- Pacman-dev - lists.archlinux.org

[pacman-dev] [PATCH] makepkg: remove build date from .PKGINFO header
by Allan McRae 17 Apr '17

17 Apr '17

This information is duplicated (in less friendly format) in the "builddate" entry and removing it improves reproducible packaging. Signed-off-by: Allan McRae <allan(a)archlinux.org> --- scripts/makepkg.sh.in | 1 - 1 file changed, 1 deletion(-) diff --git a/scripts/makepkg.sh.in b/scripts/makepkg.sh.in index 43c7e328..c019ae3b 100644 --- a/scripts/makepkg.sh.in +++ b/scripts/makepkg.sh.in @@ -635,7 +635,6 @@ write_pkginfo() { msg2 "$(gettext "Generating %s file...")" ".PKGINFO" printf "# Generated by makepkg %s\n" "$makepkg_version" printf "# using %s\n" "$(fakeroot -v)" - printf "# %s\n" "$(LC_ALL=C date -u)" write_kv_pair "pkgname" "$pkgname" if (( SPLITPKG )) || [[ "$pkgbase" != "$pkgname" ]]; then -- 2.12.0

1 0

[pacman-dev] [PATCH] be_sync: error out if a db cannot be parsed
by Andrew Gregory 17 Apr '17

17 Apr '17

Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> --- See FS#49342 for an example of how to break a database in a way that was previously silently ignored. The linked github issue includes a copy of an actual broken db. lib/libalpm/be_sync.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-) diff --git a/lib/libalpm/be_sync.c b/lib/libalpm/be_sync.c index 5f7d31ab..06f509a6 100644 --- a/lib/libalpm/be_sync.c +++ b/lib/libalpm/be_sync.c @@ -465,6 +465,7 @@ static int sync_db_populate(alpm_db_t *db) size_t est_count, count; int fd; int ret = 0; + int archive_ret; struct stat buf; struct archive *archive; struct archive_entry *entry; @@ -502,20 +503,26 @@ static int sync_db_populate(alpm_db_t *db) goto cleanup; } - while(archive_read_next_header(archive, &entry) == ARCHIVE_OK) { + while((archive_ret = archive_read_next_header(archive, &entry)) == ARCHIVE_OK) { mode_t mode = archive_entry_mode(entry); - if(S_ISDIR(mode)) { - continue; - } else { + if(!S_ISDIR(mode)) { /* we have desc, depends or deltas - parse it */ if(sync_db_read(db, archive, entry, &pkg) != 0) { _alpm_log(db->handle, ALPM_LOG_ERROR, _("could not parse package description file '%s' from db '%s'\n"), archive_entry_pathname(entry), db->treename); - continue; + ret = -1; } } } + if(archive_ret != ARCHIVE_EOF) { + _alpm_log(db->handle, ALPM_LOG_ERROR, _("could not read db '%s' (%s)\n"), + db->treename, archive_error_string(archive)); + _alpm_db_free_pkgcache(db); + db->handle->pm_errno = ALPM_ERR_LIBARCHIVE; + ret = -1; + goto cleanup; + } count = alpm_list_count(db->pkgcache->list); if(count > 0) { -- 2.12.2

2 1

[pacman-dev] [PATCH 1/2] makepkg: do not create symlinks in build directory
by Allan McRae 17 Apr '17

17 Apr '17

Setting PKGDEST and friends enables us to keep all built packages in a single location. Symlinking these files into the build directory creates unnecessary clutter and requires clean-up in multiple places when removing old version. Signed-off-by: Allan McRae <allan(a)archlinux.org> --- Does anyone use these symlinks? Are they needed? Removing this makes the next patch much easier... scripts/makepkg.sh.in | 28 ---------------------------- 1 file changed, 28 deletions(-) diff --git a/scripts/makepkg.sh.in b/scripts/makepkg.sh.in index 0218e13b..be0ea72e 100644 --- a/scripts/makepkg.sh.in +++ b/scripts/makepkg.sh.in @@ -771,20 +771,6 @@ create_package() { fi create_signature "$pkg_file" - - if (( ! ret )) && [[ ! "$PKGDEST" -ef "${startdir}" ]]; then - rm -f "${pkg_file/$PKGDEST/$startdir}" - ln -s "${pkg_file}" "${pkg_file/$PKGDEST/$startdir}" - ret=$? - if [[ -f $pkg_file.sig ]]; then - rm -f "${pkg_file/$PKGDEST/$startdir}.sig" - ln -s "$pkg_file.sig" "${pkg_file/$PKGDEST/$startdir}.sig" - fi - fi - - if (( ret )); then - warning "$(gettext "Failed to create symlink to package file.")" - fi } create_debug_package() { @@ -884,20 +870,6 @@ create_srcpackage() { create_signature "$pkg_file" - if [[ ! "$SRCPKGDEST" -ef "${startdir}" ]]; then - rm -f "${pkg_file/$SRCPKGDEST/$startdir}" - ln -s "${pkg_file}" "${pkg_file/$SRCPKGDEST/$startdir}" - ret=$? - if [[ -f $pkg_file.sig ]]; then - rm -f "${pkg_file/$SRCPKGDEST/$startdir}.sig" - ln -s "$pkg_file.sig" "${pkg_file/$SRCPKGDEST/$startdir}.sig" - fi - fi - - if (( ret )); then - warning "$(gettext "Failed to create symlink to source package file.")" - fi - cd_safe "${startdir}" rm -rf "${srclinks}" } -- 2.12.0

1 1

[pacman-dev] [GIT] The official pacman repository branch, master, updated. v5.0.1-161-g4f2fea24
by allan＠archlinux.org 17 Apr '17

17 Apr '17

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 4f2fea240d3039294f6614003206a3dd1f67cfc5 (commit) via 3c013f68ac06d791830fd722b17bc771df47d436 (commit) via fae33a1faf3f94ea46049664ef483b2a3e0d3f01 (commit) via 6a4c6a02de9b45abe4c0f78c4f5d14d92d3359d6 (commit) via 6d1dcf7937ba3739d17eda614591d1f59ac0bf3a (commit) via 16b91f798faf6413f9b043543c5c0c3467b6fdbf (commit) via f9de3b22ad9e532af14c8d5b0b90a3f0f747e331 (commit) via a8670304c9393cbff350fcff7234d8f166dbcf60 (commit) via 098e114f586df383e5c3863fadbf4958b5b244e6 (commit) via 1550938ce1a081c4edc058b55ee34b320b849299 (commit) via 9c763a0d1ba8aa56b1ff54cd8d137c6b73e19f11 (commit) from 13ec13c85e4ea39e51d58b0842d8bef50e2ae369 (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 4f2fea240d3039294f6614003206a3dd1f67cfc5 Author: Michael Straube <straubem(a)gmx.de> Date: Wed Apr 12 16:53:53 2017 +0200 makepkg: clean up pkgver and prepare log files Delete log files for the pkgver and prepare functions if -c,--clean is passed. Fixes FS#51039 and FS#51075 Includes patch submitted by Christian Braun. Signed-off-by: Michael Straube <straubem(a)gmx.de> Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 3c013f68ac06d791830fd722b17bc771df47d436 Author: Andrew Gregory <andrew.gregory.8(a)gmail.com> Date: Sun Apr 16 08:36:32 2017 -0400 run_chroot: label pipe endpoints for readability Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> commit fae33a1faf3f94ea46049664ef483b2a3e0d3f01 Author: Andrew Gregory <andrew.gregory.8(a)gmail.com> Date: Sun Apr 16 00:31:57 2017 -0400 add --sysroot option --root is not sufficient to properly operate on a mounted guest system. Using --root still uses the host system's configuration and there is no way to correctly use the guest configuration without manually modifying any Include directives. --sysroot provides an easier way to operate on a guest system by chrooting immediately after option parsing before configuration parsing or performing any operations. It is currently limited to the root user, but that's enough for restoring a guest system to a working state, which is the primary intended use case. Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> commit 6a4c6a02de9b45abe4c0f78c4f5d14d92d3359d6 Author: Andrew Gregory <andrew.gregory.8(a)gmail.com> Date: Sat Apr 15 18:22:51 2017 -0400 use sockets for scriptlet/hook communication If a scriptlet/hook dies at the wrong moment it can trigger SIGPIPE, terminating the process. For pipes, there is no way to prevent SIGPIPE other than ignoring it process-wide. This can have unintended consequences in a multi-threaded process. Using send(2) with sockets, however, allows ignoring SIGPIPE on a per-call basis, leaving other threads able to make use of SIGPIPE. Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> commit 6d1dcf7937ba3739d17eda614591d1f59ac0bf3a Author: Andrew Gregory <andrew.gregory.8(a)gmail.com> Date: Sat Apr 15 18:12:56 2017 -0400 conflict: include owner for filesystem conflicts Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> commit 16b91f798faf6413f9b043543c5c0c3467b6fdbf Author: Andrew Gregory <andrew.gregory.8(a)gmail.com> Date: Sun Apr 9 19:49:17 2017 -0400 unlink_file: strip trailing slashes If the user replaces a directory with a symlink, libalpm would get confused because the trailing slash causes system calls to resolve the symlink. This leads to errors and a misleading message during upgrades. Even though libalpm does not support this, it should not be giving misleading errors. Also adds an overflow check. Fixes FS#51377 Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> commit f9de3b22ad9e532af14c8d5b0b90a3f0f747e331 Author: Andrew Gregory <andrew.gregory.8(a)gmail.com> Date: Sat Apr 15 17:29:07 2017 -0400 sortbydeps: rename vptr -> i vptr is a simple list iterator, which are typically named i. Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> commit a8670304c9393cbff350fcff7234d8f166dbcf60 Author: Andrew Gregory <andrew.gregory.8(a)gmail.com> Date: Sat Apr 15 17:28:37 2017 -0400 sortbydeps: rename found -> switched_to_child Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> commit 098e114f586df383e5c3863fadbf4958b5b244e6 Author: Andrew Gregory <andrew.gregory.8(a)gmail.com> Date: Sat Apr 15 17:26:35 2017 -0400 sortbydeps: factor out dep cycle warning Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> commit 1550938ce1a081c4edc058b55ee34b320b849299 Author: Andrew Gregory <andrew.gregory.8(a)gmail.com> Date: Sat Apr 15 14:33:50 2017 -0400 graph.h: rename childptr -> iterator Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> commit 9c763a0d1ba8aa56b1ff54cd8d137c6b73e19f11 Author: Andrew Gregory <andrew.gregory.8(a)gmail.com> Date: Sat Apr 15 14:11:43 2017 -0400 graph.h: replace hardcoded values with an enum Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> ----------------------------------------------------------------------- Summary of changes: doc/pacman.8.txt | 11 ++- lib/libalpm/conflict.c | 25 ++++- lib/libalpm/delta.c | 14 +-- lib/libalpm/deps.c | 104 +++++++++++---------- lib/libalpm/graph.h | 10 +- lib/libalpm/remove.c | 13 ++- lib/libalpm/util.c | 46 +++++---- scripts/makepkg.sh.in | 6 ++ src/pacman/conf.h | 2 + src/pacman/pacman.c | 24 +++-- src/pacman/sync.c | 9 +- src/pacman/util.c | 3 + test/pacman/tests/TESTS | 2 + .../tests/file-conflict-with-installed-pkg.py | 17 ++++ .../remove-directory-replaced-with-symlink.py | 16 ++++ 15 files changed, 202 insertions(+), 100 deletions(-) create mode 100644 test/pacman/tests/file-conflict-with-installed-pkg.py create mode 100644 test/pacman/tests/remove-directory-replaced-with-symlink.py hooks/post-receive -- The official pacman repository

1 0

[pacman-dev] [PATCH] run_chroot: label pipe endpoints for readability
by Andrew Gregory 16 Apr '17

16 Apr '17

Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> --- lib/libalpm/util.c | 27 +++++++++++++++------------ 1 file changed, 15 insertions(+), 12 deletions(-) diff --git a/lib/libalpm/util.c b/lib/libalpm/util.c index 58efa5c7..36d8a7d7 100644 --- a/lib/libalpm/util.c +++ b/lib/libalpm/util.c @@ -573,6 +573,9 @@ int _alpm_run_chroot(alpm_handle_t *handle, const char *cmd, char *const argv[], int cwdfd; int retval = 0; +#define HEAD 1 +#define TAIL 0 + /* save the cwd so we can restore it later */ OPEN(cwdfd, ".", O_RDONLY | O_CLOEXEC); if(cwdfd < 0) { @@ -617,13 +620,13 @@ int _alpm_run_chroot(alpm_handle_t *handle, const char *cmd, char *const argv[], close(0); close(1); close(2); - while(dup2(child2parent_pipefd[1], 1) == -1 && errno == EINTR); - while(dup2(child2parent_pipefd[1], 2) == -1 && errno == EINTR); - while(dup2(parent2child_pipefd[0], 0) == -1 && errno == EINTR); - close(parent2child_pipefd[0]); - close(parent2child_pipefd[1]); - close(child2parent_pipefd[0]); - close(child2parent_pipefd[1]); + while(dup2(child2parent_pipefd[HEAD], 1) == -1 && errno == EINTR); + while(dup2(child2parent_pipefd[HEAD], 2) == -1 && errno == EINTR); + while(dup2(parent2child_pipefd[TAIL], 0) == -1 && errno == EINTR); + close(parent2child_pipefd[TAIL]); + close(parent2child_pipefd[HEAD]); + close(child2parent_pipefd[TAIL]); + close(child2parent_pipefd[HEAD]); if(cwdfd >= 0) { close(cwdfd); } @@ -652,20 +655,20 @@ int _alpm_run_chroot(alpm_handle_t *handle, const char *cmd, char *const argv[], nfds_t nfds = 2; struct pollfd fds[2], *child2parent = &(fds[0]), *parent2child = &(fds[1]); - child2parent->fd = child2parent_pipefd[0]; + child2parent->fd = child2parent_pipefd[TAIL]; child2parent->events = POLLIN; fcntl(child2parent->fd, F_SETFL, O_NONBLOCK); - close(child2parent_pipefd[1]); - close(parent2child_pipefd[0]); + close(child2parent_pipefd[HEAD]); + close(parent2child_pipefd[TAIL]); if(stdin_cb) { - parent2child->fd = parent2child_pipefd[1]; + parent2child->fd = parent2child_pipefd[HEAD]; parent2child->events = POLLOUT; fcntl(parent2child->fd, F_SETFL, O_NONBLOCK); } else { parent2child->fd = -1; parent2child->events = 0; - close(parent2child_pipefd[1]); + close(parent2child_pipefd[HEAD]); } #define STOP_POLLING(p) do { close(p->fd); p->fd = -1; } while(0) -- 2.12.2

2 2

[pacman-dev] [PATCH 1/5] graph.h: replace hardcoded values with an enum
by Andrew Gregory 16 Apr '17

16 Apr '17

Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> --- This series is an attempt to clean up sortbydeps primarily by giving variables better names and factoring out the dependency cycle warning. There shouldn't be any change in behavior. lib/libalpm/delta.c | 4 ++-- lib/libalpm/deps.c | 10 +++++----- lib/libalpm/graph.h | 8 +++++++- 3 files changed, 14 insertions(+), 8 deletions(-) diff --git a/lib/libalpm/delta.c b/lib/libalpm/delta.c index c5571b29..58b1b41e 100644 --- a/lib/libalpm/delta.c +++ b/lib/libalpm/delta.c @@ -130,7 +130,7 @@ static void dijkstra(alpm_list_t *vertices) for(i = vertices; i; i = i->next) { alpm_graph_t *v_i = i->data; - if(v_i->state == -1) { + if(v_i->state == ALPM_GRAPH_STATE_PROCESSING) { continue; } @@ -142,7 +142,7 @@ static void dijkstra(alpm_list_t *vertices) break; } - v->state = -1; + v->state = ALPM_GRAPH_STATE_PROCESSING; v->childptr = v->children; while(v->childptr) { diff --git a/lib/libalpm/deps.c b/lib/libalpm/deps.c index 40b2be6d..e5527d6c 100644 --- a/lib/libalpm/deps.c +++ b/lib/libalpm/deps.c @@ -194,16 +194,16 @@ alpm_list_t *_alpm_sortbydeps(alpm_handle_t *handle, vertex = vertices->data; while(vptr) { /* mark that we touched the vertex */ - vertex->state = -1; + vertex->state = ALPM_GRAPH_STATE_PROCESSING; int found = 0; while(vertex->childptr && !found) { alpm_graph_t *nextchild = vertex->childptr->data; vertex->childptr = vertex->childptr->next; - if(nextchild->state == 0) { + if(nextchild->state == ALPM_GRAPH_STATE_UNPROCESSED) { found = 1; nextchild->parent = vertex; vertex = nextchild; - } else if(nextchild->state == -1) { + } else if(nextchild->state == ALPM_GRAPH_STATE_PROCESSING) { /* child is an ancestor of vertex */ alpm_graph_t *transvertex = vertex; @@ -244,13 +244,13 @@ alpm_list_t *_alpm_sortbydeps(alpm_handle_t *handle, newtargs = alpm_list_add(newtargs, vertex->data); } /* mark that we've left this vertex */ - vertex->state = 1; + vertex->state = ALPM_GRAPH_STATE_PROCESSED; vertex = vertex->parent; if(!vertex) { /* top level vertex reached, move to the next unprocessed vertex */ for( vptr = vptr->next; vptr; vptr = vptr->next) { vertex = vptr->data; - if(vertex->state == 0) { + if(vertex->state == ALPM_GRAPH_STATE_UNPROCESSED) { break; } } diff --git a/lib/libalpm/graph.h b/lib/libalpm/graph.h index 2e9eac10..f1da837c 100644 --- a/lib/libalpm/graph.h +++ b/lib/libalpm/graph.h @@ -23,13 +23,19 @@ #include "alpm_list.h" +enum __alpm_graph_vertex_state { + ALPM_GRAPH_STATE_UNPROCESSED = 0, + ALPM_GRAPH_STATE_PROCESSING = -1, + ALPM_GRAPH_STATE_PROCESSED = 1 +}; + typedef struct __alpm_graph_t { void *data; struct __alpm_graph_t *parent; /* where did we come from? */ alpm_list_t *children; alpm_list_t *childptr; /* points to a child in children list */ off_t weight; /* weight of the node */ - signed char state; /* 0: untouched, -1: entered, other: leaving time */ + signed char state; } alpm_graph_t; alpm_graph_t *_alpm_graph_new(void); -- 2.12.2

2 7

[pacman-dev] [RFC] adding a --sysroot option
by Andrew Gregory 16 Apr '17

16 Apr '17

pacman's --root option is regularly (mis)used to use pacman to manage a mounted guest system, typically one whose pacman installation is currently broken. We have a few configuration defaults in place to make this sort of work, but support is incomplete. Those defaults only actually take effect if the settings haven't been set in a configuration file, several options still default to the host system resources, and using the guest's pacman configuration requires updating all configured paths to the new mounted location. Adding a --sysroot/--chroot option would allow pacman to properly operate in a mounted guest system. At the moment, there are two ways we could accomplish this: prefix all paths with the sysroot or just call chroot(2). Obviously, the problem with chroot(2) is that it requires special privileges. Unfortunately, I think my symbolic user/group patch (https://patchwork.archlinux.org/patch/3694/) will require chroot to work properly as I can't find any other way to look up users/groups in a mounted guest. So, we may have to implement both approaches so that regular users can perform queries but privileged users can perform transactions with proper symbolic name support. apg

2 5

[pacman-dev] [PATCH] use sockets for scriptlet/hook communication
by Andrew Gregory 15 Apr '17

15 Apr '17

If a scriptlet/hook dies at the wrong moment it can trigger SIGPIPE, terminating the process. For pipes, there is no way to prevent SIGPIPE other than ignoring it process-wide. This can have unintended consequences in a multi-threaded process. Using send(2) with sockets, however, allows ignoring SIGPIPE on a per-call basis, leaving other threads able to make use of SIGPIPE. Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> --- lib/libalpm/util.c | 17 ++++------------- 1 file changed, 4 insertions(+), 13 deletions(-) diff --git a/lib/libalpm/util.c b/lib/libalpm/util.c index 58efa5c7..5a7f2c87 100644 --- a/lib/libalpm/util.c +++ b/lib/libalpm/util.c @@ -30,6 +30,7 @@ #include <errno.h> #include <limits.h> #include <sys/wait.h> +#include <sys/socket.h> #include <fnmatch.h> #include <poll.h> @@ -463,7 +464,6 @@ static int _alpm_chroot_write_to_child(alpm_handle_t *handle, int fd, _alpm_cb_io out_cb, void *cb_ctx) { ssize_t nwrite; - struct sigaction newaction, oldaction; if(*buf_size == 0) { /* empty buffer, ask the callback for more */ @@ -473,16 +473,7 @@ static int _alpm_chroot_write_to_child(alpm_handle_t *handle, int fd, } } - /* ignore SIGPIPE in case the pipe has been closed */ - newaction.sa_handler = SIG_IGN; - sigemptyset(&newaction.sa_mask); - newaction.sa_flags = 0; - sigaction(SIGPIPE, &newaction, &oldaction); - - nwrite = write(fd, buf, *buf_size); - - /* restore previous SIGPIPE handler */ - sigaction(SIGPIPE, &oldaction, NULL); + nwrite = send(fd, buf, *buf_size, MSG_NOSIGNAL); if(nwrite != -1) { /* write was successful, remove the written data from the buffer */ @@ -592,13 +583,13 @@ int _alpm_run_chroot(alpm_handle_t *handle, const char *cmd, char *const argv[], /* Flush open fds before fork() to avoid cloning buffers */ fflush(NULL); - if(pipe(child2parent_pipefd) == -1) { + if(socketpair(AF_UNIX, SOCK_STREAM, 0, child2parent_pipefd) == -1) { _alpm_log(handle, ALPM_LOG_ERROR, _("could not create pipe (%s)\n"), strerror(errno)); retval = 1; goto cleanup; } - if(pipe(parent2child_pipefd) == -1) { + if(socketpair(AF_UNIX, SOCK_STREAM, 0, parent2child_pipefd) == -1) { _alpm_log(handle, ALPM_LOG_ERROR, _("could not create pipe (%s)\n"), strerror(errno)); retval = 1; goto cleanup; -- 2.12.2

1 0

[pacman-dev] [PATCH] conflict: include owner for filesystem conflicts
by Andrew Gregory 15 Apr '17

15 Apr '17

Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> --- lib/libalpm/conflict.c | 25 +++++++++++++++++----- src/pacman/sync.c | 9 ++++++-- test/pacman/tests/TESTS | 1 + .../tests/file-conflict-with-installed-pkg.py | 17 +++++++++++++++ 4 files changed, 45 insertions(+), 7 deletions(-) create mode 100644 test/pacman/tests/file-conflict-with-installed-pkg.py diff --git a/lib/libalpm/conflict.c b/lib/libalpm/conflict.c index 0e37419f..d52f8942 100644 --- a/lib/libalpm/conflict.c +++ b/lib/libalpm/conflict.c @@ -278,12 +278,15 @@ static alpm_list_t *add_fileconflict(alpm_handle_t *handle, STRDUP(conflict->target, pkg1->name, goto error); STRDUP(conflict->file, filestr, goto error); - if(pkg2) { - conflict->type = ALPM_FILECONFLICT_TARGET; - STRDUP(conflict->ctarget, pkg2->name, goto error); - } else { + if(!pkg2) { conflict->type = ALPM_FILECONFLICT_FILESYSTEM; STRDUP(conflict->ctarget, "", goto error); + } else if(pkg2->origin == ALPM_PKG_FROM_LOCALDB) { + conflict->type = ALPM_FILECONFLICT_FILESYSTEM; + STRDUP(conflict->ctarget, pkg2->name, goto error); + } else { + conflict->type = ALPM_FILECONFLICT_TARGET; + STRDUP(conflict->ctarget, pkg2->name, goto error); } conflicts = alpm_list_add(conflicts, conflict); @@ -385,6 +388,17 @@ static alpm_list_t *alpm_db_find_file_owners(alpm_db_t* db, const char *path) return owners; } +static alpm_pkg_t *_alpm_find_file_owner(alpm_handle_t *handle, const char *path) +{ + alpm_list_t *i; + for(i = alpm_db_get_pkgcache(handle->db_local); i; i = i->next) { + if(alpm_filelist_contains(alpm_pkg_get_files(i->data), path)) { + return i->data; + } + } + return NULL; +} + static int _alpm_can_overwrite_file(alpm_handle_t *handle, const char *path) { return handle->trans->flags & ALPM_TRANS_FLAG_FORCE @@ -668,7 +682,8 @@ alpm_list_t *_alpm_db_find_fileconflicts(alpm_handle_t *handle, } if(!resolved_conflict) { - conflicts = add_fileconflict(handle, conflicts, path, p1, NULL); + conflicts = add_fileconflict(handle, conflicts, path, p1, + _alpm_find_file_owner(handle, relative_path)); if(handle->pm_errno == ALPM_ERR_MEMORY) { alpm_list_free_inner(conflicts, (alpm_list_fn_free) alpm_conflict_free); diff --git a/src/pacman/sync.c b/src/pacman/sync.c index e37b91c7..680d7b3e 100644 --- a/src/pacman/sync.c +++ b/src/pacman/sync.c @@ -823,8 +823,13 @@ int sync_prepare_execute(void) conflict->file, conflict->target, conflict->ctarget); break; case ALPM_FILECONFLICT_FILESYSTEM: - printf(_("%s: %s exists in filesystem\n"), - conflict->target, conflict->file); + if(conflict->ctarget[0]) { + printf(_("%s: %s exists in filesystem (owned by %s)\n"), + conflict->target, conflict->file, conflict->ctarget); + } else { + printf(_("%s: %s exists in filesystem\n"), + conflict->target, conflict->file); + } break; } alpm_fileconflict_free(conflict); diff --git a/test/pacman/tests/TESTS b/test/pacman/tests/TESTS index eadb63b4..9dbeb767 100644 --- a/test/pacman/tests/TESTS +++ b/test/pacman/tests/TESTS @@ -28,6 +28,7 @@ TESTS += test/pacman/tests/epoch005.py TESTS += test/pacman/tests/epoch010.py TESTS += test/pacman/tests/epoch011.py TESTS += test/pacman/tests/epoch012.py +TESTS += test/pacman/tests/file-conflict-with-installed-pkg.py TESTS += test/pacman/tests/fileconflict001.py TESTS += test/pacman/tests/fileconflict002.py TESTS += test/pacman/tests/fileconflict003.py diff --git a/test/pacman/tests/file-conflict-with-installed-pkg.py b/test/pacman/tests/file-conflict-with-installed-pkg.py new file mode 100644 index 00000000..cf88e8a9 --- /dev/null +++ b/test/pacman/tests/file-conflict-with-installed-pkg.py @@ -0,0 +1,17 @@ +self.description = "File conflict with an installed package" + +lp = pmpkg("foobar") +lp.files = ["conflicting-file"] +self.addpkg2db("local", lp) + +p1 = pmpkg("pkg1") +p1.files = ["conflicting-file"] +self.addpkg(p1) + +self.args = "-U %s" % (p1.filename()) + +self.addrule("!PACMAN_RETCODE=0") +self.addrule("PKG_EXIST=foobar") +self.addrule("!PKG_EXIST=pkg1") +self.addrule("FILE_EXIST=conflicting-file") +self.addrule("PACMAN_OUTPUT=foobar") -- 2.12.2

1 0

[pacman-dev] [PATCH] makepkg: clean up pkgver and prepare log files
by Michael Straube 12 Apr '17

12 Apr '17

Delete log files for the pkgver and prepare functions if -c,--clean is passed. Fixes FS#51039 and FS#51075 Includes patch submitted by Christian Braun. Signed-off-by: Michael Straube <straubem(a)gmx.de> --- scripts/makepkg.sh.in | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/scripts/makepkg.sh.in b/scripts/makepkg.sh.in index 29408929..0218e13b 100644 --- a/scripts/makepkg.sh.in +++ b/scripts/makepkg.sh.in @@ -137,6 +137,12 @@ clean_up() { if [[ -n $pkgbase ]]; then local fullver=$(get_full_version) # Can't do this unless the BUILDSCRIPT has been sourced. + if (( PKGVERFUNC )); then + rm -f "${pkgbase}-${fullver}-${CARCH}-pkgver.log"* + fi + if (( PREPAREFUNC )); then + rm -f "${pkgbase}-${fullver}-${CARCH}-prepare.log"* + fi if (( BUILDFUNC )); then rm -f "${pkgbase}-${fullver}-${CARCH}-build.log"* fi -- 2.12.2

1 0