- Pacman-dev - lists.archlinux.org

[pacman-dev] [GIT] The official pacman repository branch, maint, updated. v4.0.0-19-gd6e3446
by dan＠archlinux.org 22 Oct '11

22 Oct '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, maint has been updated via d6e3446e70bcda9c0f28d80110d588106daa4b5c (commit) via 20a47aba8ec859e920bf965ab7da74c26a44339d (commit) via 9e9ecf218344cea86d6b8c6cf7cd246bcfcca98b (commit) via 45f86ca1ca712857e6f84288cf818ce5707fc345 (commit) via 4bada454643e6d91f0b53415bc8d135bc7f578b5 (commit) via bac670ddc9a9dcad2f0665f9f77d04b6736c6db4 (commit) via 842c4422eda231892d4bce75d8e315eb652533b6 (commit) from 020bdb4298cd1bc53df7ca4d911cda7aaa65329c (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 d6e3446e70bcda9c0f28d80110d588106daa4b5c Author: lolilolicon <lolilolicon(a)gmail.com> Date: Thu Oct 20 11:35:00 2011 +0800 bacman: pkgrel does not have to be an integer pkgrel, as with pkgver, simply mustn't contain hyphens. Signed-off-by: lolilolicon <lolilolicon(a)gmail.com> Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 20a47aba8ec859e920bf965ab7da74c26a44339d Author: Dan McGee <dan(a)archlinux.org> Date: Fri Oct 21 10:51:49 2011 -0500 Remove remaining usages of fprintf() from frontend These can either be replaced with pm_printf() if they are error related, or in the fprintf(stdout, ...) case a bare printf() will do. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 9e9ecf218344cea86d6b8c6cf7cd246bcfcca98b Author: Dan McGee <dan(a)archlinux.org> Date: Fri Oct 21 10:44:19 2011 -0500 Remove pm_fprintf() in favor of pm_printf() Now that pm_printf() always prints to stderr, we don't need this second function that was always used with stderr as the first argument. Thus, this patch removes the function and makes the following sed replacement: sed -i -e 's#pm_fprintf(stderr, #pm_printf(#g' src/pacman/*.c Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 45f86ca1ca712857e6f84288cf818ce5707fc345 Author: Dan McGee <dan(a)archlinux.org> Date: Fri Oct 21 10:38:02 2011 -0500 Use stderr as output stream for pm_printf() This matches what we now do in our backend callback function- all debug/info/warning/error/etc. messages should be on stderr. These are all the messages with a "warning:" or other type prefix, so does not affect general pacman output. This should fix the output confusion noted in FS#26555. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 4bada454643e6d91f0b53415bc8d135bc7f578b5 Author: Dan McGee <dan(a)archlinux.org> Date: Mon Oct 17 09:42:46 2011 -0500 pacman-key: add a default keyserver timeout value on --init The default is supposidely 30 seconds from the gpg manpage, but that sure wasn't what I was seeing- it was somewhere closer to two minutes of silence. Add a more reasonable 10 second timeout value which should be good enough for any keyserver that doesn't totally stink at it's job. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit bac670ddc9a9dcad2f0665f9f77d04b6736c6db4 Author: Dan McGee <dan(a)archlinux.org> Date: Mon Oct 17 09:38:35 2011 -0500 Show an error message on failed remote key lookup The absolutely terrible part about this is the failure on GPGME's part to distinguish between "key not found" and "keyserver timeout". Instead, it returns the same silly GPG_ERR_EOF in both cases (why isn't GPG_ERR_TIMEOUT being used?), leaving us helpless to tell them apart. Spit out a generic enough error message that covers both cases; unfortunately we can't provide much guidance to the user because we aren't sure what actually happened. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 842c4422eda231892d4bce75d8e315eb652533b6 Author: Dan McGee <dan(a)archlinux.org> Date: Mon Oct 17 08:28:57 2011 -0500 Table display: print message with warning: prefix Use the normal error functions here rather than a bare fprintf(). Signed-off-by: Dan McGee <dan(a)archlinux.org> ----------------------------------------------------------------------- Summary of changes: contrib/bacman.in | 2 +- lib/libalpm/signing.c | 4 ++-- scripts/pacman-key.sh.in | 1 + src/pacman/conf.c | 2 +- src/pacman/package.c | 8 ++++---- src/pacman/query.c | 28 ++++++++++++++-------------- src/pacman/remove.c | 10 +++++----- src/pacman/sync.c | 32 ++++++++++++++++---------------- src/pacman/upgrade.c | 6 +++--- src/pacman/util.c | 31 ++++++++++--------------------- src/pacman/util.h | 2 +- 11 files changed, 58 insertions(+), 68 deletions(-) hooks/post-receive -- The official pacman repository

1 0

[pacman-dev] [PATCH] Hide empty columns in table package list display
by Dan McGee 21 Oct '11

21 Oct '11

On -R operations, the "New Version" column is always empty, taking up space and not really showing the user anything valuable. The same is true on -S or -U operations for the "Old Version" column when packages are only being installed and not upgraded. Remove this column so we get a few screen columns back, especially now that we show repo/packagename style output. This also makes some adjustment to the padding logic. We no longer include padding in column widths but it is included in the total table width. We also ensure the last displayed column is always right aligned, even if this is not the actual rightmost column. Example output, before: $ sudo pacman -R eclipse checking dependencies... Targets (1): Name Old Version New Version Net Change eclipse 3.7-1 -194.02 MiB Total Removed Size: 194.02 MiB And after: $ sudo pacman -R eclipse checking dependencies... Targets (1): Name Old Version Net Change eclipse 3.7-1 -194.02 MiB Total Removed Size: 194.02 MiB Signed-off-by: Dan McGee <dan(a)archlinux.org> --- src/pacman/util.c | 50 +++++++++++++++++++++++++++++++++++++++----------- 1 files changed, 39 insertions(+), 11 deletions(-) diff --git a/src/pacman/util.c b/src/pacman/util.c index 77215de..06f8c3b 100644 --- a/src/pacman/util.c +++ b/src/pacman/util.c @@ -488,17 +488,23 @@ void string_display(const char *title, const char *string) } static void table_print_line(const alpm_list_t *line, - size_t colcount, size_t *widths) + size_t colcount, size_t *widths, int *has_data) { size_t i; const alpm_list_t *curcell; for(i = 0, curcell = line; curcell && i < colcount; i++, curcell = alpm_list_next(curcell)) { - const char *value = curcell->data; - size_t len = string_length(value); + const char *value; + int padding; + + if(!has_data[i]) { + continue; + } + + value = curcell->data; /* silly printf requires padding size to be an int */ - int padding = (int)widths[i] - (int)len; + padding = (int)widths[i] - (int)string_length(value); if(padding < 0) { padding = 0; } @@ -513,26 +519,42 @@ static void table_print_line(const alpm_list_t *line, printf("\n"); } -/* find the max string width of each column */ + + +/** + * Find the max string width of each column. Also determines whether values + * exist in the column and sets the value in has_data accordingly. + * @param header a list of header strings + * @param rows a list of lists of rows as strings + * @param totalcols the total number of columns in the header and each row + * @param widths a pointer to store width data + * @param has_data a pointer to store whether column has data + * + * @return the total width of the table; 0 on failure + */ static size_t table_calc_widths(const alpm_list_t *header, - const alpm_list_t *rows, size_t totalcols, size_t **widths) + const alpm_list_t *rows, size_t totalcols, size_t **widths, + int **has_data) { const alpm_list_t *i; const unsigned short padding = 2; size_t curcol, totalwidth = 0; size_t *colwidths; + int *coldata; if(totalcols <= 0) { return 0; } colwidths = malloc(totalcols * sizeof(size_t)); - if(!colwidths) { + coldata = calloc(totalcols, sizeof(int)); + if(!colwidths || !coldata) { return 0; } /* header determines column count and initial values of longest_strs */ for(i = header, curcol = 0; i; i = alpm_list_next(i), curcol++) { colwidths[curcol] = string_length(i->data); + /* note: header does not determine whether column has data */ } /* now find the longest string in each column */ @@ -546,6 +568,9 @@ static size_t table_calc_widths(const alpm_list_t *header, if(str_len > colwidths[curcol]) { colwidths[curcol] = str_len; } + if(str_len > 0) { + coldata[curcol] = 1; + } } } @@ -558,6 +583,7 @@ static size_t table_calc_widths(const alpm_list_t *header, } *widths = colwidths; + *has_data = coldata; return totalwidth; } @@ -576,20 +602,21 @@ int table_display(const char *title, const alpm_list_t *header, { const alpm_list_t *i; size_t *widths = NULL, totalcols, totalwidth; + int *has_data = NULL; if(rows == NULL || header == NULL) { return 0; } totalcols = alpm_list_count(header); - totalwidth = table_calc_widths(header, rows, totalcols, &widths); + totalwidth = table_calc_widths(header, rows, totalcols, &widths, &has_data); /* return -1 if terminal is not wide enough */ if(totalwidth > getcols()) { pm_printf(ALPM_LOG_WARNING, _("insufficient columns available for table display\n")); return -1; } - if(!totalwidth || !widths) { + if(!totalwidth || !widths || !has_data) { return -1; } @@ -597,14 +624,15 @@ int table_display(const char *title, const alpm_list_t *header, printf("%s\n\n", title); } - table_print_line(header, totalcols, widths); + table_print_line(header, totalcols, widths, has_data); printf("\n"); for(i = rows; i; i = alpm_list_next(i)) { - table_print_line(i->data, totalcols, widths); + table_print_line(i->data, totalcols, widths, has_data); } free(widths); + free(has_data); return 0; } -- 1.7.7

1 1

[pacman-dev] [PATCH] sync: extract build_payload() method from find_dl_candidates
by Dan McGee 21 Oct '11

21 Oct '11

This is done by both the delta and regular file code, so we can extract a little helper method. Done mostly to satisfy my "why are we repeating code here" itch. Signed-off-by: Dan McGee <dan(a)archlinux.org> --- lib/libalpm/sync.c | 32 ++++++++++++++++++-------------- 1 files changed, 18 insertions(+), 14 deletions(-) diff --git a/lib/libalpm/sync.c b/lib/libalpm/sync.c index a5e8a5e..04c557b 100644 --- a/lib/libalpm/sync.c +++ b/lib/libalpm/sync.c @@ -790,6 +790,18 @@ static int validate_deltas(alpm_handle_t *handle, alpm_list_t *deltas) return 0; } +static struct dload_payload *build_payload(alpm_handle_t *handle, + const char *filename, size_t size, alpm_list_t *servers) +{ + struct dload_payload *payload; + + CALLOC(payload, 1, sizeof(*payload), RET_ERR(handle, ALPM_ERR_MEMORY, NULL)); + STRDUP(payload->remote_name, filename, RET_ERR(handle, ALPM_ERR_MEMORY, NULL)); + payload->max_size = size; + payload->servers = servers; + return payload; +} + static int find_dl_candidates(alpm_db_t *repo, alpm_list_t **files, alpm_list_t **deltas) { alpm_list_t *i; @@ -814,14 +826,10 @@ static int find_dl_candidates(alpm_db_t *repo, alpm_list_t **files, alpm_list_t for(dlts = delta_path; dlts; dlts = dlts->next) { alpm_delta_t *delta = dlts->data; if(delta->download_size != 0) { - struct dload_payload *dpayload; - - CALLOC(dpayload, 1, sizeof(*dpayload), RET_ERR(handle, ALPM_ERR_MEMORY, -1)); - STRDUP(dpayload->remote_name, delta->delta, RET_ERR(handle, ALPM_ERR_MEMORY, -1)); - dpayload->max_size = delta->download_size; - dpayload->servers = repo->servers; - - *files = alpm_list_add(*files, dpayload); + struct dload_payload *payload = build_payload( + handle, delta->delta, delta->download_size, repo->servers); + ASSERT(payload, return -1); + *files = alpm_list_add(*files, payload); } /* keep a list of all the delta files for md5sums */ *deltas = alpm_list_add(*deltas, delta); @@ -829,13 +837,9 @@ static int find_dl_candidates(alpm_db_t *repo, alpm_list_t **files, alpm_list_t } else if(spkg->download_size != 0) { struct dload_payload *payload; - ASSERT(spkg->filename != NULL, RET_ERR(handle, ALPM_ERR_PKG_INVALID_NAME, -1)); - CALLOC(payload, 1, sizeof(*payload), RET_ERR(handle, ALPM_ERR_MEMORY, -1)); - STRDUP(payload->remote_name, spkg->filename, RET_ERR(handle, ALPM_ERR_MEMORY, -1)); - payload->max_size = spkg->size; - payload->servers = repo->servers; - + payload = build_payload(handle, spkg->filename, spkg->size, repo->servers); + ASSERT(payload, return -1); *files = alpm_list_add(*files, payload); } } -- 1.7.7

1 0

[pacman-dev] [PATCH 1/3] Use stderr as output stream for pm_printf()
by Dan McGee 21 Oct '11

21 Oct '11

This matches what we now do in our backend callback function- all debug/info/warning/error/etc. messages should be on stderr. These are all the messages with a "warning:" or other type prefix, so does not affect general pacman output. This should fix the output confusion noted in FS#26555. Signed-off-by: Dan McGee <dan(a)archlinux.org> --- This and the following two patches are for maint. src/pacman/util.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/pacman/util.c b/src/pacman/util.c index 60e2914..1ca9b4d 100644 --- a/src/pacman/util.c +++ b/src/pacman/util.c @@ -1417,7 +1417,7 @@ int pm_printf(alpm_loglevel_t level, const char *format, ...) /* print the message using va_arg list */ va_start(args, format); - ret = pm_vfprintf(stdout, level, format, args); + ret = pm_vfprintf(stderr, level, format, args); va_end(args); return ret; -- 1.7.7

1 2

[pacman-dev] Feature request: Show exact matches first
by Markus Jochim 20 Oct '11

20 Oct '11

Dear developers, I'd like to request a feature. When I search for some package, let's say openconnect: $ pacman -Ss openconnect (or -Qs) it takes quite a while until I get any results on my netbook (Atom N450 cpu). Maybe this could be sped up by showing exact matches right away. I've had this situation quite some times now and it can be a real pain in the ass to wait for 10 seconds (this may vary of course) when I simply want to know whether I have a particular package installed. Kindly, Markus

6 7

[pacman-dev] [PATCH] bacman: pkgrel does not have to be an integer
by lolilolicon 20 Oct '11

20 Oct '11

pkgrel, as with pkgver, simply mustn't contain hyphens. Signed-off-by: lolilolicon <lolilolicon(a)gmail.com> --- contrib/bacman.in | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/contrib/bacman.in b/contrib/bacman.in index 9636091..c55d716 100755 --- a/contrib/bacman.in +++ b/contrib/bacman.in @@ -93,7 +93,7 @@ pkg_dest="${PKGDEST:-$PWD}" pkg_pkger=${PACKAGER:-'Unknown Packager'} pkg_name="$1" -pkg_dir=("$pac_db/$pkg_name"-+([^-])-+([0-9])) +pkg_dir=("$pac_db/$pkg_name"-+([^-])-+([^-])) pkg_namver=("${pkg_dir[@]##*/}") # -- 1.7.6.4

1 0

[pacman-dev] [PATCH] add gpg key length to import message
by Florian Pritz 18 Oct '11

18 Oct '11

Displaying the length of the key we are about to import allows the user to spot forged keys more easily. References: http://www.pgp.net/pgpnet/pgp-faq/pgp-faq-keys.html#key-public-key-forgery Signed-off-by: Florian Pritz <bluewind(a)xinu.at> --- lib/libalpm/alpm.h | 1 + lib/libalpm/signing.c | 1 + src/pacman/callback.c | 4 ++-- 3 files changed, 4 insertions(+), 2 deletions(-) diff --git a/lib/libalpm/alpm.h b/lib/libalpm/alpm.h index 9fda940..2e4de79 100644 --- a/lib/libalpm/alpm.h +++ b/lib/libalpm/alpm.h @@ -226,6 +226,7 @@ typedef struct _alpm_pgpkey_t { char *email; alpm_time_t created; alpm_time_t expires; + unsigned int length; } alpm_pgpkey_t; /** Signature result. Contains the key, status, and validity of a given diff --git a/lib/libalpm/signing.c b/lib/libalpm/signing.c index bdaa83a..3d965b4 100644 --- a/lib/libalpm/signing.c +++ b/lib/libalpm/signing.c @@ -270,6 +270,7 @@ static int key_search(alpm_handle_t *handle, const char *fpr, pgpkey->email = key->uids->email; pgpkey->created = key->subkeys->timestamp; pgpkey->expires = key->subkeys->expires; + pgpkey->length = key->subkeys->length; ret = 1; error: diff --git a/src/pacman/callback.c b/src/pacman/callback.c index d856455..cb2b080 100644 --- a/src/pacman/callback.c +++ b/src/pacman/callback.c @@ -349,8 +349,8 @@ void cb_question(alpm_question_t event, void *data1, void *data2, char created[12]; time_t time = (time_t)key->created; strftime(created, 12, "%Y-%m-%d", localtime(&time)); - *response = yesno(_(":: Import PGP key %s, \"%s\", created %s?"), - key->fingerprint, key->uid, created); + *response = yesno(_(":: Import PGP key %s, \"%s\", created %s, length %d bytes?"), + key->fingerprint, key->uid, created, key->length); } break; } -- 1.7.7

2 1

[pacman-dev] [PATCH 1/2] Show an error message on failed remote key lookup
by Dan McGee 17 Oct '11

17 Oct '11

The absolutely terrible part about this is the failure on GPGME's part to distinguish between "key not found" and "keyserver timeout". Instead, it returns the same silly GPG_ERR_EOF in both cases (why isn't GPG_ERR_TIMEOUT being used?), leaving us helpless to tell them apart. Spit out a generic enough error message that covers both cases; unfortunately we can't provide much guidance to the user because we aren't sure what actually happened. Signed-off-by: Dan McGee <dan(a)archlinux.org> --- lib/libalpm/signing.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/lib/libalpm/signing.c b/lib/libalpm/signing.c index bdaa83a..92f34b5 100644 --- a/lib/libalpm/signing.c +++ b/lib/libalpm/signing.c @@ -797,8 +797,8 @@ int _alpm_process_siglist(alpm_handle_t *handle, const char *identifier, fetch_key.fingerprint, fetch_key.uid); } } else { - _alpm_log(handle, ALPM_LOG_DEBUG, - "key could not be looked up remotely\n"); + _alpm_log(handle, ALPM_LOG_ERROR, + _("key \"%s\" could not be looked up remotely\n"), name); } gpgme_key_unref(fetch_key.data); } -- 1.7.7

1 1

[pacman-dev] [PATCH 0/5] download diskspace checking
by Dave Reisner 16 Oct '11

16 Oct '11

I don't think this patchset is ready for merging, but I figured I'd post it here for some possible feedback. This is mostly for the purposes of adding diskspace checking prior to downloads, but it also changes some ancient behavior that Judd added of grouping sync records by repo. Instead, we just gather all the records together at once and download them. I've left some comments in the header for each patch... d Dave Reisner (5): diskspace: add _alpm_check_downloadspace() dload: add RO copy of servers list to each payload sync: dont group sync records by repository sync: check for necessary disk space for download sync: move file download loop out of download_files lib/libalpm/diskspace.c | 54 ++++++++++++++ lib/libalpm/diskspace.h | 2 + lib/libalpm/dload.h | 1 + lib/libalpm/sync.c | 184 ++++++++++++++++++++++++++++------------------ src/pacman/callback.c | 2 +- 5 files changed, 170 insertions(+), 73 deletions(-) -- 1.7.7

2 11

[pacman-dev] [PATCH 1/2] Make _alpm_filelist_contains() NULL-safe
by Dan McGee 14 Oct '11

14 Oct '11

Signed-off-by: Dan McGee <dan(a)archlinux.org> --- lib/libalpm/conflict.c | 10 +++++++--- 1 files changed, 7 insertions(+), 3 deletions(-) diff --git a/lib/libalpm/conflict.c b/lib/libalpm/conflict.c index 14c23f4..f686ca8 100644 --- a/lib/libalpm/conflict.c +++ b/lib/libalpm/conflict.c @@ -318,12 +318,16 @@ const alpm_file_t *_alpm_filelist_contains(alpm_filelist_t *filelist, const char *name) { size_t i; - const alpm_file_t *file = filelist->files; - for(i = 0; i < filelist->count; i++) { + const alpm_file_t *file; + + if(!filelist) { + return NULL; + } + + for(file = filelist->files, i = 0; i < filelist->count; file++, i++) { if(strcmp(file->name, name) == 0) { return file; } - file++; } return NULL; } -- 1.7.7

1 1