| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is equivalent to this upstream pull request:
https://github.com/terencehonles/fusepy/pull/72
It is needed for BuildStream to work on ppc64 architectures. Without this
fix, the FUSE safe-hardlinks filesystem returns garbage. One visible effect
is that running integration commands fails with "executable format error"
(due to the filesystem returning garbage in place of the actual binary).
https://gitlab.com/BuildStream/buildstream/issues/83
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The upstream of fuse.py is <https://github.com/terencehonles/fusepy>. It looks
pretty abandoned -- there have been no commits there since 5th April 2016 and
there are 17 pull requests dating from 2013 onwards that are still waiting for
responses from the maintainer.
There's no support in fuse.py for ppc64 platforms which means BuildStream is
unusable on that architecture at present. My pull request to fix that upstream
is being ignored; since the module is a single .py file provided under the
ISC license it is simplest to pull the whole thing into buildstream.git instead.
The version of fuse.py that's been embedded is the one from pull request #72,
from upstream release v2.0.4 (commit 0eafeb5).
|
|
|
|
| |
Fixes another regression from validating all the loaded node attributes.
|
|
|
|
|
|
| |
On slow machines, the actual push operation can spend several minutes
preparing before actually pushing, which is annoying if the push then
fails.
|
|
|
|
|
|
|
|
| |
A previous change introduced the usage of pytest.param to our test
suite in commit 7aba0bfc906409073c4d246ec009a8622551726a.
This was only added to pytest 3.1.0, and the tests can't work with older
versions.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A project can specify its artifacts cache sharing settings, and users can
define a default artifacts cache to use as a fallback.
With this change, users can also override the project configuration with
their own.
That means for a project named "libfoo", BuildStream will resolve the
artifacts-related options in the following order:
1. the projects.libfoo.artifacts options from the user configuration;
2. if the above was not defined, then the artifacts options from the
project configuration;
3. if the above was not defined, then the artifacts options from the
user configuration;
Fixes #87
|
|
|
|
|
|
|
| |
With this commit, we first look at the artifacts options in the project
configuration, then fall back on the user configuration if necessary.
Relates to #87
|
|
|
|
|
| |
Having them there will allow overriding the user configuration (the
context) with the project configuration.
|
|
|
|
|
| |
There's no need to set a default value, then ignore a KeyError when we
can directly use dict.get().
|
|
|
|
| |
Note: original patch by Tristan Maat on MR !92, wording modified.
|
|
|
|
| |
This fixes a regression from 6691d0eb7e20c5872d7a5f439f308ae46019777f.
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
| |
That was removed anyway, now use utils.sha256sum() to calculate
the expected tar ref.
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
| |
This little bit of code was being repeated a bunch of times
already throughout buildstream, looks sane enough to offer a utility.
|
|
|
|
|
|
|
|
|
| |
Currently, if BuildStream is unable to push to the artifact cache for
whatever reason, the build will just error out. Fix it so that if we are
unable to push to the shared cache, we give a warning to the user and
continue building.
Fixes https://gitlab.com/BuildStream/buildstream/issues/90
|
|
|
|
|
| |
Including changes for newly added bst push & pull
commands and also some changes to the shell command.
|
|
|
|
| |
So that interactive sandbox mode can be used in a test or automated scenario.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
bst shell command is now an optional argument, which
defaults to ['sh', '-i']. This allows one to override
the command including the shell, so one need not invoke
the shell, thus it fixes issue #86.
Instead of an option. This means we can pick up a complete command
line after the '--' token directly from the remaining argument vector,
which makes escaping of things a bit less complicated.
E.g.:
bst shell --scope run <target.bst> -- sh -c "echo Hello World"
Further, the --scope argument is removed in favor of defaulting
to the --scope=run behavior, and adding a --build option
(use `bst shell --build <target>` for build shells, otherwise
juse use `bst shell <target>` for run shells).
Finally, the --builddir has been renamed to --sysroot; which
a bit more accurately describes what it is.
|
|
|
|
|
|
| |
o The command argument is a list, not a string
o The default value for the command list is ['sh', '-i']
o The sandbox is always run interactively
|
| |
|
| |
|
|
|
|
| |
Fixes #63
|
|
|
|
|
| |
This composite policy is on the way out the door anyway, first
lets reduce some unused code though.
|
|
|
|
| |
This saves over 100MB from the resulting image size.
|
| |
|
|
|
|
|
|
| |
This report combines the coverage from the integration tests
and the pytest suite together into one report so we can see just
what parts of the code base we're covering collectively.
|
|
|
|
| |
combined
|
|
|
|
|
|
|
|
| |
The remove() API is not used anywhere, it was previously used
by ArtifactCache.remove() which itself was unused.
If we need it in the future, it will be right here in
the history.
|
|
|
|
|
| |
The remove() API is not used anywhere, except for previously
being tested.
|
|
|
|
|
|
|
|
|
|
| |
Except for the ArtifactCache.remove() test, all of these assertions are
covered by the frontend `push` and `pull` tests. The remove() API itself
is unused except for being tested, lets leave this functionality out
until it's used.
The highlevel tests are easier to write and exercise more of the
codebase to assert the right behaviors, so lets only maintain that.
|
|
|
|
|
|
|
| |
The coverage reports do not include codepaths reached by
forked processes with multiprocessing unless we configure
this, now we have a more accurate report (and up from 69%
to 73% !)
|