This is particularly important in the case of FTP control connections,
which may be closed by rogue NAT/firewall devices detecting idle
connections on larger transfers which may take 5-10+ minutes.
Signed-off-by: Dave Reisner <dreisner@archlinux.org>
Signed-off-by: Dan McGee <dan@archlinux.org>
With pacman-4.0, using --noconfirm or --noprogressbar with -Q or -T
results in pacman reporting an "invalid option" error. Restrict the
passing of these options to pacman. Fixes FS#28012.
Signed-off-by: Allan McRae <allan@archlinux.org>
Signed-off-by: Dan McGee <dan@archlinux.org>
The grep statement used to check for a difference between the
installed package list before and after resolving dependencies
returns 1 if there is no difference. This sets of the error
trap when "-r" is used "unnecessarily".
Signed-off-by: Allan McRae <allan@archlinux.org>
Signed-off-by: Dan McGee <dan@archlinux.org>
If you are a crazy developer like me and have bogus options in your
pacman.conf file, the tab completion gets messed up by the output on
stderr. Suppress it.
Fix the same basic issue in zsh_completion, thanks to the work by
Florian Pritz <bluewind@xinu.at>.
Signed-off-by: Dan McGee <dan@archlinux.org>
Was able to get my hands on one of these boxes today, so add yet another
new way of doing this. I'm glad these calls are so standardized. This
was compile tested on Linux and Illumos and seems to still be working in
both places.
Signed-off-by: Dan McGee <dan@archlinux.org>
Rework the frontend and backend to allow passing a ratio value in for
UseDelta rather than having a hardcoded #define-d 0.7 value always used.
This is useful for those with fast connections, who would likely benefit
from tuning this ratio to lower values; it is also useful for general
testing purposes.
The libalpm API changes for this, but we do support the old config file
format with a no-value 'UseDelta' option; in this case we simply use the
old default of 0.7.
We clamp the ratio values to a sane range between 0.0 and 2.0, allowing
ratios above 1.0 for testing purposes.
Signed-off-by: Dan McGee <dan@archlinux.org>
The entry's name is only used when not "." or ".." so only print the
string then.
Signed-off-by: Olivier Brunel <i.am.jack.mail@gmail.com>
Signed-off-by: Dan McGee <dan@archlinux.org>
Currently, a transaction is considered to be purely package removal
until the first package install is found. This resulted in the
removed packages at the start of a combined upgrade/removal transaction
not getting the "[removal]" output.
Fixes FS#27981.
Signed-off-by: Allan McRae <allan@archlinux.org>
Signed-off-by: Dan McGee <dan@archlinux.org>
Return codes from gpg commands are currently lost. This adds the functionality
of taking non-zero exit statuses from gpg. This includes error reporting for all
gpg commands that are run individually, run in a loop, and run through a pipe.
Includes the check_keyids_exist function which verifies a key exists locally
prior to attempted local manipulation of the key.
If a gpg command has a non-zero status, pacman-key will now exit with a non-zero
status. It will print a gettext error message of gpg's failure.
Signed-off-by: canyonknight <canyonknight@gmail.com>
Signed-off-by: Allan McRae <allan@archlinux.org>
Signed-off-by: Dan McGee <dan@archlinux.org>
Adds functions for every gpg command. By pulling out the gpg commands from the
"program start" section, additional commands can be run before or after a
specific gpg command without adding additional clutter to the function call
section.
Adds an explicit exit status of 0 to prevent arithmetic expansions from
returning non-zero, thereby falsely causing pacman-key to have a non-zero exit
status.
This change creates the framework for additional error messages and better
exit statuses being added to every pacman-key gpg call.
Signed-off-by: canyonknight <canyonknight@gmail.com>
Signed-off-by: Allan McRae <allan@archlinux.org>
Signed-off-by: Dan McGee <dan@archlinux.org>
Use consistent blank lines across all commands, get rid of the
translated double newlines which only serve to confuse translators, and
fix -h/--help for that extra special third command this script offers.
Signed-off-by: Dan McGee <dan@archlinux.org>
When asking question and stdin is piped, the response does not get printed out,
resulting in a missing \n and broken output (FS#27909); printing the response
fixes it.
Signed-off-by: Olivier Brunel <i.am.jack.mail@gmail.com>
Signed-off-by: Dan McGee <dan@archlinux.org>
We lost this logic somewhere between the libfetch and libcurl
transition, as it existed in the internal downloader, but was pulled
back only into the sync workflow. Add a helper function that will let us
check for existance in the filecache prior to calling the downloader.
Signed-off-by: Dave Reisner <dreisner@archlinux.org>
Signed-off-by: Dan McGee <dan@archlinux.org>
This removes the hack I added to skip '*.sig' files earlier since there
are other files that also fall into the same bucket- source packages
from `makepkg --source`, delta files, etc. Rather than prompting for
each and every one, simply skip them. Doing '-Scc' rather than '-Sc'
will delete these files if that is really what you want to do.
Signed-off-by: Dan McGee <dan@archlinux.org>
We don't need to open the data to be checked if we don't have a
signature to check against, so postpone that open until we know we have
either the base64_data or a valid signature file.
Signed-off-by: Dan McGee <dan@archlinux.org>
makepkg would not abort on a missing download agent due to the
output variable being declared local on the same line as the
function call in the assignment. That would result in strange output
such as:
==> Retrieving Sources...
==> ERROR: There is no agent set up to handle foo URLs. Check /etc/makepkg.conf.
Aborting...
-> Downloading foobaz...
/home/arch/code/pacman/scripts/makepkg: line 401: foo://foobaz: No such file or directory
==> ERROR: Failure while downloading foobaz
Aborting...
Signed-off-by: Allan McRae <allan@archlinux.org>
Signed-off-by: Dan McGee <dan@archlinux.org>
Declaring the variable as local on the same line as the assignment
results in result of the assignment being returned rather than the
result of the function on the righthand side of the assignment.
Declaring the variable as local on a separate line means the result
of the function on the r.h.s. is returned and our error function
will be invoked if necessary (although it is practically impossible
to ever trigger it...).
Signed-off-by: Allan McRae <allan@archlinux.org>
Signed-off-by: Dan McGee <dan@archlinux.org>
Simplify the source tarball generation by unifying the handling of
local and remote files. This also allows local files to be found
in $SRCDEST (FS#26580) and makepkg will abort on missing local source
files (only possible to trigger in combination with --skipinteg).
Signed-off-by: Allan McRae <allan@archlinux.org>
Signed-off-by: Dan McGee <dan@archlinux.org>
"invalid" in this case simply means files that may or may not be
archives. Discovered via a `pacman -Sc` operation with delta files in
the package cache directory, but can be triggered if any file is passed
to `pacman -Ql` that isn't an archive, for instance, or if the sync
database file is not an archive.
Fix it up so we are more careful about calling archive_read_finish()
only on archives that are valid and have not already been closed, and
teach our archive open function to set the returned archive to NULL if
we aren't going to be returning something valid anyway.
Signed-off-by: Dan McGee <dan@archlinux.org>
This adds an additional check step to find files in the local database
that claim to be owned by more than one package at once, which is
definitely not a supported setup.
Signed-off-by: Dan McGee <dan@archlinux.org>
In both cases we can go with the slightly leaner <stdint.h> header
include since we aren't using the print macros.
Signed-off-by: Dan McGee <dan@archlinux.org>
A look at what this does on 64 bit systems since we were using the
unnecessarily large 'unsigned long' type before even though it was 64
bits wide:
$ ~/bin/bloat-o-meter libalpm.so.old lib/libalpm/.libs/libalpm.so
add/remove: 0/0 grow/shrink: 0/4 up/down: 0/-10412 (-10412)
function old new delta
md5_finish 370 356 -14
sha2_finish 547 531 -16
md5_process 3762 2643 -1119
sha2_process 20356 11093 -9263
The code size is nearly halved in the sha2 case (44% smaller code size),
and md5 gets a nice size reduction (27% smaller) as well.
We also move base64 code to <stdint.h> types as well; we can use
'uint32_t' rather than 'unsigned long' for at least two variables in the
decode function. This doesn't net the same size benefit as the hash code
case, but it is more proper.
Signed-off-by: Dan McGee <dan@archlinux.org>
PGP keyservers are pieces of sh** when it comes to searching for
subkeys, and only allow it if you submit an 8-character fingerprint
rather than the recommended and less chance of collision 16-character
fingerprint.
Add a second remote lookup for the 8-character version of a key ID if we
don't find anything the first time we look up the key. This fixes
FS#27612 and the deficiency has been sent upstream to the GnuPG users
mailing list as well.
Signed-off-by: Dan McGee <dan@archlinux.org>
Add 1 failing for the -Su case, and the same case using -S (and
passing).
This is based on a real (current) issue of upgrading staging chroots
with the new pacman in staging for a libarchive build, and a new
toolchain in testing.
Signed-off-by: Dave Reisner <dreisner@archlinux.org>
Signed-off-by: Dan McGee <dan@archlinux.org>
This patch changes a variety of small things related to our pkghash
implementation with an eye toward performance, especially on native
32-bit systems.
* Use `unsigned int` rather than `size_t` for hash sizes. We already
return ERANGE for any attempted creation of a hash greater than 1
million elements, so unsigned int is more than large enough for our
purposes. Switching to this type allows 32 bit systems to do native
math without helper functions from libgcc.
* _alpm_pkghash_create() now internally adds extra padding for
additional array elements, rather than that being the responsibility of
the caller.
* #define values are moved into static const values in pkghash.c; a new
`stride` value is also extracted (but remains set at 1).
* Division and modulus operators are removed from the normal find and
add paths if possible. We store the upper limit of the number of
elements in the hash so we no longer need to calculate this every
element addition. When doing wraparound position calculations, we only
apply the modulus operator if the value is greater than the number of
buckets.
Signed-off-by: Dan McGee <dan@archlinux.org>
Commit 43cad9c8 made the building of all docs depend on the Makefile.
However, the Makefile is generated after running ./configure so is
always newer than any pregenerated docs. This means that people
building from released pacman tarballs are forced to rebuild the
docs (and thus have asciidoc installed). That defeats the purpose
of prebuilding the documentation. Have the documentatin depends on
Makefile.am instead as this is probably what was intended.
Signed-off-by: Allan McRae <allan@archlinux.org>
Signed-off-by: Dan McGee <dan@archlinux.org>
be_package.c: In function 'parse_descfile':
be_package.c:181:28: error: comparison between signed and unsigned
integer expressions [-Werror=sign-compare]
ptr - key + 2 is guaranteed to be > 0 so we can cast to size_t
Signed-off-by: Allan McRae <allan@archlinux.org>
Signed-off-by: Dan McGee <dan@archlinux.org>
Now that filelist arrays know their own size, we don't need to do the
bookkeeping we used to do when they were linked lists. Remove some of
the counter variables and use math instead.
Signed-off-by: Dan McGee <dan@archlinux.org>
This reduces the number of functions we call by log(n) in this function,
and the inlined version is trivial and barely increases the size of the
function.
Signed-off-by: Dan McGee <dan@archlinux.org>
We don't need absolute floating point precision at all here; we can
stick to integer land and use milliseconds which are precise enough for
our purposes. This also removes most floating point math out of the
non-update code path.
Signed-off-by: Dan McGee <dan@archlinux.org>
This reduces the number of regcomp() calls when parsing delta entries in
the database from once per entry to once for the entire context handle
by storing the compiled regex data on the handle itself. Just as we do
with the cURL handle, we initialize it the first time it is needed and
free it when releasing the handle.
A few other small tweaks to the parsing function also take place,
including using the stack to store the transient and short file size
string while parsing it.
When parsing a sync database with 1378 delta entries, this reduces the
time of a `pacman -Sl deltas` operation by 50% from 0.22s to 0.12s.
Signed-off-by: Dan McGee <dan@archlinux.org>
In commit 4c5e7af32f, we changed this code to use the regex gathered
substrings. However, we failed to correctly store the delta file name
(leaking memory), as well as freeing the temporary string used to hold
the file size string.
Signed-off-by: Dan McGee <dan@archlinux.org>
More than likely the compiler will do the three operation breakdown we
had here before (2 shifts + subtraction), but let the compiler do the
optimizations and make the actual operation more obvious. This actually
slightly shrinks the function binary size, likely due to instruction
reordering or something.
Signed-off-by: Dan McGee <dan@archlinux.org>
