June 2011 - Pacman-dev - lists.archlinux.org

[pacman-dev] GPG remote signing
by Dan McGee 11 Jun '11

11 Jun '11

I've done a fair amount of research on what we might be able to do with this during the afternoon here. Some observations below. This is mainly addressing point four in Thomas' prior email (http://mailman.archlinux.org/mailman/private/arch-dev/2011-May/014193.html). There are hacks for forwarding a gpg-agent socket involving socat (original one here http://superuser.com/questions/161973/how-can-i-forward-a-gpg-key-via-ssh-a…, more secure one here http://www.debian-administration.org/users/dkg/weblog/68). The problem? gpg 2.0.x and down (the current version) store only passphrases in gpg-agent; either way, the gpg2/gpg binary itself still needs access to the secure keyring to perform the encryption and generate a signature for the file. So this seems like a dead end for the time being. Rumor is 2.1 will fix this problem (http://www.debian-administration.org/users/dkg/weblog/68#comment_5), but I don't think we're in a position to wait for that. I've seen suggestions revolving around signing remote files that say the following instead- rather than sign the large file directly by either putting your key out there, or pulling the big file to you, sign a secure digest of the file. Something such as dropping the SHA512 hash into a file, copying that to the machine with the key, and then signing that instead. In a way, this is what GPG signing does anyway (gpg --edit-key <yours>, showprefs, look at the "Digest" line; also, the --show-mds option). The problem is GPG doesn't allow you to separate the digest generation from the actual signing process, which sucks. This doesn't fully work for us- at least right now. I see a few possibilities as to how to proceed. 1) suck it up, sign your packages locally. 2) live dangerously, copy your secret key out, use it to sign, delete it from that location (you better have a passphrase). 3) do the above "sign the hash trick"; one would then upload the package from location A, and the "special signature", we'll call it, from location B. When these two files are on the main server, a master key/verification program (running in a walled-off VM with very little access) would basically say "is this special signature valid, and is it blessed by our trust", and if so, truly sign the package (and the special signature would not longer be needed). This is somewhat similar to what Debian does with their dsc files (signing a little more than a single hash, see for yourself: http://ftp.us.debian.org/debian/pool/main/m/magit/magit_0.7-1.1.dsc) 4) do the above "sign the hash trick"; and teach pacman how to verify both "real" signatures and "special" signatures on files. 5) ??? Thoughts? Other ideas? Things I'm forgetting? I'll withhold my preference of option for now to prevent biasing any comments; the above should not be seen as order of preference. -Dan

2 1

[pacman-dev] Status of sodeps?
by Florian Pritz 10 Jun '11

10 Jun '11

It's been quite a while since I last brought this up so what's the current status? Older discussions and the link to my repo are here [1]. [1] https://wiki.archlinux.org/index.php/Pacman_Roadmap#sodepends -- Florian Pritz -- {flo,bluewind}(a)server-speed.net

5 11

[pacman-dev] [PATCH] Add handle argument to alpm_(add|remove)_pkg()
by Dan McGee 10 Jun '11

10 Jun '11

This makes these functions consistent with the rest of the transaction related API calls. We do an additional assert to ensure the handle attached to the package is the same as the handle passed in. Signed-off-by: Dan McGee <dan(a)archlinux.org> --- lib/libalpm/add.c | 8 ++++---- lib/libalpm/alpm.h | 6 ++++-- lib/libalpm/remove.c | 8 ++++---- src/pacman/remove.c | 4 ++-- src/pacman/sync.c | 2 +- src/pacman/upgrade.c | 2 +- 6 files changed, 16 insertions(+), 14 deletions(-) diff --git a/lib/libalpm/add.c b/lib/libalpm/add.c index abce175..6f93c61 100644 --- a/lib/libalpm/add.c +++ b/lib/libalpm/add.c @@ -50,16 +50,16 @@ #include "handle.h" /** Add a package to the transaction. */ -int SYMEXPORT alpm_add_pkg(pmpkg_t *pkg) +int SYMEXPORT alpm_add_pkg(pmhandle_t *handle, pmpkg_t *pkg) { const char *pkgname, *pkgver; - pmhandle_t *handle; pmtrans_t *trans; pmpkg_t *local; /* Sanity checks */ - ASSERT(pkg != NULL, return -1); - handle = pkg->handle; + ASSERT(handle != NULL, return -1); + ASSERT(pkg != NULL, RET_ERR(handle, PM_ERR_WRONG_ARGS, -1)); + ASSERT(handle == pkg->handle, RET_ERR(handle, PM_ERR_WRONG_ARGS, -1)); trans = handle->trans; ASSERT(trans != NULL, RET_ERR(handle, PM_ERR_TRANS_NULL, -1)); ASSERT(trans->state == STATE_INITIALIZED, diff --git a/lib/libalpm/alpm.h b/lib/libalpm/alpm.h index 37bea6d..b62c62d 100644 --- a/lib/libalpm/alpm.h +++ b/lib/libalpm/alpm.h @@ -873,16 +873,18 @@ int alpm_sync_sysupgrade(pmhandle_t *handle, int enable_downgrade); /** Add a package to the transaction. * If the package was loaded by alpm_pkg_load(), it will be freed upon * alpm_trans_release() invocation. + * @param handle the context handle * @param pkg the package to add * @return 0 on success, -1 on error (pm_errno is set accordingly) */ -int alpm_add_pkg(pmpkg_t *pkg); +int alpm_add_pkg(pmhandle_t *handle, pmpkg_t *pkg); /** Add a package removal action to the transaction. + * @param handle the context handle * @param pkg the package to uninstall * @return 0 on success, -1 on error (pm_errno is set accordingly) */ -int alpm_remove_pkg(pmpkg_t *pkg); +int alpm_remove_pkg(pmhandle_t *handle, pmpkg_t *pkg); /** @} */ diff --git a/lib/libalpm/remove.c b/lib/libalpm/remove.c index 8a1bbcb..52f60fc 100644 --- a/lib/libalpm/remove.c +++ b/lib/libalpm/remove.c @@ -44,15 +44,15 @@ #include "deps.h" #include "handle.h" -int SYMEXPORT alpm_remove_pkg(pmpkg_t *pkg) +int SYMEXPORT alpm_remove_pkg(pmhandle_t *handle, pmpkg_t *pkg) { const char *pkgname; - pmhandle_t *handle; pmtrans_t *trans; /* Sanity checks */ - ASSERT(pkg != NULL, return -1); - handle = pkg->handle; + ASSERT(handle != NULL, return -1); + ASSERT(pkg != NULL, RET_ERR(handle, PM_ERR_WRONG_ARGS, -1)); + ASSERT(handle == pkg->handle, RET_ERR(handle, PM_ERR_WRONG_ARGS, -1)); trans = handle->trans; ASSERT(trans != NULL, RET_ERR(handle, PM_ERR_TRANS_NULL, -1)); ASSERT(trans->state == STATE_INITIALIZED, diff --git a/src/pacman/remove.c b/src/pacman/remove.c index 595263a..2c4a3fc 100644 --- a/src/pacman/remove.c +++ b/src/pacman/remove.c @@ -38,7 +38,7 @@ static int remove_target(const char *target) alpm_list_t *p; if((info = alpm_db_get_pkg(db_local, target)) != NULL) { - if(alpm_remove_pkg(info) == -1) { + if(alpm_remove_pkg(config->handle, info) == -1) { pm_fprintf(stderr, PM_LOG_ERROR, "'%s': %s\n", target, alpm_strerror(alpm_errno(config->handle))); return -1; @@ -54,7 +54,7 @@ static int remove_target(const char *target) } for(p = alpm_grp_get_pkgs(grp); p; p = alpm_list_next(p)) { pmpkg_t *pkg = alpm_list_getdata(p); - if(alpm_remove_pkg(pkg) == -1) { + if(alpm_remove_pkg(config->handle, pkg) == -1) { pm_fprintf(stderr, PM_LOG_ERROR, "'%s': %s\n", target, alpm_strerror(alpm_errno(config->handle))); return -1; diff --git a/src/pacman/sync.c b/src/pacman/sync.c index 36c7799..539e651 100644 --- a/src/pacman/sync.c +++ b/src/pacman/sync.c @@ -616,7 +616,7 @@ static pmdb_t *get_db(const char *dbname) static int process_pkg(pmpkg_t *pkg) { - int ret = alpm_add_pkg(pkg); + int ret = alpm_add_pkg(config->handle, pkg); if(ret == -1) { enum _pmerrno_t err = alpm_errno(config->handle); diff --git a/src/pacman/upgrade.c b/src/pacman/upgrade.c index ef273cf..35624eb 100644 --- a/src/pacman/upgrade.c +++ b/src/pacman/upgrade.c @@ -82,7 +82,7 @@ int pacman_upgrade(alpm_list_t *targets) trans_release(); return 1; } - if(alpm_add_pkg(pkg) == -1) { + if(alpm_add_pkg(config->handle, pkg) == -1) { pm_fprintf(stderr, PM_LOG_ERROR, "'%s': %s\n", targ, alpm_strerror(alpm_errno(config->handle))); alpm_pkg_free(pkg); -- 1.7.5.2

1 0

[pacman-dev] [GIT] The official pacman repository branch, master, updated. v3.5.3-216-gff7ad5f
by dan＠archlinux.org 10 Jun '11

10 Jun '11

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 ff7ad5fd7376c133bc646c4d69d06019e0b64044 (commit) via ff8e519d4b011a6f23c0366d45c396fcd1f9af97 (commit) via 7d27cf8364b9c5d9a8d89017173f8c88d40cedd5 (commit) via 24000b83c9a9ba2f25a46914d2919293dc865a2e (commit) via 17a6ac567502975d3a98a34ed58d79c05eb7b8d1 (commit) via 7968d30510de5a6031af39da498be5c821290b82 (commit) via e826c143d3e9d21485f72c5490b2a907c936024e (commit) via 9d6568da0f55fb05ea5c28e06efb51046fd270bf (commit) via 451cd2c88dcdec793ce7159331adcc2d049381e1 (commit) from dfaeb6bb2ca8311ac418a16c30006eb7d0534d71 (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 ff7ad5fd7376c133bc646c4d69d06019e0b64044 Author: Dan McGee <dan(a)archlinux.org> Date: Thu Jun 9 15:15:15 2011 -0500 commit_single_pkg(): Use handle object directly Commit e68f5d9a3067141 did something a bit silly and changed the scriptlet calls to use 'newpkg->handle' rather than the 'handle' argument passed in. Use the handle directly. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit ff8e519d4b011a6f23c0366d45c396fcd1f9af97 Author: Dan McGee <dan(a)archlinux.org> Date: Tue Jun 7 13:39:52 2011 -0500 Require handle for alpm_sync_sysupgrade() Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 7d27cf8364b9c5d9a8d89017173f8c88d40cedd5 Author: Dan McGee <dan(a)archlinux.org> Date: Tue Jun 7 13:34:30 2011 -0500 Require handle for alpm_db_register_sync() Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 24000b83c9a9ba2f25a46914d2919293dc865a2e Author: Dan McGee <dan(a)archlinux.org> Date: Tue Jun 7 13:29:05 2011 -0500 Require handle argument to all alpm_trans_*() methods Begin enforcing the need to pass a handle. This allows us to remove one more extern handle declaration from the backend. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 17a6ac567502975d3a98a34ed58d79c05eb7b8d1 Author: Dan McGee <dan(a)archlinux.org> Date: Tue Jun 7 13:15:43 2011 -0500 Require handle argument to all alpm_option_(get|set)_*() methods This requires a lot of line changes, but not many functional changes as more often than not our handle variable is already available in some fashion. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 7968d30510de5a6031af39da498be5c821290b82 Author: Dan McGee <dan(a)archlinux.org> Date: Tue Jun 7 12:41:33 2011 -0500 Require handle argument to alpm_logaction() This is the first in a series of patches to update the API to remove the implicit global handle variable. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit e826c143d3e9d21485f72c5490b2a907c936024e Author: Dan McGee <dan(a)archlinux.org> Date: Thu Jun 9 13:41:08 2011 -0500 Kill all remaining 'PATH_MAX + 1' usages The few remaining instances were utilized for buffers in calls to snprintf() and realpath(). Both of these functions will always ensure the returned value is padded with '\0', so there is no need for the extra byte. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 9d6568da0f55fb05ea5c28e06efb51046fd270bf Author: Dan McGee <dan(a)archlinux.org> Date: Thu Jun 9 13:34:13 2011 -0500 _alpm_lstat: only duplicate string if necessary The vast majority of the time we will just be passing the same string value on to the lstat() call. The only time we need to duplicate it is if the path ends in '/'. In one run using a profiler, only 400 of the 200,000 calls (0.2%) required the string to be copied first. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 451cd2c88dcdec793ce7159331adcc2d049381e1 Author: Allan McRae <allan(a)archlinux.org> Date: Wed Jun 8 23:49:57 2011 +1000 Fix bracket type in makefile Signed-off-by: Allan McRae <allan(a)archlinux.org> Signed-off-by: Dan McGee <dan(a)archlinux.org> ----------------------------------------------------------------------- Summary of changes: lib/libalpm/add.c | 48 +++++----- lib/libalpm/alpm.h | 132 +++++++++++++----------- lib/libalpm/be_local.c | 2 +- lib/libalpm/be_package.c | 2 +- lib/libalpm/be_sync.c | 2 +- lib/libalpm/conflict.c | 6 +- lib/libalpm/db.c | 7 +- lib/libalpm/delta.c | 11 +- lib/libalpm/delta.h | 2 +- lib/libalpm/deps.c | 6 +- lib/libalpm/diskspace.c | 2 +- lib/libalpm/dload.c | 6 +- lib/libalpm/handle.c | 251 ++++++++++++++++++---------------------------- lib/libalpm/log.c | 6 +- lib/libalpm/package.c | 4 +- lib/libalpm/remove.c | 6 +- lib/libalpm/signing.c | 19 ++-- lib/libalpm/signing.h | 3 +- lib/libalpm/sync.c | 33 +++--- lib/libalpm/trans.c | 31 +++--- lib/libalpm/util.c | 41 ++++---- lib/libalpm/util.h | 6 +- scripts/Makefile.am | 2 +- src/pacman/callback.c | 6 +- src/pacman/conf.c | 46 +++++---- src/pacman/database.c | 2 +- src/pacman/deptest.c | 3 +- src/pacman/package.c | 5 +- src/pacman/pacman.c | 20 ++-- src/pacman/query.c | 21 ++-- src/pacman/remove.c | 10 +- src/pacman/sync.c | 44 ++++---- src/pacman/upgrade.c | 16 ++-- src/pacman/util.c | 13 ++- src/util/cleanupdelta.c | 4 +- src/util/pactree.c | 6 +- src/util/testdb.c | 13 +-- src/util/testpkg.c | 2 +- 38 files changed, 396 insertions(+), 443 deletions(-) hooks/post-receive -- The official pacman repository

1 0

[pacman-dev] [PATCH 00/13] global handle removal patch series
by Dan McGee 10 Jun '11

10 Jun '11

This is a bitch of a diffstat, unfortunately, and the patches aren't all that fun to look through. I have it pushed to my repo as alpm-cleanups if you'd rather grab it there. The bright side is it should successfully build and test after each patch as long as they are applied in the order sent here (or in my repo). Comments/suggestion/feedback welcome. Once these are applied, I think our API makes a lot more sense from a consumer standpoint as we have a single object, tracked by the "client", tracking all state in the library rather than the state being held in a global variable in the backend library. This patch series (and the previous one sent and now applied) makes it *much* easier for the next few patches, which will ensure DBs are signature-checked and verified at the right time, and at a time all clients expect them to be checked so error codes can be checked. Look at the bright side- the overall line differential is at least negative! -Dan Dan McGee (13): Require handle argument to alpm_logaction() Require handle argument to all alpm_option_(get|set)_*() methods Require handle argument to all alpm_trans_*() methods Require handle for alpm_db_register_sync() Require handle for alpm_sync_sysupgrade() Require handle for alpm_pkg_load() Require handle for alpm_checkconflicts() Add handle argument to two more alpm methods Require handle for alpm_find_grp_pkgs() Move pm_errno onto the handle Switch all logging to use handle directly Remove global handle variable Always pass data to trans_commit() lib/libalpm/add.c | 173 +++++++++++++++--------------- lib/libalpm/alpm.c | 14 +-- lib/libalpm/alpm.h | 159 +++++++++++++++------------- lib/libalpm/be_local.c | 77 ++++++------- lib/libalpm/be_package.c | 92 ++++++++-------- lib/libalpm/be_sync.c | 88 ++++++++-------- lib/libalpm/conflict.c | 151 ++++++++++++++------------ lib/libalpm/conflict.h | 3 +- lib/libalpm/db.c | 124 ++++++++++----------- lib/libalpm/delta.c | 31 +++--- lib/libalpm/delta.h | 2 +- lib/libalpm/deps.c | 117 +++++++++----------- lib/libalpm/deps.h | 4 +- lib/libalpm/diskspace.c | 38 ++++---- lib/libalpm/dload.c | 61 +++++------ lib/libalpm/dload.h | 3 +- lib/libalpm/error.c | 16 +--- lib/libalpm/graph.c | 2 +- lib/libalpm/group.c | 9 +- lib/libalpm/handle.c | 269 ++++++++++++++++++---------------------------- lib/libalpm/handle.h | 3 + lib/libalpm/log.c | 23 ++--- lib/libalpm/log.h | 3 +- lib/libalpm/package.c | 38 ++++--- lib/libalpm/package.h | 8 +- lib/libalpm/pkghash.c | 13 +-- lib/libalpm/remove.c | 118 +++++++++++---------- lib/libalpm/signing.c | 83 ++++++++------- lib/libalpm/signing.h | 3 +- lib/libalpm/sync.c | 192 +++++++++++++++++---------------- lib/libalpm/trans.c | 115 +++++++++----------- lib/libalpm/trans.h | 1 - lib/libalpm/util.c | 127 +++++++++++----------- lib/libalpm/util.h | 23 +++-- src/pacman/callback.c | 6 +- src/pacman/conf.c | 54 +++++----- src/pacman/database.c | 4 +- src/pacman/deptest.c | 3 +- src/pacman/package.c | 5 +- src/pacman/pacman.c | 18 ++-- src/pacman/query.c | 21 ++-- src/pacman/remove.c | 27 +++-- src/pacman/sync.c | 78 +++++++------ src/pacman/upgrade.c | 34 +++--- src/pacman/util.c | 22 ++-- src/util/cleanupdelta.c | 6 +- src/util/pactree.c | 34 ++---- src/util/testdb.c | 19 ++-- src/util/testpkg.c | 9 +- 49 files changed, 1227 insertions(+), 1296 deletions(-) -- 1.7.5.2

4 20

[pacman-dev] [PATCH 1/2] _alpm_lstat: only duplicate string if necessary
by Dan McGee 09 Jun '11

09 Jun '11

The vast majority of the time we will just be passing the same string value on to the lstat() call. The only time we need to duplicate it is if the path ends in '/'. In one run using a profiler, only 400 of the 200,000 calls (0.2%) required the string to be copied first. Signed-off-by: Dan McGee <dan(a)archlinux.org> --- lib/libalpm/util.c | 15 ++++++++------- 1 files changed, 8 insertions(+), 7 deletions(-) diff --git a/lib/libalpm/util.c b/lib/libalpm/util.c index 9a8f06d..357ce50 100644 --- a/lib/libalpm/util.c +++ b/lib/libalpm/util.c @@ -636,17 +636,18 @@ const char *_alpm_filecache_setup(pmhandle_t *handle) int _alpm_lstat(const char *path, struct stat *buf) { int ret; - char *newpath = strdup(path); - size_t len = strlen(newpath); + size_t len = strlen(path); /* strip the trailing slash if one exists */ - if(len != 0 && newpath[len - 1] == '/') { - newpath[len - 1] = '\0'; + if(len != 0 && path[len - 1] == '/') { + char *newpath = strdup(path); + newpath[len - 1] = '\0'; + ret = lstat(newpath, buf); + free(newpath); + } else { + ret = lstat(path, buf); } - ret = lstat(newpath, buf); - - FREE(newpath); return ret; } -- 1.7.5.2

1 1

[pacman-dev] [PATCH 1/9] conf: _parseconfig() cleanups and documentation
by Dan McGee 09 Jun '11

09 Jun '11

* Function doxygen documentation * Reuse a single strlen() call * Prevent infinite recursion (limit to 10 levels) * Other small cleanups Signed-off-by: Dan McGee <dan(a)archlinux.org> --- src/pacman/conf.c | 59 +++++++++++++++++++++++++++++++++++------------------ 1 files changed, 39 insertions(+), 20 deletions(-) diff --git a/src/pacman/conf.c b/src/pacman/conf.c index 97e6497..68fe59c 100644 --- a/src/pacman/conf.c +++ b/src/pacman/conf.c @@ -508,32 +508,52 @@ static int setup_libalpm(void) } -/* The real parseconfig. Called with a null section argument by the publicly - * visible parseconfig so we can recall from within ourself on an include */ -static int _parseconfig(const char *file, int parse_options, - char **section, pmdb_t *db) +/** The "real" parseconfig. Each "Include" directive will recall this method so + * recursion and stack depth are limited to 10 levels. The publicly visible + * parseconfig calls this with a NULL section argument so we can recall from + * within ourself on an include. + * @param file path to the config file + * @param section the current active section name; should be freed after all + * parsing is complete + * @param db the current active alpm database object + * @param parse_options whether to parse and call methods for the options + * section; if 0, parse and call methods for the repos sections + * @param depth the current recursion depth + **/ +static int _parseconfig(const char *file, char **section, pmdb_t *db, + int parse_options, int depth) { FILE *fp = NULL; - char line[PATH_MAX+1]; + char line[PATH_MAX]; int linenum = 0; - char *ptr; int ret = 0; + const int max_depth = 10; + + if(depth >= max_depth) { + pm_printf(PM_LOG_ERROR, + _("config parsing exceeded max recursion depth of %d.\n"), max_depth); + ret = 1; + goto cleanup; + } pm_printf(PM_LOG_DEBUG, "config: attempting to read file %s\n", file); fp = fopen(file, "r"); if(fp == NULL) { pm_printf(PM_LOG_ERROR, _("config file %s could not be read.\n"), file); - return 1; + ret = 1; + goto cleanup; } while(fgets(line, PATH_MAX, fp)) { - char *key, *value; + char *key, *value, *ptr; + size_t line_len; linenum++; strtrim(line); + line_len = strlen(line); /* ignore whole line and end of line comments */ - if(strlen(line) == 0 || line[0] == '#') { + if(line_len == 0 || line[0] == '#') { continue; } if((ptr = strchr(line, '#'))) { @@ -548,19 +568,18 @@ static int _parseconfig(const char *file, int parse_options, goto cleanup; } - if(line[0] == '[' && line[strlen(line)-1] == ']') { + if(line[0] == '[' && line[line_len - 1] == ']') { char *name; - /* new config section, skip the '[' */ - ptr = line; - ptr++; - name = strdup(ptr); - name[strlen(name)-1] = '\0'; - if(!strlen(name)) { + /* only possibility here is a line == '[]' */ + if(line_len <= 2) { pm_printf(PM_LOG_ERROR, _("config file %s, line %d: bad section name.\n"), file, linenum); ret = 1; goto cleanup; } + /* new config section, skip the '[' */ + name = strdup(line + 1); + name[line_len - 2] = '\0'; pm_printf(PM_LOG_DEBUG, "config: new section '%s'\n", name); /* if we are not looking at the options section, register a db */ if(!parse_options && strcmp(name, "options") != 0) { @@ -634,7 +653,7 @@ static int _parseconfig(const char *file, int parse_options, for(gindex = 0; gindex < globbuf.gl_pathc; gindex++) { pm_printf(PM_LOG_DEBUG, "config file %s, line %d: including %s\n", file, linenum, globbuf.gl_pathv[gindex]); - _parseconfig(globbuf.gl_pathv[gindex], parse_options, section, db); + _parseconfig(globbuf.gl_pathv[gindex], section, db, parse_options, depth++); } break; } @@ -690,7 +709,7 @@ cleanup: } /** Parse a configuration file. - * @param file path to the config file. + * @param file path to the config file * @return 0 on success, non-zero on error */ int parseconfig(const char *file) @@ -703,7 +722,7 @@ int parseconfig(const char *file) /* call the real parseconfig function with a null section & db argument */ pm_printf(PM_LOG_DEBUG, "parseconfig: options pass\n"); - if((ret = _parseconfig(file, 1, &section, NULL))) { + if((ret = _parseconfig(file, &section, NULL, 1, 0))) { free(section); return ret; } @@ -714,7 +733,7 @@ int parseconfig(const char *file) /* second pass, repo section parsing */ section = NULL; pm_printf(PM_LOG_DEBUG, "parseconfig: repo pass\n"); - return _parseconfig(file, 0, &section, NULL); + return _parseconfig(file, &section, NULL, 0, 0); free(section); } -- 1.7.5.2

3 15

[pacman-dev] [PATCH 0/6] Package signing patches
by Pang Yan Han 09 Jun '11

09 Jun '11

This is a series of patches which can hopefully take package signing a small step forward. Patch 1/6 lets pacman -v print the GPG Dir Patch 2/6 updates the pacman manpage's --gpgdir section and adds a commented GPGDir option in pacman.conf which shows the user the default gpgdir. Patch 3/6 updates Makefile.am to create the default gpgdir, pubring.gpg and trustdb.gpg if they do not exist. Patch 4/6 adds a --verify option to pacman which lets the user control the level of signature verification. Valid arguments to --verify include "always", "optional", "never". Think of this as a way to override the "VerifySig" option in pacman.conf . In addition Patch 4/6 updates configure.ac to output gpgdir at the end of ./configure. Patch 5/6 updates the help text of pacman-key with regards to --gpgdir. Patch 6/6 adds a --import-trustdb option to pacman-key. This allows the user to import a selected trustdb upon confirmation. For Patch 6/6, Bash is not really my strong point and I took quite some time to get it right. Please help me check it. Thanks! Pang Yan Han (6): Let pacman -v print GPG Dir Update pacman manpage and pacman.conf for gpgdir Update Makefile.am and configure.ac for gpgdir Add --verify option for signature level pacman-key: print default gpgdir in usage pacman-key: add --import-trustdb option Makefile.am | 7 +++- configure.ac | 1 + doc/pacman-key.8.txt | 4 ++ doc/pacman.8.txt | 16 ++++++--- etc/pacman.conf.in | 1 + scripts/pacman-key.sh.in | 83 +++++++++++++++++++++++++++++++++++++++++++++- src/pacman/conf.h | 3 +- src/pacman/pacman.c | 16 +++++++++ 8 files changed, 122 insertions(+), 9 deletions(-) -- 1.7.5.rc0.101.g3d23c

6 24

[pacman-dev] [PATCH] makepkg: move option parsing code to separte file
by Allan McRae 09 Jun '11

09 Jun '11

This move the getopt replacement function parse_options out of makepkg.sh.in and into a separate file. The code is inserted into the relevant place in makepkg using m4. This will allow the reuse of the option parsing code in other scripts (i.e. pacman-key) while avoiding code duplication. Signed-off-by: Allan McRae <allan(a)archlinux.org> --- Is there a better way to do this that I am missing? scripts/Makefile.am | 10 +++-- scripts/library/parse_options.sh | 87 +++++++++++++++++++++++++++++++++++++ scripts/makepkg.sh.in | 88 +------------------------------------- 3 files changed, 94 insertions(+), 91 deletions(-) create mode 100644 scripts/library/parse_options.sh diff --git a/scripts/Makefile.am b/scripts/Makefile.am index fd9b20d..6335ba1 100644 --- a/scripts/Makefile.am +++ b/scripts/Makefile.am @@ -56,10 +56,12 @@ edit = sed \ # third 'test' line- make sure one of the two checks succeeded $(OURSCRIPTS): Makefile @echo ' ' GEN $@; - @rm -f $@ $@.tmp - @test -f $(srcdir)/$@.sh.in && $(edit) $(srcdir)/$@.sh.in >$@.tmp || true - @test -f $(srcdir)/$@.py.in && $(edit) $(srcdir)/$@.py.in >$@.tmp || true - @test -f $@.tmp || false + @rm -f $@ $@.m4 $@.tmp + @test -f $(srcdir)/$@.sh.in && $(edit) $(srcdir)/$@.sh.in >$@.m4 || true + @test -f $(srcdir)/$@.py.in && $(edit) $(srcdir)/$@.py.in >$@.m4 || true + @test -f $@.m4 || false + @m4 -P $@.m4 >$@.tmp + @rm -f $@.m4 @chmod +x $@.tmp @chmod a-w $@.tmp @mv $@.tmp $@ diff --git a/scripts/library/parse_options.sh b/scripts/library/parse_options.sh new file mode 100644 index 0000000..878cff1 --- /dev/null +++ b/scripts/library/parse_options.sh @@ -0,0 +1,87 @@ +# getopt like parser +parse_options() { + local short_options=$1; shift; + local long_options=$1; shift; + local ret=0; + local unused_options="" + local i + + while [[ -n $1 ]]; do + if [[ ${1:0:2} = '--' ]]; then + if [[ -n ${1:2} ]]; then + local match="" + for i in ${long_options//,/ }; do + if [[ ${1:2} = ${i//:} ]]; then + match=$i + break + fi + done + if [[ -n $match ]]; then + if [[ ${1:2} = $match ]]; then + printf ' %s' "$1" + else + if [[ -n $2 ]]; then + printf ' %s' "$1" + shift + printf " '%s'" "$1" + else + echo "makepkg: option '$1' $(gettext "requires an argument")" >&2 + ret=1 + fi + fi + else + echo "makepkg: $(gettext "unrecognized option") '$1'" >&2 + ret=1 + fi + else + shift + break + fi + elif [[ ${1:0:1} = '-' ]]; then + for ((i=1; i<${#1}; i++)); do + if [[ $short_options =~ ${1:i:1} ]]; then + if [[ $short_options =~ ${1:i:1}: ]]; then + if [[ -n ${1:$i+1} ]]; then + printf ' -%s' "${1:i:1}" + printf " '%s'" "${1:$i+1}" + else + if [[ -n $2 ]]; then + printf ' -%s' "${1:i:1}" + shift + printf " '%s'" "${1}" + else + echo "makepkg: option $(gettext "requires an argument") -- '${1:i:1}'" >&2 + ret=1 + fi + fi + break + else + printf ' -%s' "${1:i:1}" + fi + else + echo "makepkg: $(gettext "invalid option") -- '${1:i:1}'" >&2 + ret=1 + fi + done + else + unused_options="${unused_options} '$1'" + fi + shift + done + + printf " --" + if [[ -n $unused_options ]]; then + for i in ${unused_options[@]}; do + printf ' %s' "$i" + done + fi + if [[ -n $1 ]]; then + while [[ -n $1 ]]; do + printf " '%s'" "${1}" + shift + done + fi + printf "\n" + + return $ret +} \ No newline at end of file diff --git a/scripts/makepkg.sh.in b/scripts/makepkg.sh.in index b3081fc..0639fbd 100644 --- a/scripts/makepkg.sh.in +++ b/scripts/makepkg.sh.in @@ -1506,93 +1506,7 @@ canonicalize_path() { fi } -# getopt like parser -parse_options() { - local short_options=$1; shift; - local long_options=$1; shift; - local ret=0; - local unused_options="" - local i - - while [[ -n $1 ]]; do - if [[ ${1:0:2} = '--' ]]; then - if [[ -n ${1:2} ]]; then - local match="" - for i in ${long_options//,/ }; do - if [[ ${1:2} = ${i//:} ]]; then - match=$i - break - fi - done - if [[ -n $match ]]; then - if [[ ${1:2} = $match ]]; then - printf ' %s' "$1" - else - if [[ -n $2 ]]; then - printf ' %s' "$1" - shift - printf " '%s'" "$1" - else - echo "makepkg: option '$1' $(gettext "requires an argument")" >&2 - ret=1 - fi - fi - else - echo "makepkg: $(gettext "unrecognized option") '$1'" >&2 - ret=1 - fi - else - shift - break - fi - elif [[ ${1:0:1} = '-' ]]; then - for ((i=1; i<${#1}; i++)); do - if [[ $short_options =~ ${1:i:1} ]]; then - if [[ $short_options =~ ${1:i:1}: ]]; then - if [[ -n ${1:$i+1} ]]; then - printf ' -%s' "${1:i:1}" - printf " '%s'" "${1:$i+1}" - else - if [[ -n $2 ]]; then - printf ' -%s' "${1:i:1}" - shift - printf " '%s'" "${1}" - else - echo "makepkg: option $(gettext "requires an argument") -- '${1:i:1}'" >&2 - ret=1 - fi - fi - break - else - printf ' -%s' "${1:i:1}" - fi - else - echo "makepkg: $(gettext "invalid option") -- '${1:i:1}'" >&2 - ret=1 - fi - done - else - unused_options="${unused_options} '$1'" - fi - shift - done - - printf " --" - if [[ -n $unused_options ]]; then - for i in ${unused_options[@]}; do - printf ' %s' "$i" - done - fi - if [[ -n $1 ]]; then - while [[ -n $1 ]]; do - printf " '%s'" "${1}" - shift - done - fi - printf "\n" - - return $ret -} +m4_include(library/parse_options.sh) usage() { printf "makepkg (pacman) %s\n" "$myver" -- 1.7.5.4

3 3

[pacman-dev] Should we bump major version number too?
by Rémy Oudompheng 08 Jun '11

08 Jun '11

Hi all, It's in the mood: GTK, Gnome, Linux are bumping their major version numbers. Should we follow the same path? Several reasons why I think this is a good idea: - we have a major feature coming: signature checks - we have a major API breakage: the pmhandle_t thing, the signature check level thing - throughout the 3.x development cycle, a lot of incompatibilities/changes have happened: * database have no longer the separate "depends" file * sync DBs are now a single tarball * packages have an "epoch" field not understood by previous versions * packages can be xz-compressed * PKGBUILDs have epoch, check() and checkdepends * force is not accepted anymore by makepkg If the next pacman version were to be 3.6, it would have very little similarities with pacman 3.0. They don't seem to be forward-compatible, neither are they backward-compatible. So maybe it's time to go pacman 4.0 ? Cheers, Rémy.

2 1