| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
| |
|
|
|
|
| |
This will fix bug #386095.
|
|
|
|
|
|
|
| |
This fixes _getMaskAtom to work correctly with the repo atoms that are
generated by MaskManager, solving the issue reported here:
https://github.com/cschwan/sage-on-gentoo/issues/90#issuecomment-2324953
|
|
|
|
|
|
|
|
| |
If we unconditionally exclude myremoved from myupdates, then removed
files will also be excluded from myheaders. This allows us to avoid
unnecessary separate manifest commits in cases when "myheaders" only
contained files that were being removed. Since our code always uses
myupdates + myremoved, this change in logic doesn't break anything.
|
|
|
|
|
| |
Don't populate the "myheaders" variable when we have thin-manifests
that contain only DIST entries.
|
|
|
|
| |
bug #340475.
|
|
|
|
|
| |
This fixes unintended extra blank lines since commit
1c8ff00a26b0e2e3627e9c552374f71235ee6c39.
|
|
|
|
|
| |
This allows us to properly test behavior of best_match_to_list with
different permutations of the input.
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
This should fix the issue shown in bug 375265, comment #10.
|
|
|
|
|
| |
These manifest-hashes warnings are irrelevant to regular users, so suppress
them by default.
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The manifest-hashes layout.conf setting simply overrides the hashes
that are generated. When unspecified, the MANIFEST2_HASH_DEFAULTS
constant is used instead (current value contains SHA1, SHA256, and
RMD160).
In order to enable SHA512 and WHIRLPOOL, and drop RMD160 and SHA1,
simply set "manifest-hashes = SHA256 SHA512 WHIRLPOOL" in layout.conf.
As with the previous boolean flags approach, manifest-hashes settings
become redundant and can be removed from layout.conf after a version
of portage with equivalent defaults is stable.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The portage.const changes are derived from an earlier patch that was
submitted by Robin H. Johnson <robbat2@gentoo.org>:
http://archives.gentoo.org/gentoo-portage-dev/msg_911574e7cb615f67e4c21fc75c043f65.xml
This patch has no affect on a repository unless it contains a
metadata/layout.conf file which contains boolean flags that enable or
disable manifest hash types. The plan is to deploy portage with this
patch, and once it has been stabilized, add the following
metadata/layout.conf settings to gentoo-x86:
manifest-rmd160 = false
manifest-sha1 = false
manifest-whirlpool = true
The above settings will become the default settings in a future portage
release, making them redundant and eligible for removal from
layout.conf (in order to avoid cluttering layout.conf with obsolete
information).
Future events:
After WHIRLPOOL is supported in stable portage:
- Add WHIRLPOOL to MANIFEST2_HASH_DEFAULTS.
- Remove SHA1 and RMD160 from MANIFEST2_HASH_*.
- Toggle gentoo-x86/metadata/layout.conf settings to match.
After WHIRLPOOL is supported in stable portage for at least 1 year:
- Change MANIFEST2_REQUIRED_HASH to WHIRLPOOL.
- Remove SHA256 from MANIFEST2_HASH_*.
- Toggle gentoo-x86/metadata/layout.conf settings to match.
After SHA-3 is approved:
- Add new hashes to MANIFEST2_HASH_*.
After SHA-3 is supported in stable portage:
- Toggle gentoo-x86/metadata/layout.conf settings to match.
After layout.conf settings correspond to defaults in stable portage:
- Remove redundant settings from gentoo-x86/metadata/layout.conf.
|
|
|
|
|
|
| |
If we have a local file, before we assume that we don't need to compute
any digests, make sure that the pre-computed digest types are exactly
the same types that we desire.
|
|
|
|
|
| |
This fixes commit f3101b3adce6731790f80f83fafece54b7bd8a63 to ensure
that Manifest.create() works correctly.
|
|
|
|
|
|
|
|
| |
The slot conflict display has better noise reduction than the
unsatisfied blockers display, so skip unsatisfied blockers display if
there are slot conflicts (see bug #385391). Note that this reverses
the logic from bug 159310, since the slot conflict display has evolved
a lot since then.
|
|
|
|
| |
This will fix bug #385413.
|
|
|
|
| |
This will fix bug #385341.
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Offer mhash as a provider for Manifest2 hash generation and validation.
This is important as either of pycrypto or fchksum offer an accelerated
Whirlpool implementation, and hashlib might not offer it. Additionally,
the mhash implementation is accelerated and ships with a rigorious
testsuite.
Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
|
| |
|
|
|
|
|
|
| |
Provide SHA512 hash algorithm to be used as new Manifest2 hash.
Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
|
|
|
|
|
|
|
| |
Provide public-domain implementation of the Whirlpool hash algorithm to
be used as new Manifest2 hash.
Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
|
|
|
|
|
|
| |
To be used shortly for WHIRLPOOL as well as RMD160.
Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
This allows ~* in KEYWORDS to be match by any unstable keyword in the
user's configuration, similar to how * in KEYWORDS already matches any
user configuration. Thanks to Daniel Robbins <drobbins@funtoo.org> for
the suggestion.
Also, the warning about * or -* in KEYWORDS is now gone, so that
ebuilds from the funtoo tree will be usable without triggering
warnings like this. If necessary, we can add ways to selectively
enable these kinds of warnings via layout.conf and/or repos.conf.
|
|
|
|
| |
This will fix bug #375613.
|
|
|
|
|
|
|
|
| |
This controls whether or not the layout.conf
"authoritative-cache = true" setting will be trusted for a particular
repo. It can be enabled globally by setting
"trust-authoritative-cache = true" in the [DEFAULT] section of
repos.conf.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
By authorative, this means "the cache is accurate; skip validation". While
a useful hint for a slight speedup in validation, the true gain is for
repositories that are distributed in a fashion that doesn't preserve mtime;
git primarily. Setting authorative-cache = true results in portage
skipping mtime validation checks for the bundled cache, allowing
for git vcs based repos to distribute a cache.
BUG=chromium-os:21049
TEST=dump a cache into metadata/cache, touch it to now, set layout.conf
to authorative-cache=true, verify it doesn't generate cache entries
for that repo.
Change-Id: I92423e679bc171d2411a18d6d3ac22e8ef457753
|
|
|
|
|
|
| |
This fixes a false --binpkg-respect-use warning that's triggered by
erroneous USE/IUSE comparison between the new-style virtual and an
old-style virtual match.
|
|
|
|
|
|
|
| |
This causes new-style virtuals to get pulled in for virtuals that are
already satisfied by installed old-style virtuals. This case is common,
due to PROVIDE being (removed without revision bump) from lots of
ebuilds.
|
|
|
|
|
|
|
|
| |
This removes a special case where manifests would first be commited
without signatures for category-level or greater commits. This case
behavior wasn't very useful anyway, and would be a problem if the
unsigned manifests got rejected by a commit hook (as may happen in the
near future).
|
| |
|
|
|
|
|
|
| |
Since the change involving slot conflict parent atoms in commit
6cea2091526659521d35be6c8dc7733f69f1a760, we want to check to make
sure we don't display any false positives in the "missed updates".
|