- Pacman-dev - lists.archlinux.org

[pacman-dev] [PATCH] update NEWS for 6.0.0 release
by Eli Schwartz 20 May '21

20 May '21

Signed-off-by: Eli Schwartz <eschwartz(a)archlinux.org> --- crudely semi-sorted, feel free to re-order NEWS | 64 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 64 insertions(+) diff --git a/NEWS b/NEWS index fd2a6e9e0..2cc29a8b8 100644 --- a/NEWS +++ b/NEWS @@ -1,5 +1,69 @@ VERSION DESCRIPTION ----------------------------------------------------------------------------- +6.0.0 - teach the internal downloader to use parallel downloads + (FS#20056) + - TotalDownload now adds an additional progress bar, and the + option to disable it has been removed + - add support for multiple 'Architecture' values + - remove support for the autotools build system + - -Qk now validates file checksums in addition to date/size + - download via -U with a cached package file now tries to + download required signatures, rather than failing (FS#33992) + - allow package signatures to come from the CacheDir if not + embedded in the database + - download detached signature for packages even if signature is + embedded in the database + - when verifying detached signatures, verify the package hash + from the database too + - zsh completion: add pacman-conf support + - clarify misleading error message on init (FS#59595) # not newsworthy? + - -Fx now reports error for invalid regex + - colored upgrade summary now dulls version numbers for contrast + - support xattr when extracting packages + - allow setting --noprogressbar in pacman.conf + - fix output alignment for CJK translated text (FS#59229) + - meson: properly compile internal symbols as hidden + - fix reading targets from stdin when using --sysroot (FS#68630) + - fix deleting signatures for existing databases with -Sc + - skip mirror servers with too many errors (FS#29293) + - fix build errors on systems like FreeBSD + - meson: make -uninstalled.pc correct + - libalpm frontends can now supply context to callbacks (FS#12721) + - fix download rates becoming negative + - check for and forbid duplicate download filenames (FS#67850) + - makepkg: + - add support for sources using the fossil VCS + - strip: fix removing file attributes such as xattr + - add link time optimization support to makepkg + - allow specifying alternative authentication commands when + running pacman as root (FS#32621) + - switch to CRC as default integrity checksum + - add optional argument support to parseopts + - remove dependency on GNU sed + - record $startdir for reproducible builds + - record name of build orchestration tool for reproducible builds + - reduce dependency on file for detecting ELF files + - avoid trailing whitespace in --printsrcinfo output + - libprovides: don't provide both versioned and unversioned + sonames + - fix signing of source packages + - don't double-layer distcc on ccache + - support zstd decompression for sources + - fix detection of source file names for debug packages with + gcc 11 + - strip: silence warnings emitted by readelf while detecting + source filenames + - fix use of spaces in source file renaming (FS#70254) + - pacman-key: + - be less noisy when populating the keyring (FS#64142) + - warn about time taken for master key generation + - --refresh-keys queries WKD before keyserver + - repo-add: + - support the same compression methods as makepkg + - various documentation updates + - after a decade and a half of promising libalpm.3 documentation + "once we get around to doing good Doxygen documentation", it + has happened and is now documented 5.2.2 - fix pacman test suite under python 3.8 - only prompt to import new keys once in a transaction - fix handling of fully downloaded .part files -- 2.31.1

1 0

[pacman-dev] [PATCH] update old NEWS entry with bugtracker ticket reference
by Eli Schwartz 20 May '21

20 May '21

Signed-off-by: Eli Schwartz <eschwartz(a)archlinux.org> --- NEWS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/NEWS b/NEWS index d7c0b1f4f..fd2a6e9e0 100644 --- a/NEWS +++ b/NEWS @@ -40,7 +40,7 @@ VERSION DESCRIPTION - do not warn about PACKAGER format if not set - only run --clean when a package is built - repo-add: - - fix compression of databases with zstd + - fix compression of databases with zstd (FS#64213) 5.2.0 - completely remove delta support (CVE-2019-18183) - add support to pacman and pacman-key for downloading PGP signing keys using the WKD protocol (FS#63171) -- 2.31.1

1 0

[pacman-dev] [GIT] The official pacman repository branch, master, updated. v6.0.0alpha1-118-ge176b8ec
by Allan McRae 19 May '21

19 May '21

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 e176b8ec198ef34ae70a1f46b7b7fe998aa0c475 (commit) via 0e655dd4923495ffcf478e2e695f402670e5ac7b (commit) via 7464062ad6b3a60be6f11e080f11a6b091e496f9 (commit) via 58fde8666896639e5cfb3923fbbfd52d166bffe7 (commit) via 17ccc2f9c22914d1c20b4f7849130c4171fed54b (commit) from 618fa4f675594798825a01bd0a3462903dda6ef7 (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 e176b8ec198ef34ae70a1f46b7b7fe998aa0c475 Author: Eli Schwartz <eschwartz(a)archlinux.org> Date: Wed May 19 03:23:51 2021 -0400 makepkg: squelch readelf warnings from debug stripping readelf --debug-dump sometimes reports inscrutable warnings which don't actually affect our extraction of source filenames. For example: readelf: Warning: There is a hole [0xd3d - 0xd89] in .debug_loc section. Now gcc 11 seems to have dramatically increased the number of warnings: readelf: Warning: Corrupt offset (0x0000008e) in range entry 9 [...] readelf: Warning: Corrupt offset (0x000010f0) in range entry 250 The resulting debuginfo created by the very same toolchain works fine, as does the list of source filenames. But the warnings are quite noisy... send them to /dev/null since they are not actionable in the context of getting source files Signed-off-by: Eli Schwartz <eschwartz(a)archlinux.org> Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 0e655dd4923495ffcf478e2e695f402670e5ac7b Author: Allan McRae <allan(a)archlinux.org> Date: Tue May 18 21:00:03 2021 +1000 Update README for pacman-6.0 Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 7464062ad6b3a60be6f11e080f11a6b091e496f9 Author: Allan McRae <allan(a)archlinux.org> Date: Mon May 17 20:13:48 2021 +1000 Update maintainers list Update to relflect the current state of affairs. Add Morgan, drop Dave. Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 58fde8666896639e5cfb3923fbbfd52d166bffe7 Author: Allan McRae <allan(a)archlinux.org> Date: Mon May 17 20:02:17 2021 +1000 Use attribute to declare case statement fallthroughs Clang does not recognise the comment style notification of expected case statement fallthrough. Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 17ccc2f9c22914d1c20b4f7849130c4171fed54b Author: Allan McRae <allan(a)archlinux.org> Date: Mon May 17 17:09:18 2021 +1000 Silence new warning with gcc-11 Signed-off-by: Allan McRae <allan(a)archlinux.org> ----------------------------------------------------------------------- Summary of changes: README | 34 +++++++++++++++++++++++++++++++--- doc/footer.asciidoc | 3 ++- lib/libalpm/sync.c | 2 +- scripts/libmakepkg/tidy/strip.sh.in | 2 +- src/pacman/callback.c | 2 +- src/pacman/pacman.c | 5 +++-- src/pacman/util.c | 1 + 7 files changed, 40 insertions(+), 9 deletions(-) hooks/post-receive -- The official pacman repository

1 0

[pacman-dev] [PATCH] Use attribute to declare case statement fallthroughs
by Allan McRae 19 May '21

19 May '21

Clang does not recognise the comment style notification of expected case statement fallthrough. Signed-off-by: Allan McRae <allan(a)archlinux.org> --- lib/libalpm/sync.c | 2 +- src/pacman/pacman.c | 5 +++-- src/pacman/util.c | 1 + 3 files changed, 5 insertions(+), 3 deletions(-) diff --git a/lib/libalpm/sync.c b/lib/libalpm/sync.c index 7ef558c4..36ad6242 100644 --- a/lib/libalpm/sync.c +++ b/lib/libalpm/sync.c @@ -1026,7 +1026,7 @@ static int check_validity(alpm_handle_t *handle, v->siglevel & ALPM_SIG_PACKAGE_OPTIONAL, v->siglevel & ALPM_SIG_PACKAGE_MARGINAL_OK, v->siglevel & ALPM_SIG_PACKAGE_UNKNOWN_OK); - /* fallthrough */ + __attribute__((fallthrough)); case ALPM_ERR_PKG_INVALID_CHECKSUM: prompt_to_delete(handle, v->path, v->error); break; diff --git a/src/pacman/pacman.c b/src/pacman/pacman.c index 7e810127..e398855a 100644 --- a/src/pacman/pacman.c +++ b/src/pacman/pacman.c @@ -196,7 +196,7 @@ static void usage(int op, const char * const myname) addlist(_(" --ignore <pkg> ignore a package upgrade (can be used more than once)\n")); addlist(_(" --ignoregroup <grp>\n" " ignore a group upgrade (can be used more than once)\n")); - /* fall through */ + __attribute__((fallthrough)); case PM_OP_REMOVE: addlist(_(" -d, --nodeps skip dependency version checks (-dd to skip all checks)\n")); addlist(_(" --assume-installed <package=version>\n" @@ -412,7 +412,8 @@ static int parsearg_global(int opt) unsigned short debug = (unsigned short)atoi(optarg); switch(debug) { case 2: - config->logmask |= ALPM_LOG_FUNCTION; /* fall through */ + config->logmask |= ALPM_LOG_FUNCTION; + __attribute__((fallthrough)); case 1: config->logmask |= ALPM_LOG_DEBUG; break; diff --git a/src/pacman/util.c b/src/pacman/util.c index 59d6a72e..5486e7a5 100644 --- a/src/pacman/util.c +++ b/src/pacman/util.c @@ -1069,6 +1069,7 @@ static char *pkg_get_location(alpm_pkg_t *pkg) } /* fallthrough - for theoretical serverless repos */ + __attribute__((fallthrough)); case ALPM_PKG_FROM_FILE: return strdup(alpm_pkg_get_filename(pkg)); -- 2.31.1

2 1

[pacman-dev] [PATCH] makepkg: squelch readelf warnings from debug stripping
by Eli Schwartz 19 May '21

19 May '21

readelf --debug-dump sometimes reports inscrutable warnings which don't actually affect our extraction of source filenames. For example: readelf: Warning: There is a hole [0xd3d - 0xd89] in .debug_loc section. Now gcc 11 seems to have dramatically increased the number of warnings: readelf: Warning: Corrupt offset (0x0000008e) in range entry 9 [...] readelf: Warning: Corrupt offset (0x000010f0) in range entry 250 The resulting debuginfo created by the very same toolchain works fine, as does the list of source filenames. But the warnings are quite noisy... send them to /dev/null since they are not actionable in the context of getting source files Signed-off-by: Eli Schwartz <eschwartz(a)archlinux.org> --- scripts/libmakepkg/tidy/strip.sh.in | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/libmakepkg/tidy/strip.sh.in b/scripts/libmakepkg/tidy/strip.sh.in index 737c51153..5d03b28f2 100644 --- a/scripts/libmakepkg/tidy/strip.sh.in +++ b/scripts/libmakepkg/tidy/strip.sh.in @@ -36,7 +36,7 @@ build_id() { } source_files() { - LANG=C readelf "$1" --debug-dump | \ + LANG=C readelf "$1" --debug-dump 2>/dev/null | \ awk '/DW_AT_name +:/{name=$NF}/DW_AT_comp_dir +:/{{if (name == "<artificial>") next}{if (name !~ /^[<\/]/) {printf "%s/", $NF}}{print name}}' } -- 2.31.1

1 0

[pacman-dev] [PATCH] Update README for pacman-6.0
by Allan McRae 18 May '21

18 May '21

Signed-off-by: Allan McRae <allan(a)archlinux.org> --- README | 34 +++++++++++++++++++++++++++++++--- 1 file changed, 31 insertions(+), 3 deletions(-) diff --git a/README b/README index 4465432a..71f9a58f 100644 --- a/README +++ b/README @@ -661,10 +661,38 @@ API CHANGES BETWEEN 5.2 AND 6.0 [REMOVED] - ALPM_EVENT_PKGDOWNLOAD_START, ALPM_EVENT_PKGDOWNLOAD_DONE, ALPM_EVENT_PKGDOWNLOAD_FAILED +- old TotalDownload implementation + - alpm_cb_totaldl + - alpm_option_get_totaldlcb() + - alpm_option_set_totaldlcb() [CHANGED] -- alpm_db_update() changed its signature and now accepts a list of databases - rather than a single database. This is need to handle database downloading - in a multiplexed way. +- alpm_db_update() now accepts a list of databases rather than a single database. +- alpm_fetch_pkgurl() accepts a list of packages to download. +- alpm_db_search() now has an additional parameter and returns success status +- ALPM_EVENT_RETRIEVE_* -> ALPM_EVENT_DB_RETRIEVE_* and ALPM_EVENT_PKG_RETRIEVE_* +- alpm_cb_download pass event and data +- multi architecture support + - alpm_option_get_arch() -> alpm_option_get_architectures() + - alpm_option_set_arch() -> alpm_option_set_architectures() +- alpm_db_get_servers() copies parameter data [ADDED] +- parallel download support + - alpm_option_set_parallel_downloads() + - alpm_option_get_parallel_downloads() +- file download events + - alpm_download_event_type_t + - alpm_download_event_init_t + - alpm_download_event_progress_t + - alpm_download_event_completed_t +- download misc + - ALPM_DOWNLOAD_RETRY + - alpm_download_event_retry_t + - alpm_event_pkg_retrieve_t +- multiarchitecture support + - alpm_option_add_architecture() + - alpm_option_remove_architecture() +- misc + - alpm_pkg_get_sig() +- callbacks add front-end provided context -- 2.31.1

1 0

[pacman-dev] [PATCH] libalpm: remove unused error value
by morganamilo 18 May '21

18 May '21

--- lib/libalpm/alpm.h | 2 -- lib/libalpm/error.c | 2 -- 2 files changed, 4 deletions(-) diff --git a/lib/libalpm/alpm.h b/lib/libalpm/alpm.h index c4acc062..a5f4a6ae 100644 --- a/lib/libalpm/alpm.h +++ b/lib/libalpm/alpm.h @@ -292,8 +292,6 @@ typedef enum _alpm_errno_t { ALPM_ERR_PKG_INVALID_NAME, /** Package has an invalid architecture */ ALPM_ERR_PKG_INVALID_ARCH, - /** Unused */ - ALPM_ERR_PKG_REPO_NOT_FOUND, /* Signatures */ /** Signatures are missing */ ALPM_ERR_SIG_MISSING, diff --git a/lib/libalpm/error.c b/lib/libalpm/error.c index c06d091a..c1a16cee 100644 --- a/lib/libalpm/error.c +++ b/lib/libalpm/error.c @@ -125,8 +125,6 @@ const char SYMEXPORT *alpm_strerror(alpm_errno_t err) return _("package filename is not valid"); case ALPM_ERR_PKG_INVALID_ARCH: return _("package architecture is not valid"); - case ALPM_ERR_PKG_REPO_NOT_FOUND: - return _("could not find repository for target"); /* Signatures */ case ALPM_ERR_SIG_MISSING: return _("missing PGP signature"); -- 2.31.1

3 4

[pacman-dev] [PATCH] libalpm: fix alpm_fetch_pkgurl with a fetch callback
by Guillaume Benoit 18 May '21

18 May '21

After download, alpm_fetch_pkgurl uses payload->destfile_name and payload->tempfile_name in order to find the downloaded file path. Those fields are not set if a custom fetch callback is defined. Use filecache_find_url instead, like in the beginning of the function. --- lib/libalpm/dload.c | 13 ++++--------- 1 file changed, 4 insertions(+), 9 deletions(-) diff --git a/lib/libalpm/dload.c b/lib/libalpm/dload.c index 6f33451a..d45c7707 100644 --- a/lib/libalpm/dload.c +++ b/lib/libalpm/dload.c @@ -1007,16 +1007,11 @@ int SYMEXPORT alpm_fetch_pkgurl(alpm_handle_t *handle, const alpm_list_t *urls, EVENT(handle, &event); } - for(i = payloads; i; i = i->next) { - struct dload_payload *payload = i->data; - char *filepath; + for(i = urls; i; i = i->next) { + char *url = i->data; - if(payload->destfile_name) { - const char *filename = mbasename(payload->destfile_name); - filepath = _alpm_filecache_find(handle, filename); - } else { - STRDUP(filepath, payload->tempfile_name, GOTO_ERR(handle, ALPM_ERR_MEMORY, err)); - } + /* attempt again to find the file in our pkgcache */ + char *filepath = filecache_find_url(handle, url); if(filepath) { alpm_list_append(fetched, filepath); } else { -- 2.31.1

3 9

[pacman-dev] [PATCH] Update maintainers list
by Allan McRae 18 May '21

18 May '21

Update to relflect the current state of affairs. Add Morgan, drop Dave. Signed-off-by: Allan McRae <allan(a)archlinux.org> --- Morgan has been contributing patches consistently over the last couple of releases - in the top 3 patch contributors for both the 5.2 and 6.0 releases, and in the top 10 of patch submitters overall. Being a maintainer comes with the added responsibility of doing a bit of patch review! (I am always looking for more eyes on patches!) doc/footer.asciidoc | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/doc/footer.asciidoc b/doc/footer.asciidoc index 1bae8b20..edf36080 100644 --- a/doc/footer.asciidoc +++ b/doc/footer.asciidoc @@ -18,8 +18,8 @@ Current maintainers: * Allan McRae <allan(a)archlinux.org> * Andrew Gregory <andrew.gregory.8(a)gmail.com> -* Dave Reisner <dreisner(a)archlinux.org> * Eli Schwartz <eschwartz(a)archlinux.org> +* Morgan Adamiec <morganamilo(a)archlinux.org> Past major contributors: @@ -29,6 +29,7 @@ Past major contributors: * Dan McGee <dan(a)archlinux.org> * Xavier Chantry <shiningxc(a)gmail.com> * Nagy Gabor <ngaba(a)bibl.u-szeged.hu> +* Dave Reisner <dreisner(a)archlinux.org> For additional contributors, use `git shortlog -s` on the pacman.git repository. -- 2.31.1

1 0

[pacman-dev] [PATCH] Silence new warning with gcc-11
by Allan McRae 17 May '21

17 May '21

Signed-off-by: Allan McRae <allan(a)archlinux.org> --- src/pacman/callback.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/pacman/callback.c b/src/pacman/callback.c index 99ad716e..75c74f8b 100644 --- a/src/pacman/callback.c +++ b/src/pacman/callback.c @@ -677,7 +677,7 @@ void cb_progress(void *ctx, alpm_progress_t event, const char *pkgname, int i = textlen - 3; wchar_t *p = wcstr; /* grab the max number of char columns we can fill */ - while(i - wcwidth(*p) > 0) { + while(i > wcwidth(*p)) { i -= wcwidth(*p); p++; } -- 2.31.1

1 0

[pacman-dev] [GIT] The official pacman repository branch, master, updated. v6.0.0alpha1-113-g618fa4f6
by Allan McRae 17 May '21

17 May '21

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 618fa4f675594798825a01bd0a3462903dda6ef7 (commit) via d6ffa7f56134b37efb0ea798dc6a0013e8a4d561 (commit) via 8bf17b29a24802c587afda0e96d93b22c5f5caeb (commit) via 4fead44e3ce97d3a83ca6bab393d480746ba9227 (commit) via 15be417c17809cbf2721a3fb0c519ee910b8d0a6 (commit) via 31c42c338b717ea6e0cae92c2279d859a882d1b3 (commit) from ef14c3380da0ed0e3eeba7e8b6c66294db316883 (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 618fa4f675594798825a01bd0a3462903dda6ef7 Author: morganamilo <morganamilo(a)archlinux.org> Date: Sun May 9 15:18:17 2021 +0100 libalpm: check for duplicate filenames This partially fixes FS#67850 It fixes the case for -S'ing packages but not -U'ing urls. pacman -S a/a b/b resolving dependencies... error: packages a and b have the same filename: a-1-1-any.pkg.tar.zst error: failed to prepare transaction (duplicate filename) Signed-off-by: Allan McRae <allan(a)archlinux.org> commit d6ffa7f56134b37efb0ea798dc6a0013e8a4d561 Author: morganamilo <morganamilo(a)archlinux.org> Date: Sun May 9 15:18:18 2021 +0100 fix formatting mistake Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 8bf17b29a24802c587afda0e96d93b22c5f5caeb Author: morganamilo <morganamilo(a)archlinux.org> Date: Sun May 9 14:23:57 2021 +0100 libalpm: fix download rates becoming negative When a download fails on one mirror a new download is started on the next mirror. This causes the ammount downloaded to reset, confusing the rate math and making it display a negative rate. This is further complicated by the fact that a download may be resumed from where it is or started over. To account for this we alert the frontend that the download was restarted. Pacman then starts the progress bar over. Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 4fead44e3ce97d3a83ca6bab393d480746ba9227 Author: morganamilo <morganamilo(a)archlinux.org> Date: Sun May 9 13:18:38 2021 +0100 libalpm: clone data on alpm_db_set_servers Every alpm_option_set function clones the input so lets be more consistent. Also this fixes servers not being sanatized. Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 15be417c17809cbf2721a3fb0c519ee910b8d0a6 Author: morganamilo <morganamilo(a)archlinux.org> Date: Fri May 7 07:26:12 2021 +0100 remove unused variable in meson.build Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 31c42c338b717ea6e0cae92c2279d859a882d1b3 Author: morganamilo <morganamilo(a)archlinux.org> Date: Sun Apr 18 20:59:45 2021 +0100 Match initial download message with progress message When initially downloading a package, pacman will display a message like: wine-6.6-1-x86_64.pkg.tar.zst downloading... Then when the download progresses the message will change to: wine-6.6-1-x86_64.pkg.tar.zst So instead lets match the progress message so there's no sudden change. Signed-off-by: Allan McRae <allan(a)archlinux.org> ----------------------------------------------------------------------- Summary of changes: lib/libalpm/alpm.h | 16 ++++++++++++---- lib/libalpm/db.c | 8 +++++++- lib/libalpm/dload.c | 33 +++++++++++++++++++++++++-------- lib/libalpm/error.c | 2 ++ lib/libalpm/sync.c | 19 +++++++++++++++++++ lib/libalpm/trans.c | 2 +- meson.build | 1 - src/pacman/callback.c | 34 ++++++++++++++++++++++++++++++++-- 8 files changed, 98 insertions(+), 17 deletions(-) hooks/post-receive -- The official pacman repository

1 0

[pacman-dev] [PATCH v2 3/4] libalpm: check for duplicate filenames
by morganamilo 09 May '21

09 May '21

This partially fixes FS#67850 It fixes the case for -S'ing packages but not -U'ing urls. pacman -S a/a b/b resolving dependencies... error: packages a and b have the same filename: a-1-1-any.pkg.tar.zst error: failed to prepare transaction (duplicate filename) --- lib/libalpm/alpm.h | 2 ++ lib/libalpm/error.c | 2 ++ lib/libalpm/sync.c | 19 +++++++++++++++++++ 3 files changed, 23 insertions(+) diff --git a/lib/libalpm/alpm.h b/lib/libalpm/alpm.h index 833df829..ccf2edc2 100644 --- a/lib/libalpm/alpm.h +++ b/lib/libalpm/alpm.h @@ -257,6 +257,8 @@ typedef enum _alpm_errno_t { ALPM_ERR_TRANS_NULL, /** Duplicate target in transaction */ ALPM_ERR_TRANS_DUP_TARGET, + /** Duplicate filename in transaction */ + ALPM_ERR_TRANS_DUP_FILENAME, /** A transaction has not been initialized */ ALPM_ERR_TRANS_NOT_INITIALIZED, /** Transaction has not been prepared */ diff --git a/lib/libalpm/error.c b/lib/libalpm/error.c index c6f19599..c06d091a 100644 --- a/lib/libalpm/error.c +++ b/lib/libalpm/error.c @@ -90,6 +90,8 @@ const char SYMEXPORT *alpm_strerror(alpm_errno_t err) return _("transaction not initialized"); case ALPM_ERR_TRANS_DUP_TARGET: return _("duplicate target"); + case ALPM_ERR_TRANS_DUP_FILENAME: + return _("duplicate filename"); case ALPM_ERR_TRANS_NOT_INITIALIZED: return _("transaction not initialized"); case ALPM_ERR_TRANS_NOT_PREPARED: diff --git a/lib/libalpm/sync.c b/lib/libalpm/sync.c index 7fa50a13..8ccb0ba0 100644 --- a/lib/libalpm/sync.c +++ b/lib/libalpm/sync.c @@ -464,6 +464,25 @@ int _alpm_sync_prepare(alpm_handle_t *handle, alpm_list_t **data) } } + /* Ensure two packages don't have the same filename */ + for(i = resolved; i; i = i->next) { + alpm_pkg_t *pkg1 = i->data; + for(j = i->next; j; j = j->next) { + alpm_pkg_t *pkg2 = j->data; + if(strcmp(pkg1->filename, pkg2->filename) == 0) { + alpm_list_free(resolved); + ret = -1; + handle->pm_errno = ALPM_ERR_TRANS_DUP_FILENAME; + _alpm_log(handle, ALPM_LOG_ERROR, _("packages %s and %s have the same filename: %s\n"), + pkg1->name, pkg2->name, pkg1->filename); + } + } + } + + if(ret != 0) { + goto cleanup; + } + /* Set DEPEND reason for pulled packages */ for(i = resolved; i; i = i->next) { alpm_pkg_t *pkg = i->data; -- 2.31.1

1 1

[pacman-dev] [PATCH v3] libalpm: fix download rates becoming negative
by morganamilo 09 May '21

09 May '21

When a download fails on one mirror a new download is started on the next mirror. This causes the ammount downloaded to reset, confusing the rate math and making it display a negative rate. This is further complicated by the fact that a download may be resumed from where it is or started over. To account for this we alert the frontend that the download was restarted. Pacman then starts the progress bar over. --- v2: stat tempfile to get initial_size v3: review fixes --- lib/libalpm/alpm.h | 7 +++++++ lib/libalpm/dload.c | 33 +++++++++++++++++++++++++-------- src/pacman/callback.c | 30 ++++++++++++++++++++++++++++++ 3 files changed, 62 insertions(+), 8 deletions(-) diff --git a/lib/libalpm/alpm.h b/lib/libalpm/alpm.h index 101d686b..226fd953 100644 --- a/lib/libalpm/alpm.h +++ b/lib/libalpm/alpm.h @@ -1172,6 +1172,8 @@ typedef enum _alpm_download_event_type_t { ALPM_DOWNLOAD_INIT, /** A download made progress */ ALPM_DOWNLOAD_PROGRESS, + /** Download will be retried */ + ALPM_DOWNLOAD_RETRY, /** A download completed */ ALPM_DOWNLOAD_COMPLETED } alpm_download_event_type_t; @@ -1190,6 +1192,11 @@ typedef struct _alpm_download_event_progress_t { off_t total; } alpm_download_event_progress_t; +/** Context struct for when a download retries. */ +typedef struct _alpm_download_event_retry_t { + /** If the download will resume or start over */ + int resume; +} alpm_download_event_retry_t; /** Context struct for when a download completes. */ typedef struct _alpm_download_event_completed_t { diff --git a/lib/libalpm/dload.c b/lib/libalpm/dload.c index 8fa46bb2..2d8b4d6d 100644 --- a/lib/libalpm/dload.c +++ b/lib/libalpm/dload.c @@ -408,6 +408,7 @@ static int curl_retry_next_server(CURLM *curlm, CURL *curl, struct dload_payload { const char *server; size_t len; + struct stat st; alpm_handle_t *handle = payload->handle; while(payload->servers && should_skip_server(handle, payload->servers->data)) { @@ -427,15 +428,31 @@ static int curl_retry_next_server(CURLM *curlm, CURL *curl, struct dload_payload MALLOC(payload->fileurl, len, RET_ERR(handle, ALPM_ERR_MEMORY, -1)); snprintf(payload->fileurl, len, "%s/%s", server, payload->filepath); - if(payload->unlink_on_fail) { + + fflush(payload->localf); + + if(payload->allow_resume && stat(payload->tempfile_name, &st) == 0) { + /* a previous partial download exists, resume from end of file. */ + payload->tempfile_openmode = "ab"; + curl_easy_setopt(curl, CURLOPT_RESUME_FROM_LARGE, (curl_off_t)st.st_size); + _alpm_log(handle, ALPM_LOG_DEBUG, + "%s: tempfile found, attempting continuation from %jd bytes\n", + payload->remote_name, (intmax_t)st.st_size); + payload->initial_size = st.st_size; + } else { /* we keep the file for a new retry but remove its data if any */ - fflush(payload->localf); if(ftruncate(fileno(payload->localf), 0)) { RET_ERR(handle, ALPM_ERR_SYSTEM, -1); } fseek(payload->localf, 0, SEEK_SET); } + if(handle->dlcb && !payload->signature) { + alpm_download_event_retry_t cb_data; + cb_data.resume = payload->allow_resume; + handle->dlcb(handle->dlcb_ctx, payload->remote_name, ALPM_DOWNLOAD_RETRY, &cb_data); + } + /* Set curl with the new URL */ curl_easy_setopt(curl, CURLOPT_URL, payload->fileurl); @@ -483,7 +500,6 @@ static int curl_check_finished_download(CURLM *curlm, CURLMsg *msg, _alpm_log(handle, ALPM_LOG_DEBUG, "%s: response code %ld\n", payload->remote_name, payload->respcode); if(payload->respcode >= 400) { - payload->unlink_on_fail = 1; if(!payload->errors_ok) { handle->pm_errno = ALPM_ERR_RETRIEVE; /* non-translated message is same as libcurl */ @@ -498,6 +514,7 @@ static int curl_check_finished_download(CURLM *curlm, CURLMsg *msg, (*active_downloads_num)++; return 2; } else { + payload->unlink_on_fail = 1; goto cleanup; } } @@ -515,7 +532,6 @@ static int curl_check_finished_download(CURLM *curlm, CURLMsg *msg, } goto cleanup; case CURLE_COULDNT_RESOLVE_HOST: - payload->unlink_on_fail = 1; handle->pm_errno = ALPM_ERR_SERVER_BAD_URL; _alpm_log(handle, ALPM_LOG_ERROR, _("failed retrieving file '%s' from %s : %s\n"), @@ -525,13 +541,10 @@ static int curl_check_finished_download(CURLM *curlm, CURLMsg *msg, (*active_downloads_num)++; return 2; } else { + payload->unlink_on_fail = 1; goto cleanup; } default: - /* delete zero length downloads */ - if(fstat(fileno(payload->localf), &st) == 0 && st.st_size == 0) { - payload->unlink_on_fail = 1; - } if(!payload->errors_ok) { handle->pm_errno = ALPM_ERR_LIBCURL; _alpm_log(handle, ALPM_LOG_ERROR, @@ -547,6 +560,10 @@ static int curl_check_finished_download(CURLM *curlm, CURLMsg *msg, (*active_downloads_num)++; return 2; } else { + /* delete zero length downloads */ + if(fstat(fileno(payload->localf), &st) == 0 && st.st_size == 0) { + payload->unlink_on_fail = 1; + } goto cleanup; } } diff --git a/src/pacman/callback.c b/src/pacman/callback.c index ae7e1bed..1881f8ac 100644 --- a/src/pacman/callback.c +++ b/src/pacman/callback.c @@ -994,6 +994,34 @@ static void dload_progress_event(const char *filename, alpm_download_event_progr fflush(stdout); } +/* download retried */ +static void dload_retry_event(const char *filename, alpm_download_event_retry_t *data) { + if(!dload_progressbar_enabled()) { + return; + } + + int index; + struct pacman_progress_bar *bar; + bool ok = find_bar_for_filename(filename, &index, &bar); + assert(ok); + + if(!data->resume) { + if(total_enabled) { + /* note total download does not reflect partial downloads that are restarted */ + totalbar->xfered -= bar->xfered; + } + } + + bar->xfered = 0; + bar->total_size = 0; + bar->init_time = get_time_ms(); + bar->sync_time = 0; + bar->sync_xfered = 0; + bar->rate = 0.0; + bar->eta = 0.0; +} + + /* download completed */ static void dload_complete_event(const char *filename, alpm_download_event_completed_t *data) { @@ -1090,6 +1118,8 @@ void cb_download(void *ctx, const char *filename, alpm_download_event_type_t eve dload_init_event(filename, data); } else if(event == ALPM_DOWNLOAD_PROGRESS) { dload_progress_event(filename, data); + } else if(event == ALPM_DOWNLOAD_RETRY) { + dload_retry_event(filename, data); } else if(event == ALPM_DOWNLOAD_COMPLETED) { dload_complete_event(filename, data); } else { -- 2.31.1

1 0

[pacman-dev] [PATCH] libalpm: check for duplicate filenames
by morganamilo 09 May '21

09 May '21

This partially fixes FS#67850 It fixes the case for -S'ing packages but not -U'ing urls. morganamilo@Octavia dup ~git/pacman % sudo ./build/pacman -S a/a b/b resolving dependencies... error: a and b have the same filename: a-1-1-any.pkg.tar.zst error: failed to prepare transaction (duplicate filename) diff --git a/lib/libalpm/alpm.h b/lib/libalpm/alpm.h index 833df829..ccf2edc2 100644 --- a/lib/libalpm/alpm.h +++ b/lib/libalpm/alpm.h @@ -257,6 +257,8 @@ typedef enum _alpm_errno_t { ALPM_ERR_TRANS_NULL, /** Duplicate target in transaction */ ALPM_ERR_TRANS_DUP_TARGET, + /** Duplicate filename in transaction */ + ALPM_ERR_TRANS_DUP_FILENAME, /** A transaction has not been initialized */ ALPM_ERR_TRANS_NOT_INITIALIZED, /** Transaction has not been prepared */ diff --git a/lib/libalpm/error.c b/lib/libalpm/error.c index c6f19599..c06d091a 100644 --- a/lib/libalpm/error.c +++ b/lib/libalpm/error.c @@ -90,6 +90,8 @@ const char SYMEXPORT *alpm_strerror(alpm_errno_t err) return _("transaction not initialized"); case ALPM_ERR_TRANS_DUP_TARGET: return _("duplicate target"); + case ALPM_ERR_TRANS_DUP_FILENAME: + return _("duplicate filename"); case ALPM_ERR_TRANS_NOT_INITIALIZED: return _("transaction not initialized"); case ALPM_ERR_TRANS_NOT_PREPARED: diff --git a/lib/libalpm/sync.c b/lib/libalpm/sync.c index 7fa50a13..73e9e77d 100644 --- a/lib/libalpm/sync.c +++ b/lib/libalpm/sync.c @@ -464,6 +464,25 @@ int _alpm_sync_prepare(alpm_handle_t *handle, alpm_list_t **data) } } + /* Ensure two packages don't have the same filename */ + for(i = resolved; i; i = i->next) { + alpm_pkg_t *pkg1 = i->data; + for(j = i->next; j; j = j->next) { + alpm_pkg_t *pkg2 = j->data; + if(strcmp(pkg1->filename, pkg2->filename) == 0) { + alpm_list_free(resolved); + ret = -1; + handle->pm_errno = ALPM_ERR_TRANS_DUP_FILENAME; + _alpm_log(handle, ALPM_LOG_ERROR, _("%s and %s have the same filename: %s\n"), + pkg1->name, pkg2->name, pkg1->filename); + } + } + } + + if(ret != 0) { + goto cleanup; + } + /* Set DEPEND reason for pulled packages */ for(i = resolved; i; i = i->next) { alpm_pkg_t *pkg = i->data; diff --git a/lib/libalpm/trans.c b/lib/libalpm/trans.c index c6ec7eea..4cee603c 100644 --- a/lib/libalpm/trans.c +++ b/lib/libalpm/trans.c @@ -122,7 +122,7 @@ int SYMEXPORT alpm_trans_prepare(alpm_handle_t *handle, alpm_list_t **data) /* pm_errno is set by _alpm_remove_prepare() */ return -1; } - } else { + } else { if(_alpm_sync_prepare(handle, data) == -1) { /* pm_errno is set by _alpm_sync_prepare() */ return -1; -- 2.31.1

2 1

[pacman-dev] [PATCH] libalpm: download rates becoming negative
by morganamilo 09 May '21

09 May '21

When a download fails on one mirror a new download is started on the next mirror. This new download will have the initial size of whatever has been downloaded so far, as well as the ammount downloaded reset to 0. To account for this, when a download changes mirror, save how much has been downloaded so far and add that to dlcb calls. --- lib/libalpm/dload.c | 14 ++++++++++++-- lib/libalpm/dload.h | 1 + 2 files changed, 13 insertions(+), 2 deletions(-) diff --git a/lib/libalpm/dload.c b/lib/libalpm/dload.c index a4c42f8d..27a7748a 100644 --- a/lib/libalpm/dload.c +++ b/lib/libalpm/dload.c @@ -207,8 +207,8 @@ static int dload_progress_cb(void *file, curl_off_t dltotal, curl_off_t dlnow, /* do NOT include initial_size since it wasn't part of the package's * download_size (nor included in the total download size callback) */ - cb_data.total = dltotal; - cb_data.downloaded = dlnow; + cb_data.total = dltotal + payload->resetprogress; + cb_data.downloaded = dlnow + payload->resetprogress; payload->handle->dlcb(payload->remote_name, ALPM_DOWNLOAD_PROGRESS, &cb_data); payload->prevprogress = current_size; @@ -440,6 +440,16 @@ static int curl_retry_next_server(CURLM *curlm, CURL *curl, struct dload_payload fseek(payload->localf, 0, SEEK_SET); } + + /* reset progress for next server */ + payload->resetprogress += payload->prevprogress - payload->initial_size; + payload->unlink_on_fail = 0; + + if(payload->allow_resume) { + payload->initial_size = payload->prevprogress; + } + + /* Set curl with the new URL */ curl_easy_setopt(curl, CURLOPT_URL, payload->fileurl); diff --git a/lib/libalpm/dload.h b/lib/libalpm/dload.h index 8f3d17b4..abfd9466 100644 --- a/lib/libalpm/dload.h +++ b/lib/libalpm/dload.h @@ -42,6 +42,7 @@ struct dload_payload { off_t initial_size; off_t max_size; off_t prevprogress; + off_t resetprogress; /* progress made before we try a new server */ int force; int allow_resume; int errors_ok; -- 2.31.1

5 15

[pacman-dev] [PATCH v3] libalpm: clone data on alpm_db_set_servers
by morganamilo 09 May '21

09 May '21

Every alpm_option_set function clones the input so lets be more consistent. Also this fixes servers not being sanatized. --- v2: fix copy paste error --- lib/libalpm/alpm.h | 4 +--- lib/libalpm/db.c | 8 +++++++- 2 files changed, 8 insertions(+), 4 deletions(-) diff --git a/lib/libalpm/alpm.h b/lib/libalpm/alpm.h index 65b8e2fb..e6aefd4c 100644 --- a/lib/libalpm/alpm.h +++ b/lib/libalpm/alpm.h @@ -1312,9 +1312,7 @@ alpm_list_t *alpm_db_get_servers(const alpm_db_t *db); /** Sets the list of servers for the database to use. * @param db the database to set the servers - * @param servers a char* list of servers. Note: the database will - * take ownership of the list and it should no longer be - * freed by the caller + * @param servers a char* list of servers. */ int alpm_db_set_servers(alpm_db_t *db, alpm_list_t *servers); diff --git a/lib/libalpm/db.c b/lib/libalpm/db.c index 8511bb83..b8d1b157 100644 --- a/lib/libalpm/db.c +++ b/lib/libalpm/db.c @@ -139,9 +139,15 @@ alpm_list_t SYMEXPORT *alpm_db_get_servers(const alpm_db_t *db) int SYMEXPORT alpm_db_set_servers(alpm_db_t *db, alpm_list_t *servers) { + alpm_list_t *i; ASSERT(db != NULL, return -1); FREELIST(db->servers); - db->servers = servers; + for(i = servers; i; i = i->next) { + char *url = i->data; + if(alpm_db_add_server(db, url) != 0) { + return -1; + } + } return 0; } -- 2.31.1

1 0

[pacman-dev] [PATCH 0/1] signhandler add SIGQUIT soft_interrupt_handler
by ZheNing Hu 09 May '21

09 May '21

Hello, eveyone! I'am a newbie, because of the experience of using pacman downloading with the `Ctrl +\` exit was not very good,it leave the database lock, so I am thinking can we just deal with SIGQUIT rather than doing nothing. Hopefully this will improve the experience of users. Thanks. ZheNing Hu (1): signhandler: add SIGQUIT soft_interrupt_handler src/pacman/sighandler.c | 3 +++ 1 file changed, 3 insertions(+) -- 2.31.0

3 3

[pacman-dev] [PATCH] libalpm: clone data on alpm_db_set_servers
by morganamilo 09 May '21

09 May '21

Every alpm_option_set function clones the input so lets be more consistent. Also this fixes servers not being sanatized. --- lib/libalpm/db.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/lib/libalpm/db.c b/lib/libalpm/db.c index 8511bb83..6ddf5461 100644 --- a/lib/libalpm/db.c +++ b/lib/libalpm/db.c @@ -139,9 +139,16 @@ alpm_list_t SYMEXPORT *alpm_db_get_servers(const alpm_db_t *db) int SYMEXPORT alpm_db_set_servers(alpm_db_t *db, alpm_list_t *servers) { + alpm_list_t *i; ASSERT(db != NULL, return -1); FREELIST(db->servers); - db->servers = servers; + for(i = servers; i; i = i->next) { + char *url = i->data; + if(alpm_db_add_server(db, url) != 0) { + return -1; + } + servers = servers->next; + } return 0; } -- 2.31.1

3 3

[pacman-dev] [PATCH] remove unused variable in meson.build
by morganamilo 07 May '21

07 May '21

--- meson.build | 1 - 1 file changed, 1 deletion(-) diff --git a/meson.build b/meson.build index c8835ab6..46f38de6 100644 --- a/meson.build +++ b/meson.build @@ -28,7 +28,6 @@ LIBMAKEPKGDIR = join_paths(PREFIX, DATAROOTDIR, 'makepkg') PKGDATADIR = join_paths(PREFIX, DATAROOTDIR, meson.project_name()) PYTHON = import('python').find_installation('python3') -DU = find_program('du') LDCONFIG = get_option('ldconfig') MESON_MAKE_SYMLINK = join_paths(meson.source_root(), 'build-aux/meson-make-symlink.sh') -- 2.31.1

1 0

[pacman-dev] [GIT] The official pacman repository branch, master, updated. v6.0.0alpha1-107-gef14c338
by Allan McRae 03 May '21

03 May '21

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 ef14c3380da0ed0e3eeba7e8b6c66294db316883 (commit) from 72238aa046b8a8de1c7b9110726986c7538023b5 (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 ef14c3380da0ed0e3eeba7e8b6c66294db316883 Author: morganamilo <morganamilo(a)archlinux.org> Date: Sat May 1 21:25:35 2021 +0100 libalpm: fix error when installing cached packages without signatures With a repo using "SigLevel = Optional" and a package already downloaded into the cache, download_files() returns 1 (via _alpm_download) to indicate no files were downloaded. This causes installation of the package to fail. Explicitly check that download_files() returns -1 (error) rather than non-zero. Signed-off-by: Allan McRae <allan(a)archlinux.org> ----------------------------------------------------------------------- Summary of changes: lib/libalpm/sync.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) hooks/post-receive -- The official pacman repository

1 0

[pacman-dev] (no subject)
by Jazmine Breaux 03 May '21

03 May '21

1 0

[pacman-dev] [PATCH v2] libalpm: fix error when downloading up to date files
by morganamilo 03 May '21

03 May '21

--- lib/libalpm/sync.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/libalpm/sync.c b/lib/libalpm/sync.c index 2419cc69..bdfab635 100644 --- a/lib/libalpm/sync.c +++ b/lib/libalpm/sync.c @@ -1128,7 +1128,7 @@ int _alpm_sync_load(alpm_handle_t *handle, alpm_list_t **data) uint64_t total_bytes = 0; alpm_trans_t *trans = handle->trans; - if(download_files(handle)) { + if(download_files(handle) == -1) { return -1; } -- 2.31.1

2 1

Re: [pacman-dev] [PATCH] libalpm: fix alpm_fetch_pkgurl with a fetch callback
by Guillaume Benoit 02 May '21

02 May '21

Le 02/05/2021 à 11:47, Allan McRae a écrit :> On 2/5/21 12:33 am, Guillaume Benoit wrote:>> Le 01/05/2021 à 05:53, Andrew Gregory a écrit :>>> On 04/30/21 at 12:09pm, Guillaume Benoit wrote:>>>> After download, alpm_fetch_pkgurl uses payload->destfile_name>>>> and payload->tempfile_name in order to find the downloaded file path.>>>> Those fields are not set if a custom fetch callback is defined.>>>> Use filecache_find_url instead, like in the beginning of the function.>>>>>>>> --->>>> lib/libalpm/dload.c | 13 ++++--------->>>> 1 file changed, 4 insertions(+), 9 deletions(-)>>>>>>>> diff --git a/lib/libalpm/dload.c b/lib/libalpm/dload.c>>>> index 6f33451a..d45c7707 100644>>>> --- a/lib/libalpm/dload.c>>>> +++ b/lib/libalpm/dload.c>>>> @@ -1007,16 +1007,11 @@ int SYMEXPORT alpm_fetch_pkgurl(alpm_handle_t>>>> *handle, const alpm_list_t *urls,>>>> EVENT(handle, &event);>>>> }>>>> - for(i = payloads; i; i = i->next) {>>>> - struct dload_payload *payload = i->data;>>>> - char *filepath;>>>> + for(i = urls; i; i = i->next) {>>>> + char *url = i->data;>>>> - if(payload->destfile_name) {>>>> - const char *filename =>>>> mbasename(payload->destfile_name);>>>> - filepath = _alpm_filecache_find(handle, filename);>>>> - } else {>>>> - STRDUP(filepath, payload->tempfile_name,>>>> GOTO_ERR(handle, ALPM_ERR_MEMORY, err));>>>> - }>>>> + /* attempt again to find the file in our pkgcache */>>>> + char *filepath = filecache_find_url(handle, url);>>>> if(filepath) {>>>> alpm_list_append(fetched, filepath);>>>> } else {>>>>>> Without testing since this needs to be rebased now; this looks broken.>>> filecache_find_url with fail for any url where destfile_name doesn't>>> match the file name in the url. Notably, that includes the download>>> links on Arch's package pages:>>> https://archlinux.org/packages/core/x86_64/pacman/download/>>>>> Yes, with or without this patch, the current implementation doesn't work>> for this kind of url when using a fetch callback.>> The solution I see is to change the fetch callback signature by making>> it return the path of the downloaded file.>> Otherwise this patch works for standard mirror urls.>> .> > While this patch does improve the current situation, it is a step> backwards in terms the full fix, which will need to pass the full> payload to the front-end so it can fill in the needed values.> > On that basis, I will not be accepting this patch.> > Given fetching a URL with XferCommand has not worked for a long time (or> ever?), this issue is not a blocker for 6.0.> > Allan> Yes, with pacman 5, fetching a URL with XferCommand doesn't work.With pacman 6, it doesn't work with the current code but it works with this patch,except for the special URLs given by Andrew Gregory.I can try to provide a different patch that pass the full payload to the front-endbut it can't help making pacman 6 correctly fetching a URL with XferCommand when thefilename can be guessed from the URL.

3 2

[pacman-dev] [PATCH] libalpm: fix error when downloading up to date files
by morganamilo 01 May '21

01 May '21

--- lib/libalpm/sync.c | 2 +- lib/libalpm/trans.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/lib/libalpm/sync.c b/lib/libalpm/sync.c index 2419cc69..bdfab635 100644 --- a/lib/libalpm/sync.c +++ b/lib/libalpm/sync.c @@ -1128,7 +1128,7 @@ int _alpm_sync_load(alpm_handle_t *handle, alpm_list_t **data) uint64_t total_bytes = 0; alpm_trans_t *trans = handle->trans; - if(download_files(handle)) { + if(download_files(handle) == -1) { return -1; } diff --git a/lib/libalpm/trans.c b/lib/libalpm/trans.c index 939ab05a..e608555b 100644 --- a/lib/libalpm/trans.c +++ b/lib/libalpm/trans.c @@ -185,7 +185,7 @@ int SYMEXPORT alpm_trans_commit(alpm_handle_t *handle, alpm_list_t **data) } if(trans->add) { - if(_alpm_sync_load(handle, data) != 0) { + if(_alpm_sync_load(handle, data) == -1) { /* pm_errno is set by _alpm_sync_load() */ return -1; } -- 2.31.1

1 0

Re: [pacman-dev] Patch Policy Inquiry
by Sijie Bu 01 May '21

01 May '21

Hello, Sorry for the late reply. I was a bit busy for the past two days. The specific Arch Linux 32 changes in question, though I have not confirmed yet, seems to be detecting Pentium 4 CPUs and optimizing accordingly. I will have to discuss with my project mentor to see if we need these too (because we're targeting the R6 sub-architecture, which is relatively new). But otherwise, looks like the pacman code should compile and run cleanly. Thank you for the clarifications. Sincerely

1 0