, thoughts? -- Hugo Osvaldo Barrera

2 2

[Git][pacman/pacman][master] 5 commits: meson: replace deprecated get_pkgconfig_variable()
by Allan McRae (＠allan) 29 Jul '22

29 Jul '22

Allan McRae pushed to branch master at Pacman / Pacman Commits: daa2db6a by Joe Baldino at 2022-07-29T09:41:47+10:00 meson: replace deprecated get_pkgconfig_variable() Apparently that function was deprecated in 0.56, so use the generic getter introduced in 0.51 instead. This squashes a warning. Signed-off-by: Joe Baldino <pedanticdm(a)gmx.us> Signed-off-by: Allan McRae <allan(a)archlinux.org> - - - - - 5abe1455 by Alexander F. Rødseth at 2022-07-29T09:42:38+10:00 pacman: remove redundant argument The fill_progress function is called from two locations, and both locations pass in the same percentage value twice. This patch modifies the function signature to to receive the percentage value just once. Signed-off-by: Alexander F. Rødseth <xyproto(a)archlinux.org> Signed-off-by: Allan McRae <allan(a)archlinux.org> - - - - - 51b2e1c9 by Alexander F. Rødseth at 2022-07-29T09:44:04+10:00 pacman: let the progress bar "C"/"c" move up and down at a more regular rate For some terminal widths, the "C"/"c" character does not alternate at regular intervals, but may look like it is stuck at either lowercase or uppercase. The previous behavior toggled based on the character position, while this new behavior toggles the chomp alternation based on the progress percentage value. This leads to slightly improved chomping. Signed-off-by: Alexander F. Rødseth <xyproto(a)archlinux.org> Signed-off-by: Allan McRae <allan(a)archlinux.org> - - - - - 79bd5121 by Daan De Meyer at 2022-07-29T10:17:52+10:00 Add --keyringdir meson option to configure the keyring directory On Debian, keyrings are stored in /usr/share/keyrings. To support this, let's add a new --keyringdir option that allows configuring the directory under datarootdir where the keyrings should be imported from. We default to 'pacman/keyrings' for backwards compatibility. - - - - - e017a597 by soloturn at 2022-07-29T11:00:01+10:00 makepkg: Add GITFLAGS environmental variable to customise checkout The default flag used to clone a git repository when using makepkg is "--mirror". However, when working with huge repositories, the use of different flags during cloning can allow an faster checkout. For example, using "--filter=blob:none" allows for small checkouts, at the expense of requiring downloads during the build stage if anything but the HEAD commit is used for the build. In addition, this example would serve as a replacement for the often requested (but broken) addition of --depth=1. Add support for the environment variable GITFLAG to pass flags for the git clone command. Note that this overrides the default rather than adding to it in order to prevent incompatibilities. - - - - - 10 changed files: - build-aux/edit-script.sh.in - doc/makepkg.8.asciidoc - doc/meson.build - doc/pacman-key.8.asciidoc - meson.build - meson_options.txt - scripts/completion/zsh_completion.in - scripts/libmakepkg/source/git.sh.in - scripts/pacman-key.sh.in - src/pacman/callback.c View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/compare/ff7c6c8e570bcf5e078763… -- View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/compare/ff7c6c8e570bcf5e078763… You're receiving this email because of your account on gitlab.archlinux.org.

1 0

[PATCH 1/2] pacman: remove redundant argument
by Alexander F. Rødseth 28 Jul '22

28 Jul '22

The fill_progress function is called from two locations, and both locations pass in the same percentage value twice. This patch modifies the function signature to to receive the percentage value just once. Signed-off-by: Alexander F. Rødseth <xyproto(a)archlinux.org> --- src/pacman/callback.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/src/pacman/callback.c b/src/pacman/callback.c index a06f3bcc..df4032a4 100644 --- a/src/pacman/callback.c +++ b/src/pacman/callback.c @@ -150,12 +150,11 @@ static int64_t get_update_timediff(int first_call) } /* refactored from cb_trans_progress */ -static void fill_progress(const int bar_percent, const int disp_percent, - const int proglen) +static void fill_progress(const int percent, const int proglen) { /* 8 = 1 space + 1 [ + 1 ] + 5 for percent */ const int hashlen = proglen > 8 ? proglen - 8 : 0; - const int hash = bar_percent * hashlen / 100; + const int hash = percent * hashlen / 100; int i; if(hashlen > 0) { @@ -188,7 +187,7 @@ static void fill_progress(const int bar_percent, const int disp_percent, /* print display percent after progress bar */ /* 5 = 1 space + 3 digits + 1 % */ if(proglen >= 5) { - printf(" %3d%%", disp_percent); + printf(" %3d%%", percent); } putchar('\r'); @@ -677,7 +676,7 @@ void cb_progress(void *ctx, alpm_progress_t event, const char *pkgname, free(wcstr); /* call refactored fill progress function */ - fill_progress(percent, percent, cols - infolen); + fill_progress(percent, cols - infolen); if(percent == 100) { putchar('\n'); @@ -855,7 +854,7 @@ static void draw_pacman_progress_bar(struct pacman_progress_bar *bar) free(fname); free(wcfname); - fill_progress(file_percent, file_percent, cols - infolen); + fill_progress(file_percent, cols - infolen); return; } -- 2.37.1

2 3

[PATCH] meson: replace deprecated get_pkgconfig_variable()
by Joe Baldino 28 Jul '22

28 Jul '22

Apparently that function was deprecated in 0.56, so use the generic getter introduced in 0.51 instead. This squashes a warning. Signed-off-by: Joe Baldino <pedanticdm(a)gmx.us> --- Notes: WARNING: Deprecated features used: * 0.56.0: {'dependency.get_pkgconfig_variable'} See: https://mesonbuild.com/Reference-manual_returned_dep.html#depget_variable meson.build | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/meson.build b/meson.build index 1519a2bbb4ef..6ea97672e33d 100644 --- a/meson.build +++ b/meson.build @@ -43,7 +43,7 @@ endif bashcompletion = dependency('bash-completion', required : false) if bashcompletion.found() - BASHCOMPDIR = bashcompletion.get_pkgconfig_variable('completionsdir') + BASHCOMPDIR = bashcompletion.get_variable(pkgconfig: 'completionsdir') else BASHCOMPDIR = join_paths(DATAROOTDIR, 'bash-completion/completions') endif -- 2.37.1

2 1

Fwd: [PATCH] This simple patch allows comments in the middle of a line.
by Fabiano Furtado 25 Jul '22

25 Jul '22

>From 4a87185f577d61f51c94e9b158d602460c795330 Mon Sep 17 00:00:00 2001 From: Fabiano Furtado <fabianofurtado <at> gmail <dot> com> Date: Sun, 24 Jul 2022 16:44:31 -0300 Subject: [PATCH] As described in https://archlinux.org/pacman/pacman.conf.5.html#_description, "Comments are only supported by beginning a line with the hash (#) symbol. Comments cannot begin in the middle of a line.". This patch allows comments in the middle of a line. Example: LogFile = /tmp/pacman.log#/var/log/pacman.log Signed-off-by: Fabiano Furtado <fabianofurtado <at> gmail <dot> com> --- src/common/ini.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/src/common/ini.c b/src/common/ini.c index 410a2843..4d09d844 100644 --- a/src/common/ini.c +++ b/src/common/ini.c @@ -60,14 +60,22 @@ int parse_ini(const char *file, ini_parser_fn cb, void *data) } while(safe_fgets(line, PATH_MAX, fp)) { - char *key, *value; + char *key, *value, *comment; size_t line_len; linenum++; + if(line[0] == '#') { + continue; + } + + if((comment = strchr(line,'#')) != NULL) { + *comment = '\0'; + } + line_len = strtrim(line); - if(line_len == 0 || line[0] == '#') { + if(line_len == 0) { continue; } -- 2.37.1

2 2

[PATCH] proto: Change the default checksum from md5 to sha256
by Ben Westover 22 Jul '22

22 Jul '22

MD5 isn't a very good checksum, and the PKGBUILD page on the Arch Wiki states that it should not be used, instead recommending sha256 or b2. This patch changes the default from md5 to sha256 because that seems to be the most commonly used checksum today. Signed-off-by: Ben Westover <kwestover.kw(a)gmail.com> --- proto/PKGBUILD-split.proto | 2 +- proto/PKGBUILD-vcs.proto | 2 +- proto/PKGBUILD.proto | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/proto/PKGBUILD-split.proto b/proto/PKGBUILD-split.proto index 9898ef81..eea97e56 100644 --- a/proto/PKGBUILD-split.proto +++ b/proto/PKGBUILD-split.proto @@ -28,7 +28,7 @@ changelog= source=("$pkgbase-$pkgver.tar.gz" "$pkgname-$pkgver.patch") noextract=() -md5sums=() +sha256sums=() validpgpkeys=() prepare() { diff --git a/proto/PKGBUILD-vcs.proto b/proto/PKGBUILD-vcs.proto index ae9956a9..49c6759f 100644 --- a/proto/PKGBUILD-vcs.proto +++ b/proto/PKGBUILD-vcs.proto @@ -25,7 +25,7 @@ options=() install= source=('FOLDER::VCS+URL#FRAGMENT') noextract=() -md5sums=('SKIP') +sha256sums=('SKIP') # Please refer to the 'USING VCS SOURCES' section of the PKGBUILD man page for # a description of each element in the source array. diff --git a/proto/PKGBUILD.proto b/proto/PKGBUILD.proto index a2c600d5..9aff797c 100644 --- a/proto/PKGBUILD.proto +++ b/proto/PKGBUILD.proto @@ -27,7 +27,7 @@ changelog= source=("$pkgname-$pkgver.tar.gz" "$pkgname-$pkgver.patch") noextract=() -md5sums=() +sha256sums=() validpgpkeys=() prepare() { -- 2.36.1

1 0

[PATCH v2] util: Flush cursor state to stdout before removing signal handler
by Chris Down 22 Jul '22

22 Jul '22

It's possible that the cursor does not reappear after pressing ^C during shutdown. In my case, I noticed this when pressing ^C after getting results from `pacman -F` -- this can reasonably reliably be triggered by issuing a file query and pressing ^C shortly after results are shown. There are two reasons for this issue: 1. The graceful SIGINT handler is removed at the start of cleanup(), but the window from entering cleanup() to reaching exit() is non trivial. The main offender is FREELIST(pm_targets), which on my T14s takes >0.1s to execute. This means that if you are unlucky enough to press ^C while there, the cursor isn't coming back, because we haven't issued any command to show the cursor again yet, and the userspace signal handler is already blown away. 2. Moving console_cursor_show() to earlier in cleanup() only half solves the issue. While it's fine not to flush after _hiding_ the cursor, since it will at least make itself apparent before any other text reaches the screen, _showing_ the cursor must be followed by flushing stdout, because once the graceful SIGINT handler is gone, if you press ^C, no flush will be triggered (and thus there will be no cursor). This fixes the issue by always starting out by showing the cursor again at cleanup() time. This means that no matter where we get caught at ^C, we will not end up leaving the terminal without its beloved ensign. Signed-off-by: Chris Down <chris(a)chrisdown.name> --- src/pacman/pacman.c | 2 +- src/pacman/util.c | 10 ++++++++++ 2 files changed, 11 insertions(+), 1 deletion(-) diff --git a/src/pacman/pacman.c b/src/pacman/pacman.c index ea536a47..e5c6e420 100644 --- a/src/pacman/pacman.c +++ b/src/pacman/pacman.c @@ -289,6 +289,7 @@ static void setuseragent(void) */ static void cleanup(int ret) { + console_cursor_show(); remove_soft_interrupt_handler(); if(config) { /* free alpm library resources */ @@ -302,7 +303,6 @@ static void cleanup(int ret) /* free memory */ FREELIST(pm_targets); - console_cursor_show(); exit(ret); } diff --git a/src/pacman/util.c b/src/pacman/util.c index 53833d6f..e92b2012 100644 --- a/src/pacman/util.c +++ b/src/pacman/util.c @@ -1426,6 +1426,16 @@ void console_cursor_hide(void) { void console_cursor_show(void) { if(isatty(fileno(stdout))) { printf(CURSOR_SHOW_ANSICODE); + + /* We typically explicitly show the cursor either when we are + * getting input from stdin, or when we are in the process of + * exiting. In the former case, it's not guaranteed that the + * terminal will see the command before reading from stdin. In + * the latter case, we need to make sure that if we get a + * further TERM/INT after we return signal disposition to + * SIG_DFL, it doesn't leave the cursor invisible. + */ + fflush(stdout); } } -- 2.35.1

2 4

[Git][pacman/pacman][master] util: Flush cursor state to stdout before removing signal handler
by Allan McRae (＠allan) 22 Jul '22

22 Jul '22

Allan McRae pushed to branch master at Pacman / Pacman Commits: ff7c6c8e by Chris Down at 2022-07-22T10:20:55+10:00 util: Flush cursor state to stdout before removing signal handler It's possible that the cursor does not reappear after pressing ^C during shutdown. In my case, I noticed this when pressing ^C after getting results from `pacman -F` -- this can reasonably reliably be triggered by issuing a file query and pressing ^C shortly after results are shown. There are two reasons for this issue: 1. The graceful SIGINT handler is removed at the start of cleanup(), but the window from entering cleanup() to reaching exit() is non trivial. The main offender is FREELIST(pm_targets), which on my T14s takes >0.1s to execute. This means that if you are unlucky enough to press ^C while there, the cursor isn't coming back, because we haven't issued any command to show the cursor again yet, and the userspace signal handler is already blown away. 2. Moving console_cursor_show() to earlier in cleanup() only half solves the issue. While it's fine not to flush after _hiding_ the cursor, since it will at least make itself apparent before any other text reaches the screen, _showing_ the cursor must be followed by flushing stdout, because once the graceful SIGINT handler is gone, if you press ^C, no flush will be triggered (and thus there will be no cursor). This fixes the issue by always starting out by showing the cursor again at cleanup() time. This means that no matter where we get caught at ^C, we will not end up leaving the terminal without its beloved ensign. Signed-off-by: Chris Down <chris(a)chrisdown.name> Signed-off-by: Allan McRae <allan(a)archlinux.org> - - - - - 2 changed files: - src/pacman/pacman.c - src/pacman/util.c View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/commit/ff7c6c8e570bcf5e0787639… -- View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/commit/ff7c6c8e570bcf5e0787639… You're receiving this email because of your account on gitlab.archlinux.org.

1 0

[Git][pacman/pacman][master] 8 commits: add extended data field for arbitrary package data
by Allan McRae (＠allan) 21 Jul '22

21 Jul '22

Allan McRae pushed to branch master at Pacman / Pacman Commits: 3405709b by Andrew Gregory at 2022-07-21T19:01:27+10:00 add extended data field for arbitrary package data This adds a mechanism for package builders to add arbitrary data to packages that is not necessarily relevant enough to the package installation process to gain first-class support in alpm. Currently these fields have to be added to parsers with a "not actually used" comment and can't be retrieved through the API. Extended data is stored in "name=value" format in the xdata field (%XDATA% in desc files): xdata = pkgtype=debug or %XDATA% pkgtype=debug Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> - - - - - fe028c7f by Andrew Gregory at 2022-07-21T19:01:38+10:00 query: print extended data when extra info is requested Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> - - - - - 6dcd4b53 by Andrew Gregory at 2022-07-21T19:01:45+10:00 makepkg: store pkgtype in xdata Package type is not relevant to alpm or even exposed to front-ends in any way. Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> - - - - - 819a0c29 by Jelle van der Waa at 2022-07-21T19:09:59+10:00 util.c: extend --print-format with expac options Extend --print-format with all expac format strings which can be easily added without conversions and through a simple C macro. Signed-off-by: Jelle van der Waa <jelle(a)archlinux.org> Signed-off-by: Allan McRae <allan(a)archlinux.org> - - - - - 7bcc2d9b by Allan McRae at 2022-07-21T19:50:10+10:00 Correctly handle failure in getting build or install dates alpm_pkg_get_builddate() and alpm_pkg_get_installdate() both return -1 on error. Correctly handle the error condition in pacman. Signed-off-by: Allan McRae <allan(a)archlinux.org> - - - - - 00bc386d by Jelle van der Waa at 2022-07-21T19:56:18+10:00 util.c: extend --print-format with %b for builddate Signed-off-by: Jelle van der Waa <jelle(a)archlinux.org> Signed-off-by: Allan McRae <allan(a)archlinux.org> - - - - - cb9776a0 by Jelle van der Waa at 2022-07-21T19:58:20+10:00 util.c: extend --print-format with %C, %D, %M Extend print-format with checkdepends, depends and makedepends. Signed-off-by: Jelle van der Waa <jelle(a)archlinux.org> Signed-off-by: Allan McRae <allan(a)archlinux.org> - - - - - 015eb31c by Chris Down at 2022-07-21T20:00:44+10:00 dload: Remove unused ABORT_SIGINT The last user of ABORT_SIGINT was removed in commit 84723cab5dfc ("Cleanup the old sequential download code"), and this isn't exported as part of the public API. Signed-off-by: Chris Down <chris(a)chrisdown.name> Signed-off-by: Allan McRae <allan(a)archlinux.org> - - - - - 11 changed files: - doc/pacman.8.asciidoc - lib/libalpm/alpm.h - lib/libalpm/be_local.c - lib/libalpm/be_package.c - lib/libalpm/be_sync.c - lib/libalpm/dload.c - lib/libalpm/package.c - lib/libalpm/package.h - scripts/makepkg.sh.in - src/pacman/package.c - src/pacman/util.c View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/compare/2bf67416c17c99851e2d0b… -- View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/compare/2bf67416c17c99851e2d0b… You're receiving this email because of your account on gitlab.archlinux.org.

1 0

[PATCH] Fix buffer overflow for 'Include' parameter in pacman.conf
by tim174 21 Jul '22

21 Jul '22

If the 'Include' parameter in the config file is set to a long string (for example 3000x '/') the pacman config parser will crash when calling glob in conf.c with this value. If the string is shorter than approx. 3000 symbols the normal error message is printed and the segfault does not occur. I was able to reproduce this on my own system and on the official Arch Linux iso image. The PATH_MAX variable is too large to prevent this bug, hence the new variable GLOB_LIMIT with a security buffer of 1000 since I don't know how persistent this limit is across systems. Its origin is unclear to me and I am not sure if it is a fixed value. This is why I would appreciate any help to make this a sustainable patch. Thanks Tim Signed-off-by: Tim <tim174 at protonmail.com> --- src/pacman/conf.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/src/pacman/conf.c b/src/pacman/conf.c index f9edf75b..d8222752 100644 --- a/src/pacman/conf.c +++ b/src/pacman/conf.c @@ -65,6 +65,9 @@ config_t *config = NULL; #define BOLDWHITE "\033[1;37m" #define GREY46 "\033[38;5;243m" +/* limit for glob input variable */ +#define GLOB_LIMIT 2000 + void enable_colors(int colors) { colstr_t *colstr = &config->colstr; @@ -1042,6 +1045,13 @@ static int process_include(const char *value, void *data, return 1; } + if(strlen(value) > GLOB_LIMIT) { + pm_printf(ALPM_LOG_ERROR, + ("config file %s, line %d, directive '%s': value too long\n"), + file, linenum, "Include"); + return 1; + } + if(section->depth >= config_max_recursion) { pm_printf(ALPM_LOG_ERROR, _("config parsing exceeded max recursion depth of %d.\n"), -- 2.34.1

2 1

[PATCH] dload: Remove unused ABORT_SIGINT
by Chris Down 21 Jul '22

21 Jul '22

The last user of ABORT_SIGINT was removed in commit 84723cab5dfc ("Cleanup the old sequential download code"), and this isn't exported as part of the public API. Signed-off-by: Chris Down <chris(a)chrisdown.name> --- lib/libalpm/dload.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/lib/libalpm/dload.c b/lib/libalpm/dload.c index 7c27c3ea..4fa17b35 100644 --- a/lib/libalpm/dload.c +++ b/lib/libalpm/dload.c @@ -153,8 +153,7 @@ static char *get_fullpath(const char *path, const char *filename, } enum { - ABORT_SIGINT = 1, - ABORT_OVER_MAXFILESIZE + ABORT_OVER_MAXFILESIZE = 1, }; static int dload_interrupted; -- 2.35.1

2 1

Extend print-format with expac options
by Jelle van der Waa 21 Jul '22

21 Jul '22

This patch series extends --print-format with a subset of the expac options. More options to be added in further patch series.

3 8

[PATCH] Correctly handle failure in getting build or install dates
by Allan McRae 21 Jul '22

21 Jul '22

alpm_pkg_get_builddate() and alpm_pkg_get_installdate() both return -1 on error. Correctly handle the error condition in pacman. Signed-off-by: Allan McRae <allan(a)archlinux.org> --- src/pacman/package.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/pacman/package.c b/src/pacman/package.c index f2b6fa19..4fcc11f2 100644 --- a/src/pacman/package.c +++ b/src/pacman/package.c @@ -210,11 +210,11 @@ void dump_pkg_full(alpm_pkg_t *pkg, int extra) /* set variables here, do all output below */ bdate = (time_t)alpm_pkg_get_builddate(pkg); - if(bdate) { + if(bdate != -1) { strftime(bdatestr, 50, "%c", localtime(&bdate)); } idate = (time_t)alpm_pkg_get_installdate(pkg); - if(idate) { + if(idate != -1) { strftime(idatestr, 50, "%c", localtime(&idate)); } -- 2.37.0

1 0

[RFC][WIP][PATCH 0/5] adding extended package data
by Andrew Gregory 21 Jul '22

21 Jul '22

ALPM is currently parsing several fields that are not particularly relevant to managing package installation (e.g. pkgbase, pkgtype, makedepends, checkdepends). pkgtype in particular is not even exposed in any way through the API, it has to be included in the parser with a "not actually used" comment. This patchset adds an extended data field to packages that can be used by packagers or package-building infrastructure to store arbitrary package data in a way that doesn't require altering alpm for every field and allows the data to actually be retrieved via alpm (though not as easily as first-class fields). Andrew Gregory (5): query: only strip leading local/ for db packages query: allow querying extra info for package files add data field for arbitrary package data query: print extended data when extra info is requested makepkg: store package type in extended data lib/libalpm/alpm.h | 7 +++++++ lib/libalpm/be_local.c | 25 +++++++++++++++++++++++++ lib/libalpm/be_package.c | 6 ++++++ lib/libalpm/be_sync.c | 16 ++++++++++++++++ lib/libalpm/package.c | 17 +++++++++++++++++ lib/libalpm/package.h | 6 ++++++ scripts/makepkg.sh.in | 2 +- src/pacman/package.c | 13 +++++++++++++ src/pacman/query.c | 16 ++++++---------- 9 files changed, 97 insertions(+), 11 deletions(-) -- 2.35.0

2 13

makedepends field of alpm_pkg_t
by Colin Woodbury 16 Jul '22

16 Jul '22

Hi guys, I have a brief question about the `makedepends` field of the `alpm_pkg_t` type. Try as I might, I'm unable to get this field to report anything (i.e. a non-empty alpm_list) for installed packages originally from the AUR that definitely had makedepends. This seems to be the case even when those makedepends are still installed. Under what circumstances should this field be expected to yield results? To be clear, I'm accessing it via the top-level `_pkg_get_makedepends` function. Thanks, Colin

2 2

[PATCH] libmakepkg/integrity: handle PGP signature files containing multiple signatures
by Jonas Witschel 26 Jun '22

26 Jun '22

PGP signature files can contain multiple signatures from different keys. In this case, verification is considered to be successful if *all* signatures are valid, compare e.g. the behaviour of the gpgv verification tool. parse_gpg_statusfile() is currently not equipped to handle the gpg output for such signature files: it assumes that the file contains only one signature and therefore overwrites the verification results of previous signatures, effectively only considering the last signature in the file. Therefore the logic needs to be refined so that $success and $trusted are only true if all signatures are valid and trusted, respectively. We also need to store the verification results per key (in order to present more specific error messages) and all the validated fingerprints (in order to check whether they are in validpgpkeys). The signatures for the source tarballs of recent GnuPG releases, e.g. version 2.2.35, are examples of such signature files. Signed-off-by: Jonas Witschel <diabonas(a)archlinux.org> --- .../integrity/verify_signature.sh.in | 115 ++++++++++-------- 1 file changed, 65 insertions(+), 50 deletions(-) diff --git a/scripts/libmakepkg/integrity/verify_signature.sh.in b/scripts/libmakepkg/integrity/verify_signature.sh.in index ad5bb66d..4aa14d94 100644 --- a/scripts/libmakepkg/integrity/verify_signature.sh.in +++ b/scripts/libmakepkg/integrity/verify_signature.sh.in @@ -26,6 +26,18 @@ LIBRARY=${LIBRARY:-'@libmakepkgdir@'} source "$LIBRARY/util/message.sh" source "$LIBRARY/util/pkgbuild.sh" +# Returns the elements of the first array that are not found in the second +# array as the third array. All arrays are passed by name. +set_difference() { + local -n first=$1 + local -n second=$2 + local -n result=$3 + result=() + for element in "${first[@]}"; do + ! in_array "$element" "${second[@]}" && result+=("$element") + done +} + check_pgpsigs() { (( SKIPPGPCHECK )) && return 0 ! source_has_signatures && return 0 @@ -56,48 +68,53 @@ check_pgpsigs() { fi # these variables are assigned values in parse_gpg_statusfile - success=0 - status= - pubkey= - fingerprint= + success= + declare -A status + fingerprints=() trusted= parse_gpg_statusfile "$statusfile" if (( ! $success )); then printf '%s' "$(gettext "FAILED")" >&2 - case "$status" in - "missingkey") - printf ' (%s)' "$(gettext "unknown public key") $pubkey" >&2 - ;; - "revokedkey") - printf " ($(gettext "public key %s has been revoked"))" "$pubkey" >&2 - ;; - "bad") - printf ' (%s)' "$(gettext "bad signature from public key") $pubkey" >&2 - ;; - "error") - printf ' (%s)' "$(gettext "error during signature verification")" >&2 - ;; - esac + for pubkey in "${!status[@]}"; do + case "${status["$pubkey"]}" in + "missingkey") + printf ' (%s)' "$(gettext "unknown public key") $pubkey" >&2 + ;; + "revokedkey") + printf " ($(gettext "public key %s has been revoked"))" "$pubkey" >&2 + ;; + "bad") + printf ' (%s)' "$(gettext "bad signature from public key") $pubkey" >&2 + ;; + "error") + printf ' (%s)' "$(gettext "error during signature verification")" >&2 + ;; + esac + done errors=1 else + local invalid_keys + set_difference fingerprints validpgpkeys invalid_keys if (( ${#validpgpkeys[@]} == 0 && !trusted )); then - printf "%s ($(gettext "the public key %s is not trusted"))" $(gettext "FAILED") "$fingerprint" >&2 + printf "%s ($(gettext "the public key %s is not trusted"))" $(gettext "FAILED") "${fingerprints[*]}" >&2 errors=1 - elif (( ${#validpgpkeys[@]} > 0 )) && ! in_array "$fingerprint" "${validpgpkeys[@]}"; then - printf "%s (%s %s)" "$(gettext "FAILED")" "$(gettext "invalid public key")" "$fingerprint" >&2 + elif (( ${#validpgpkeys[@]} > 0 )) && (( ${#invalid_keys[@]} > 0 )); then + printf "%s (%s %s)" "$(gettext "FAILED")" "$(gettext "invalid public key")" "${invalid_keys[*]}" >&2 errors=1 else printf '%s' "$(gettext "Passed")" >&2 - case "$status" in - "expired") - printf ' (%s)' "$(gettext "WARNING:") $(gettext "the signature has expired.")" >&2 - warnings=1 - ;; - "expiredkey") - printf ' (%s)' "$(gettext "WARNING:") $(gettext "the key has expired.")" >&2 - warnings=1 - ;; - esac + for pubkey in "${!status[@]}"; do + case "${status["$pubkey"]}" in + "expired") + printf " (%s $(gettext "the signature by key %s has expired."))" "$(gettext "WARNING:")" "$pubkey" >&2 + warnings=1 + ;; + "expiredkey") + printf " (%s $(gettext "the key %s has expired."))" "$(gettext "WARNING:")" "$pubkey" >&2 + warnings=1 + ;; + esac + done fi fi printf '\n' >&2 @@ -201,59 +218,57 @@ verify_git_signature() { parse_gpg_statusfile() { local type arg1 arg6 arg10 + success=-1 + trusted=-1 while read -r _ type arg1 _ _ _ _ arg6 _ _ _ arg10 _; do case "$type" in GOODSIG) - pubkey=$arg1 - success=1 - status="good" + success=$((success!=0)) + status["$arg1"]="good" ;; EXPSIG) - pubkey=$arg1 - success=1 - status="expired" + success=$((success!=0)) + status["$arg1"]="expired" ;; EXPKEYSIG) - pubkey=$arg1 - success=1 - status="expiredkey" + success=$((success!=0)) + status["$arg1"]="expiredkey" ;; REVKEYSIG) - pubkey=$arg1 success=0 - status="revokedkey" + status["$arg1"]="revokedkey" ;; BADSIG) - pubkey=$arg1 success=0 - status="bad" + status["$arg1"]="bad" ;; ERRSIG) - pubkey=$arg1 success=0 if [[ $arg6 == 9 ]]; then - status="missingkey" + status["$arg1"]="missingkey" else - status="error" + status["$arg1"]="error" fi ;; VALIDSIG) if [[ $arg10 ]]; then # If the file was signed with a subkey, arg10 contains # the fingerprint of the primary key - fingerprint=$arg10 + fingerprints+=("$arg10") else - fingerprint=$arg1 + fingerprints+=("$arg1") fi ;; TRUST_UNDEFINED|TRUST_NEVER) trusted=0 ;; TRUST_MARGINAL|TRUST_FULLY|TRUST_ULTIMATE) - trusted=1 + trusted=$((trusted!=0)) ;; esac done < "$1" + success=$((success>0)) + trusted=$((trusted>0)) } source_has_signatures() { -- 2.36.1

3 8

[PATCH] makepkg: Implement the verify function
by Morten Linderud 25 Jun '22

25 Jun '22

From: Morten Linderud <morten(a)linderud.pw> This patch implements a new verify function in makepkg. It allows us to do arbitrary authentication on sources before extraction. There are several new signing and validation methods being implemented and it would be hard to have `makepkg` implement support for things such as sequoia, cosign or minisign. This would allow us to distribute generic validation functions. This also implements a new `copy_` routine for our protocols as we need to have a separation between extracting sources and copying sources. Signed-off-by: Morten Linderud <morten(a)linderud.pw> --- doc/PKGBUILD.5.asciidoc | 4 ++++ doc/makepkg.8.asciidoc | 3 +++ scripts/libmakepkg/integrity.sh.in | 3 +++ scripts/libmakepkg/source.sh.in | 15 +++++++++++++++ scripts/libmakepkg/source/file.sh.in | 9 ++++++++- scripts/makepkg.sh.in | 17 ++++++++++++++++- 6 files changed, 49 insertions(+), 2 deletions(-) diff --git a/doc/PKGBUILD.5.asciidoc b/doc/PKGBUILD.5.asciidoc index 4ca8eb3b..e7743c88 100644 --- a/doc/PKGBUILD.5.asciidoc +++ b/doc/PKGBUILD.5.asciidoc @@ -344,6 +344,10 @@ function. fakeroot to ensure correct file permissions in the resulting package. All other functions will be run as the user calling makepkg. +*verify() Function*:: + An optional `verify()` function can be specified to implement arbiterary + source authentication. This function is run before sources are extracted. + *prepare() Function*:: An optional `prepare()` function can be specified in which operations to prepare the sources for building, such as patching, are performed. This diff --git a/doc/makepkg.8.asciidoc b/doc/makepkg.8.asciidoc index 38032e7b..75b2139f 100644 --- a/doc/makepkg.8.asciidoc +++ b/doc/makepkg.8.asciidoc @@ -168,6 +168,9 @@ Options *\--noprepare*:: Do not run the prepare() function in the PKGBUILD. +*\--noverify*:: + Do not run the verify() function in the PKGBUILD. + *\--sign*:: Sign the resulting package with gpg, overriding the setting in linkman:makepkg.conf[5]. diff --git a/scripts/libmakepkg/integrity.sh.in b/scripts/libmakepkg/integrity.sh.in index 070392fa..81f935df 100644 --- a/scripts/libmakepkg/integrity.sh.in +++ b/scripts/libmakepkg/integrity.sh.in @@ -42,4 +42,7 @@ check_source_integrity() { check_checksums "$@" check_pgpsigs "$@" fi + if (( VERIFYFUNC )); then + run_verify + fi } diff --git a/scripts/libmakepkg/source.sh.in b/scripts/libmakepkg/source.sh.in index e39dd16c..92dc71e4 100644 --- a/scripts/libmakepkg/source.sh.in +++ b/scripts/libmakepkg/source.sh.in @@ -69,6 +69,21 @@ download_sources() { done } +copy_sources(){ + msg "$(gettext "Copying sources...")" + local netfile all_sources + + get_all_sources_for_arch 'all_sources' + for netfile in "${all_sources[@]}"; do + local proto=$(get_protocol "$netfile") + if declare -f copy_$proto > /dev/null; then + copy_$proto "$netfile" + else + copy_file "$netfile" + fi + done +} + extract_sources() { msg "$(gettext "Extracting sources...")" local netfile all_sources diff --git a/scripts/libmakepkg/source/file.sh.in b/scripts/libmakepkg/source/file.sh.in index fa09d446..51452550 100644 --- a/scripts/libmakepkg/source/file.sh.in +++ b/scripts/libmakepkg/source/file.sh.in @@ -82,13 +82,20 @@ download_file() { fi } -extract_file() { +copy_file(){ local netfile=$1 local file=$(get_filename "$netfile") local filepath=$(get_filepath "$file") rm -f "$srcdir/${file}" ln -s "$filepath" "$srcdir/" +} + +extract_file() { + local netfile=$1 + + local file=$(get_filename "$netfile") + local filepath=$(get_filepath "$file") if in_array "$file" "${noextract[@]}"; then # skip source files in the noextract=() array diff --git a/scripts/makepkg.sh.in b/scripts/makepkg.sh.in index 5aaabf63..b7b21af1 100644 --- a/scripts/makepkg.sh.in +++ b/scripts/makepkg.sh.in @@ -144,6 +144,9 @@ clean_up() { if (( PKGVERFUNC )); then rm -f "${pkgbase}-${fullver}-${CARCH}-pkgver.log"* fi + if (( VERIFYFUNC )); then + rm -f "${pkgbase}-${fullver}-${CARCH}-verify.log"* + fi if (( PREPAREFUNC )); then rm -f "${pkgbase}-${fullver}-${CARCH}-prepare.log"* fi @@ -447,6 +450,10 @@ run_function() { fi } +run_verify() { + run_function_safe "verify" +} + run_prepare() { run_function_safe "prepare" } @@ -973,6 +980,7 @@ while true; do -m|--nocolor) USE_COLOR='n'; PACMAN_OPTS+=("--color" "never") ;; --noarchive) NOARCHIVE=1 ;; --nocheck) RUN_CHECK='n' ;; + --noverify) RUN_VERIFY='n' ;; --noprepare) RUN_PREPARE='n' ;; --nosign) SIGNPKG='n' ;; -o|--nobuild) BUILDPKG=0 NOBUILD=1 ;; @@ -1093,7 +1101,7 @@ fi unset pkgname "${pkgbuild_schema_strings[@]}" "${pkgbuild_schema_arrays[@]}" unset "${known_hash_algos[@]/%/sums}" -unset -f pkgver prepare build check package "${!package_@}" +unset -f pkgver verify prepare build check package "${!package_@}" unset "${!makedepends_@}" "${!depends_@}" "${!source_@}" "${!checkdepends_@}" unset "${!optdepends_@}" "${!conflicts_@}" "${!provides_@}" "${!replaces_@}" unset "${!cksums_@}" "${!md5sums_@}" "${!sha1sums_@}" "${!sha224sums_@}" @@ -1165,6 +1173,12 @@ if (( ${#pkgname[@]} > 1 )) || have_function package_${pkgname}; then fi # test for available PKGBUILD functions +if have_function verify; then + # "Hide" verify() function if not going to be run + if [[ $RUN_VERIFY != "n" ]]; then + VERIFYFUNC=1 + fi +fi if have_function prepare; then # "Hide" prepare() function if not going to be run if [[ $RUN_PREPARE != "n" ]]; then @@ -1312,6 +1326,7 @@ if (( !REPKG )); then warning "$(gettext "Using existing %s tree")" "\$srcdir/" else download_sources + copy_sources check_source_integrity (( VERIFYSOURCE )) && exit $E_OK -- 2.36.1

2 3

[Git][pacman/pacman][master] Fix remaining email list address
by Allan McRae (＠allan) 25 Jun '22

25 Jun '22

Allan McRae pushed to branch master at Pacman / Pacman Commits: 2bf67416 by Allan McRae at 2022-06-25T22:59:20+10:00 Fix remaining email list address One instance was missed in b0a2fd75b2f4ba12075646f0f3f18911d71a42c7. Signed-off-by: Allan McRae <allan(a)archlinux.org> - - - - - 1 changed file: - lib/libalpm/po/Makevars View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/commit/2bf67416c17c99851e2d0bb… -- View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/commit/2bf67416c17c99851e2d0bb… You're receiving this email because of your account on gitlab.archlinux.org.

1 0

[Git][pacman/pacman][master] 8 commits: makepkg: use bsdtar --no-read-sparse for archive creation if available
by Allan McRae (＠allan) 25 Jun '22

25 Jun '22

Allan McRae pushed to branch master at Pacman / Pacman Commits: f5f9e0f4 by Jonas Witschel at 2022-03-11T09:43:59+10:00 makepkg: use bsdtar --no-read-sparse for archive creation if available bsdtar uses the "pax" TAR archive format by default, which has support for storing sparse file information in the archive. Unfortunately this is a source of unreproducibility because the sparse encoding is taken from the file system and different file systems handle sparse files differently: some file systems have no support for sparsely encoded files at all, and even file systems with sparse file support can report different file information for identical files due to differing implementations. As a real world example where this happens, consider the Arch Linux package "brotli-testdata 1.0.9-7", which contains a sparsely encoded all-zeros file "usr/share/brotli/testdata/zeros". Building this package on a btrfs file system yields a different package than building it on tmpfs or ext4 solely due to different sparse file information that gets recorded in the package tarball. To improve the reproducibility of archives containing sparsely encoded files, libarchive version 3.6.0 introduces a new --no-read-sparse option. This skips reading sparse file information from disk entirely and therefore stores files "expanded" in the archive, which is the only way to make them reliably reproducible across file systems. makepkg will use this option if libarchive is recent enough to support it, which is detected at build time. Signed-off-by: Allan McRae <allan(a)archlinux.org> - - - - - 18152a21 by Allan McRae at 2022-05-12T09:45:12+10:00 libmakepkg: Do not "strip" files when the format is not recognised We use a multi step process during stripping to ensure permissions do not get changed. However, if the initial objcopy fails, the subsequent cat results in a blank file. Abandon early if objcopy fails. Fixes FS#74486 Signed-off-by: Allan McRae <allan(a)archlinux.org> - - - - - fdf04cda by Allan McRae at 2022-05-18T22:06:07+10:00 doc/PKGBUILD: remove Arch Linux specific advise The documentation for the license array was specific to Arch Linux. Remove it and some minor other Arch Linux specific references. Signed-off-by: Allan McRae <allan(a)archlinux.org> - - - - - 84dc662d by Allan McRae at 2022-05-31T20:13:14+10:00 makepkg: add source signing PGP keys to source package if available Arch Linux is adding source signing PGP keys to their package source tree alongside PKGBUILDs in the form keys/pgp/$fingerprint.asc. As the PGP keyserver infrastructure is a mess, this helps other people validate sources in a PKGBUILD. Add the keys to source packages if found alongside the PKGBUILD. Signed-off-by: Allan McRae <allan(a)archlinux.org> - - - - - d1e95c4e by Allan McRae at 2022-06-16T09:00:18+10:00 Gitlab CI: Install glibc-debug valgrind test Currently our gitlab CI is failing due to valgrind breakage. With Arch stripping glibc, valgrind now requires debuginfod to be active. However the gitlab CI system combined without our testsuite does not retrieve these symbols, even when the appropriate environmental variable is set. Work around this by installing the glibc-debug package directly using a slight kludge... All blame for this approach is assigned to foutrelis! Signed-off-by: Allan McRae <allan(a)archlinux.org> - - - - - 0031cd1e by Levente Polyak at 2022-06-25T10:25:36+10:00 doc/BUILDINFO.5: document buildtool and buildtoolver properties The feature has been introduced in commit a33cdac10bf1a682 The buildinfo version has been bumped in commit 0428f6213bcb5586 Signed-off-by: Levente Polyak <anthraxx(a)archlinux.org> - - - - - 776b7c1e by Morten Linderud at 2022-06-25T10:45:04+10:00 debugflags: Ensure we have unique source paths In some cases packages are built outside of a directory which contains pkgname-pkgver, this results in source listing in debug packages having a conflicting path like `/usr/src/debug/build/` which is not ideal. This patch ensures we always include the pkgbase to ensure the paths are unique. Signed-off-by: Morten Linderud <morten(a)linderud.pw> Signed-off-by: Allan McRae <allan(a)archlinux.org> - - - - - 6daeb37f by Jean Raby at 2022-06-25T22:45:00+10:00 pacman: silence "info" msg when using --print This allows for parsing the output of: pacman --upgrade --print-format '<format>' pkg.zst without having to remove info messages from it. Signed-off-by: Allan McRae <allan(a)archlinux.org> - - - - - 9 changed files: - .gitlab-ci.yml - build-aux/edit-script.sh.in - doc/BUILDINFO.5.asciidoc - doc/PKGBUILD.5.asciidoc - meson.build - scripts/libmakepkg/buildenv/debugflags.sh.in - scripts/libmakepkg/tidy/strip.sh.in - scripts/makepkg.sh.in - src/pacman/upgrade.c View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/compare/bddfcc3f40ce1a19d4c955… -- View it on GitLab: https://gitlab.archlinux.org/pacman/pacman/-/compare/bddfcc3f40ce1a19d4c955… You're receiving this email because of your account on gitlab.archlinux.org.

1 0

[PATCH] pacman: send 'loading packages...' msg to stderr
by Jean Raby 25 Jun '22

25 Jun '22

This allows for parsing the output of: pacman --upgrade --print-format '<format>' pkg.zst without having to remove info messages from it. Signed-off-by: Jean Raby <jean(a)raby.sh> --- src/pacman/upgrade.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/pacman/upgrade.c b/src/pacman/upgrade.c index 0691856f..6f70b115 100644 --- a/src/pacman/upgrade.c +++ b/src/pacman/upgrade.c @@ -99,7 +99,7 @@ int pacman_upgrade(alpm_list_t *targets) goto fail_free; } - printf(_("loading packages...\n")); + fprintf(stderr,_("loading packages...\n")); retval |= load_packages(local_targets, alpm_option_get_local_file_siglevel(config->handle)); retval |= load_packages(fetched_files, alpm_option_get_remote_file_siglevel(config->handle)); -- 2.35.1

2 3

[PATCH v2] strip: Use dwz to compress dwarf headers
by Morten Linderud 25 Jun '22

25 Jun '22

From: Morten Linderud <morten(a)linderud.pw> `dwz` allows us to compress the DWARF files when building debug packages. Potentially shaving off some of the sizes of the symbols we distribute. https://sourceware.org/dwz/ A sample of packages built with dwz; pacman: Original debug info size: 1520kB Size after compression: 1252kB systemd: Original debug info size: 46692kB Size after compression: 41036kB Signed-off-by: Morten Linderud <morten(a)linderud.pw> --- doc/makepkg.conf.5.asciidoc | 3 ++ scripts/libmakepkg/executable/dwz.sh.in | 38 ++++++++++++++++ .../executable/sepdebugcrcfix.sh.in | 38 ++++++++++++++++ scripts/libmakepkg/tidy/strip.sh.in | 44 +++++++++++++++++-- 4 files changed, 120 insertions(+), 3 deletions(-) create mode 100644 scripts/libmakepkg/executable/dwz.sh.in create mode 100644 scripts/libmakepkg/executable/sepdebugcrcfix.sh.in diff --git a/doc/makepkg.conf.5.asciidoc b/doc/makepkg.conf.5.asciidoc index a0d9a6d4..2e40f27f 100644 --- a/doc/makepkg.conf.5.asciidoc +++ b/doc/makepkg.conf.5.asciidoc @@ -194,6 +194,9 @@ Options DEBUG_CXXFLAGS to their counterpart buildflags. Creates a separate package containing the debug symbols when used with `strip'. + *dwz*;; + Compress dwarf files inside the debug package. + *lto*;; Enable building packages using link time optimization. Adds the flags specified in LTOFLAGS to CFLAGS, CXXFLAGS and LDFLAGS (or diff --git a/scripts/libmakepkg/executable/dwz.sh.in b/scripts/libmakepkg/executable/dwz.sh.in new file mode 100644 index 00000000..a540697c --- /dev/null +++ b/scripts/libmakepkg/executable/dwz.sh.in @@ -0,0 +1,38 @@ +#!/usr/bin/bash +# +# dwz.sh - Confirm presence of dwz binary +# +# Copyright (c) 2011-2022 Pacman Development Team <pacman-dev(a)lists.archlinux.org> +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program. If not, see <http://www.gnu.org/licenses/>. +# + +[[ -n "$LIBMAKEPKG_EXECUTABLE_DWZ_SH" ]] && return +LIBMAKEPKG_EXECUTABLE_DWZ_SH=1 + +LIBRARY=${LIBRARY:-'@libmakepkgdir@'} + +source "$LIBRARY/util/message.sh" +source "$LIBRARY/util/option.sh" + +executable_functions+=('executable_dwz') + +executable_dwz() { + if check_option "dwz" "y"; then + if ! type -p dwz>/dev/null; then + error "$(gettext "Cannot find the %s binary required to compress dwarf files.")" "dwz" + return 1 + fi + fi +} diff --git a/scripts/libmakepkg/executable/sepdebugcrcfix.sh.in b/scripts/libmakepkg/executable/sepdebugcrcfix.sh.in new file mode 100644 index 00000000..6259ac32 --- /dev/null +++ b/scripts/libmakepkg/executable/sepdebugcrcfix.sh.in @@ -0,0 +1,38 @@ +#!/usr/bin/bash +# +# sepdebugcrcfix.sh - Confirm presence of sepdebugcrcfix binary +# +# Copyright (c) 2011-2022 Pacman Development Team <pacman-dev(a)lists.archlinux.org> +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program. If not, see <http://www.gnu.org/licenses/>. +# + +[[ -n "$LIBMAKEPKG_EXECUTABLE_SEPDEBUGCRCFIX_SH" ]] && return +LIBMAKEPKG_EXECUTABLE_SEPDEBUGCRCFIX_SH=1 + +LIBRARY=${LIBRARY:-'@libmakepkgdir@'} + +source "$LIBRARY/util/message.sh" +source "$LIBRARY/util/option.sh" + +executable_functions+=('executable_sepdebugcrcfix') + +executable_sepdebugcrcfix() { + if check_option "dwz" "y"; then + if ! type -p sepdebugcrcfix>/dev/null; then + error "$(gettext "Cannot find the %s binary required to fix CRC.")" "sepdebugcrcfix" + return 1 + fi + fi +} diff --git a/scripts/libmakepkg/tidy/strip.sh.in b/scripts/libmakepkg/tidy/strip.sh.in index 688bcf1b..1365b77b 100644 --- a/scripts/libmakepkg/tidy/strip.sh.in +++ b/scripts/libmakepkg/tidy/strip.sh.in @@ -27,7 +27,7 @@ source "$LIBRARY/util/message.sh" source "$LIBRARY/util/option.sh" -packaging_options+=('strip' 'debug') +packaging_options+=('strip' 'debug' 'dwz') tidy_modify+=('tidy_strip') @@ -52,6 +52,37 @@ source_files() { | sort -zu | tr '\0' '\n' } +compress_dwarf_files(){ + binary_files="$@" + dwz_opts="" + readarray dwz_files < <(find "$dbgdir" -type f -name \*.debug | LC_ALL=C sort) + + # dwz multifile attempts to remove common symols found in all ELF files and + # throw them into one file. Replacing the references with .gnu_debugaltlink + local fullver=$(get_full_version) + dwz_multifile_name="${pkgbase}-${fullver}-${CARCH}" + if [ ${#dwz_files[@]} -gt 1 ]; then + dwz_opts="-m .dwz/${dwz_multifile_name}" + fi + mkdir -p "$dbgdir/.dwz" + + # The dwz options are taken from rpm/debugedit + (cd "$dbgdir" && + LANG=C dwz --hardlink \ + --quiet \ + --relative \ + --low-mem-die-limit 10000000 \ + --max-die-limit 50000000 \ + $dwz_opts \ + ${dwz_files[@]} 2> /dev/null) + + # Remove dir if empty + rmdir "$dbgdir/.dwz" 2> /dev/null + + # Some debuggers might use the CRC to find relevant debug files instead of the build-id + LANG=C sepdebugcrcfix "$dbgdir" $binary_files &>/dev/null +} + strip_file() { local binary=$1; shift @@ -142,7 +173,8 @@ tidy_strip() { fi local binary strip_flags - find . -type f -perm -u+w -print0 2>/dev/null | while IFS= read -rd '' binary ; do + declare -a binary_files + while IFS= read -rd '' binary ; do local STRIPLTO=0 case "$(LC_ALL=C readelf -h "$binary" 2>/dev/null)" in *Type:*'DYN (Shared object file)'*) # Libraries (.so) or Relocatable binaries @@ -166,6 +198,12 @@ tidy_strip() { esac strip_file "$binary" ${strip_flags} (( STRIPLTO )) && strip_lto "$binary" - done + binary_files+=("$binary") + done < <(find . -type f -perm -u+w -print0 2>/dev/null) + + if check_option "debug" "y" && check_option "dwz" "y"; then + msg2 "$(gettext "Compressing dwarf files...")" + compress_dwarf_files ${binary_files[@]} + fi fi } -- 2.36.1

3 3

[pacman-dev] [PATCH 1/3] alpm: add ALPM_TRANS_FLAG_NOKEEP
by morganamilo 25 Jun '22

25 Jun '22

this flag prevents backup files from being kept on package installation. This is useful for resetting a package's config files back to their original state. Implements FS#59908 although with it's own flag name instead of reusing nosave. This allows nokeep to optionally create a pacnew that you can then choose to disable by also setting nosave. --- I actually very dislike NOKEEP but it was the best I could come up with I would have prefered overwrite or nosave but they are taken. Better names are welcome. --- lib/libalpm/alpm.h | 3 ++- lib/libalpm/remove.c | 4 +++- 2 files changed, 5 insertions(+), 2 deletions(-) diff --git a/lib/libalpm/alpm.h b/lib/libalpm/alpm.h index 8d8fe243..c6048b63 100644 --- a/lib/libalpm/alpm.h +++ b/lib/libalpm/alpm.h @@ -2704,7 +2704,8 @@ int alpm_pkg_mtree_close(const alpm_pkg_t *pkg, struct archive *archive); typedef enum _alpm_transflag_t { /** Ignore dependency checks. */ ALPM_TRANS_FLAG_NODEPS = 1, - /* (1 << 1) flag can go here */ + /** Don't keep backup files when installing packages. */ + ALPM_TRANS_FLAG_NOKEEP = (1 << 1), /** Delete files even if they are tagged as backup. */ ALPM_TRANS_FLAG_NOSAVE = (1 << 2), /** Ignore version numbers when checking dependencies. */ diff --git a/lib/libalpm/remove.c b/lib/libalpm/remove.c index de39724a..233fff0c 100644 --- a/lib/libalpm/remove.c +++ b/lib/libalpm/remove.c @@ -575,7 +575,9 @@ static int should_skip_file(alpm_handle_t *handle, { return _alpm_fnmatch_patterns(handle->noupgrade, path) == 0 || alpm_list_find_str(handle->trans->skip_remove, path) - || (newpkg && _alpm_needbackup(path, newpkg) + || (!(handle->trans->flags & ALPM_TRANS_FLAG_NOKEEP) + && newpkg + && _alpm_needbackup(path, newpkg) && alpm_filelist_contains(alpm_pkg_get_files(newpkg), path)); } -- 2.33.0

2 3

[PATCH] debugflags: Ensure we have unique source paths
by Morten Linderud 25 Jun '22

25 Jun '22

From: Morten Linderud <morten(a)linderud.pw> In some cases packages are built outside of a directory which contains pkgname-pkgver, this results in source listing in debug packages having a conflicting path like `/usr/src/debug/build/` which is not ideal. This patch ensures we always include the pkgbase to ensure the paths are unique. Signed-off-by: Morten Linderud <morten(a)linderud.pw> --- scripts/libmakepkg/buildenv/debugflags.sh.in | 6 +++--- scripts/libmakepkg/tidy/strip.sh.in | 2 +- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/scripts/libmakepkg/buildenv/debugflags.sh.in b/scripts/libmakepkg/buildenv/debugflags.sh.in index 1cb58613..2bdb8b0b 100644 --- a/scripts/libmakepkg/buildenv/debugflags.sh.in +++ b/scripts/libmakepkg/buildenv/debugflags.sh.in @@ -30,9 +30,9 @@ buildenv_functions+=('buildenv_debugflags') buildenv_debugflags() { if check_option "debug" "y" && ! check_option "buildflags" "n"; then - DEBUG_CFLAGS+=" -ffile-prefix-map=$srcdir=${DBGSRCDIR:-/usr/src/debug}" - DEBUG_CXXFLAGS+=" -ffile-prefix-map=$srcdir=${DBGSRCDIR:-/usr/src/debug}" - DEBUG_RUSTFLAGS+=" --remap-path-prefix=$srcdir=${DBGSRCDIR:-/usr/src/debug}" + DEBUG_CFLAGS+=" -ffile-prefix-map=$srcdir=${DBGSRCDIR:-/usr/src/debug}/${pkgbase}" + DEBUG_CXXFLAGS+=" -ffile-prefix-map=$srcdir=${DBGSRCDIR:-/usr/src/debug}/${pkgbase}" + DEBUG_RUSTFLAGS+=" --remap-path-prefix=$srcdir=${DBGSRCDIR:-/usr/src/debug}/${pkgbase}" CFLAGS+=" $DEBUG_CFLAGS" CXXFLAGS+=" $DEBUG_CXXFLAGS" RUSTFLAGS+=" $DEBUG_RUSTFLAGS" diff --git a/scripts/libmakepkg/tidy/strip.sh.in b/scripts/libmakepkg/tidy/strip.sh.in index 688bcf1b..25bc649e 100644 --- a/scripts/libmakepkg/tidy/strip.sh.in +++ b/scripts/libmakepkg/tidy/strip.sh.in @@ -47,7 +47,7 @@ source_files() { LANG=C debugedit --no-recompute-build-id \ --base-dir "${srcdir}" \ - --dest-dir "${dbgsrcdir}" \ + --dest-dir "${dbgsrcdir}/${pkgbase}" \ --list-file /dev/stdout "$1" \ | sort -zu | tr '\0' '\n' } -- 2.36.1

2 1

[PATCH] proto: Change the default checksum from md5 to sha256
by Ben Westover 18 Jun '22

18 Jun '22

MD5 isn't a very good checksum, and the PKGBUILD page on the Arch Wiki states that it should not be used, instead recommending sha256 or b2. This patch changes the default from md5 to sha256 because that seems to be the most commonly used checksum today. Signed-off-by: Ben Westover <kwestover.kw(a)gmail.com> --- proto/PKGBUILD-split.proto | 2 +- proto/PKGBUILD-vcs.proto | 2 +- proto/PKGBUILD.proto | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/proto/PKGBUILD-split.proto b/proto/PKGBUILD-split.proto index 9898ef81..eea97e56 100644 --- a/proto/PKGBUILD-split.proto +++ b/proto/PKGBUILD-split.proto @@ -28,7 +28,7 @@ changelog= source=("$pkgbase-$pkgver.tar.gz" "$pkgname-$pkgver.patch") noextract=() -md5sums=() +sha256sums=() validpgpkeys=() prepare() { diff --git a/proto/PKGBUILD-vcs.proto b/proto/PKGBUILD-vcs.proto index ae9956a9..49c6759f 100644 --- a/proto/PKGBUILD-vcs.proto +++ b/proto/PKGBUILD-vcs.proto @@ -25,7 +25,7 @@ options=() install= source=('FOLDER::VCS+URL#FRAGMENT') noextract=() -md5sums=('SKIP') +sha256sums=('SKIP') # Please refer to the 'USING VCS SOURCES' section of the PKGBUILD man page for # a description of each element in the source array. diff --git a/proto/PKGBUILD.proto b/proto/PKGBUILD.proto index a2c600d5..9aff797c 100644 --- a/proto/PKGBUILD.proto +++ b/proto/PKGBUILD.proto @@ -27,7 +27,7 @@ changelog= source=("$pkgname-$pkgver.tar.gz" "$pkgname-$pkgver.patch") noextract=() -md5sums=() +sha256sums=() validpgpkeys=() prepare() { -- 2.36.1

1 0

[PATCH] Gitlab CI: remove valgrind test
by Allan McRae 17 Jun '22

17 Jun '22

Currently our gitlab CI is failing due to valgrind breakage. With Arch stripping glibc, valgrind now requires debuginfod to be active. However the gitlab CI system does not retrieve these symbols, even when the appropriate environmental variable is set. If Arch is not appropriate for this test, we may want to switch to another distribution for valgrind. Signed-off-by: Allan McRae <allan(a)archlinux.org> --- This is a temporary patch to get CI running clean again. Still looking at how this can be fixed for Arch, otherwise I will switch the valgrind run to a distro with appropriate dependencies for valgrind to run. .gitlab-ci.yml | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml index 186ad3dc..d9659746 100644 --- a/.gitlab-ci.yml +++ b/.gitlab-ci.yml @@ -46,13 +46,14 @@ arch-clang: - ninja -C build - fakechroot meson test -C build -arch-valgrind: - extends: .arch-test - script: - - pacman -Syu --needed --noconfirm valgrind - - meson build - - ninja -C build - - PACTEST_VALGRIND=1 fakechroot meson test -C build +# Fails due to lack of debug symbols in libc.so.6 and debuginfod not working +#arch-valgrind: +# extends: .arch-test +# script: +# - pacman -Syu --needed --noconfirm valgrind +# - meson build +# - ninja -C build +# - PACTEST_VALGRIND=1 fakechroot meson test -C build arch-nettle: extends: .arch-test -- 2.36.1

2 2