This commit adds an iterator interface for reading files from the
syncdbs. Instead of using alpm_pkg_get_files(), you now get the files
from the database using alpm_db_files_open(), you then use
alpm_db_files_next() to iterate through the files for each package. If
you want to actually load the files from that package you then use
alpm_db_files_load().
This means alpm_pkg_get_files() will always return empty for syncdbs,
even on .files databases, however these functions still work on the
localdb and loaded packages.
This aproach is faster when dumping the entire file list but slower when
searching for a specific package.
The memory usage of pacman is drastically less. See below.
build/pacman -Fl 0.55s user 0.01s system 99% cpu 0.556 total
build/pacman -Fl pacman 0.46s user 0.01s system 99% cpu 0.472 total
build/pacman -Fx pacman 2.88s user 0.09s system 99% cpu 2.965 total
pacman -Fl 1.60s user 0.13s system 99% cpu 1.731 total
pacman -Fl pacman 0.24s user 0.04s system 99% cpu 0.283 total
pacman -Fx pacman 2.45s user 0.14s system 99% cpu 2.593 total
Peak Memory
build/pacman -Fl 43.52MB
build/pacman -Fl pacmam 11.292MB
pacman -Fl 677.048MB
pacman -Fl pacman 163.288MB
Many of these are pointless (e.g. there is no need to explicitly turn on
spellchecking and language dictionaries for the manpages by default).
The only useful modelines are the ones enforcing the project coding
standards for indentation style (and "maybe" filetype/syntax, but
everything except the asciidoc manpages and makepkg.conf is already
autodetected), and indent style can be applied more easily with
.editorconfig
Signed-off-by: Eli Schwartz <eschwartz@archlinux.org>
Signed-off-by: Allan McRae <allan@archlinux.org>
Forcing vim users to view files with a tabstop of 2 seems really
unnecessary when noet is set. I find it much easier to read code with
ts=4 and I dislike having to override the modeline by hand.
Command run:
find . -type f -exec sed -i '/vim.* noet/s# ts=2 sw=2##' {} +
Signed-off-by: Florian Pritz <bluewind@xinu.at>
Signed-off-by: Allan McRae <allan@archlinux.org>
Return -1 if a path is too long to resolve or we run out of memory.
Signed-off-by: Andrew Gregory <andrew.gregory.8@gmail.com>
Signed-off-by: Allan McRae <allan@archlinux.org>
The _alpm_filelist_resolve function takes a filelist and creates
a list with any symlinks in directory paths resolved.
Signed-off-by: Andrew Gregory <andrew.gregory.8@gmail.com>
Signed-off-by: Allan McRae <allan@archlinux.org>
To improve conflict checking, we will need to make these functions
diverge to an extent where having two separate functions will be
preferable.
Signed-off-by: Allan McRae <allan@archlinux.org>
Signed-off-by: Dan McGee <dan@archlinux.org>
We have a few of these and might as well gather them together. This also
cleans up the code a bit by using an enum instead of integer values, as
well as makes a "search for file in filelist" function public so
frontends can do better than straight linear search of the filelists.
Signed-off-by: Dan McGee <dan@archlinux.org>
