| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
this also convers scripts file to a tar archive.
|
|
was left there unintentionally while debugging stuff.
|
|
|
|
snprintf is dog slow. make the blob stuff have some helper functions
so we can use them in code paths that are executed often.
|
|
avoid recalculating hashes, and store the lengths of names, so
we can optimize some operations.
|
|
also make the array code more explicit to have gcc optimizer happy.
|
|
|
|
some fixes on index reading code too.
|
|
in future we want to checksum on gzip boundary basis, not the
full file.
|
|
to actually get hardware acceleration if available.
|
|
allow .apk to consist of multiple separate gzip streams which
are just concatenated together.
|
|
We use APK_UPDATE_CACHE apk_flag instead.
|
|
Support for limiting output for only showing the packages who have
the status we are interested in. For example --limit '<' will only
list packages that have a newer version available.
While here, we also fix so packages that are not in any repository
are displayed with '?'.
|
|
This will update the repository cache upon db_open.
|
|
we need to know if caching is enabled, before the repositories are
added. otherwise the cache is not used properly at db opening time.
|
|
add all ciphers and built-in engines so we get e.g. hw acceleration
of sha1 when available.
|
|
use pkgconfig of zlib instead of hard requiring it. and link in
the pkg-config libs with --as-needed as not all of the openssl
libs are really required.
|
|
instead of having static md5 implemenation, use the openssl
library for digest functions.
|
|
|
|
and use this option in initramfs, so we are likely to get an usable
environment even if some packages from world are missing. fixes #50.
|
|
|
|
check for the full filename to match an expected pattern or
delete it (so we delete files with .new, .new.backup, etc.).
final part of remote package caching: fixes #49.
|
|
that way we get the same result as the 'apk add world' at boot
time; it might differ from currently installed set (ref #49).
|
|
more fine grained control what to load, and rename some of the
flags to be shorter.
|
|
|
|
printf %i does not work so well with size_t on 64 bit.
Use %zu which seems to be portable enough.
|
|
Administrative tool to download or delete files to/from the cache.
|
|
If /etc/apk/cache is a symlink to directory, a copy of all installed
packages is stored there, and the index of remote repositories will
be there instead of /var/lib/apk. This enables to reconstruct running
system during boot.
Left as todo: remove cached copy when the package is removed, and
additional apk applet to download missing packages to cache and/or
remove extra items.
|
|
And use it in couple of places. Some whitespace fixes too.
|
|
And add some more verbosity to the help message.
|
|
That will make the upgrade prefer packages available in repositories.
This is good if one want's to downgrade packages by removing an experimental
repository. Or to force re-install of locally built vs. repository version
when the package version are same, but checksum is different. Fixes #51.
|
|
Currently just goes through all world dependencies and updates them
where possible (ref #51).
|
|
|
|
We want be able to compare blobs so we basicly revert the old change,
and make a wrapper that takes version strings.
|
|
make apk_version_compare() take strings rather than blobs
add apk_pkgversion_compare(), a wrapper that takes packages
|
|
We might want to add an --update-index option to misc applets. For
example:
apk add --update-index -u package
apk version --update-index
|
|
and re-arrange the order of displayed info
|
|
By using a bitmask we can combine subactions
|
|
|
|
|
|
apk --help will list the generic options only and give a list of commands
To get the details for a spefic command, 'apk command --help' should be used.
|
|
Support version numbers specified with packages. For example:
apk add 'busybox<1.14'
apk add 'squid=>3.0'
|
|
since we dont verify the checksum we dont need to calculate it
Speed up when you try fetch lots of stuff thats already there.
|
|
|
|
We need to close the fd on destruction. This is what the corresponding
istream variant does too.
|
|
According the manpage readlink(2) does not append a null byte to buf.
So we have to do it ourselves.
|
|
Otherwise the iso image will have softlinks which is not what we want
|
|
leftovers from a debug session.
|
|
fixes #42
This will also fix a bug that left an empty file in destination dir when
source file did not exist in repository.
There are still issues with paths longer than 255 chars.
|
|
we should always close the in-stream, not only on failure.
|
