| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
| |
Change-Id: I992dc0c1d40f563ade56a833162d409b02be90a0
|
| | |
|
| |\
| |
| |
| |
| | |
Reviewed-By: Adam Coldrick <adam.coldrick@codethink.co.uk>
Reviewed-By: Richard Maw <richard.maw@codethink.co.uk>
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This consolidates a bunch of code paths that were previously duplicated.
This also changes the API for local cached repos to match the
function names GitDirectory uses. Note that the remote repo cache still
uses the old names, and should be fixed when time permits.
Some unit tests that use the CachedRepo module required a bit of
inelegant monkey-patching in order that they continue to work. A better
way to do this would be with the 'mock' library (which would need to be
added to Baserock 'build' and 'devel' systems before we could use it).
|
| |/
|
|
| |
Also fix wrong indent.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
Previously, creating a GitDirectory object for something that wasn't a
Git repository would succeed, but most operations would raise a
cliapp.AppException with the following error message:
ERROR: Command failed: git config -z core.bare
Now, the constructor will raise a NoGitRepoError if the directory isn't
a Git repo, which (unless handled) gives the user the following sort of
message:
ERROR: Directory /src/ws/definitions is not a Git repository
|
| |\
| |
| |
| |
| | |
Reviewed-By: Pedro Alvarez <pedro.alvarez@codethink.co.uk>
Reviewed-By: Adam Coldrick <adam.coldrick@codethink.co.uk>
|
| | |
| |
| |
| | |
Clarify that bare repositories are supported, and other fixes.
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The GitDirectory() constructor, if passed a 'dirname' that doesn't
contain a '.git' subdirectory, can search upwards to find the real root
of the repository. This is used by the `add-binary`, `push`, and `pull`
commands.
This causes very confusing behaviour in the case that 'dirname' points to
a directory that should be a Git repository, but isn't, and that directory
is a path inside the working tree of another Git repository. Rather than
raising an error, in this case the GitDirectory class would perform
operations on a different repository to the one the caller expected.
This 'search_for_root' behaviour is now opt-in, to avoid confusion.
|
| |/
|
|
|
| |
This allows checking if a ref exists without requiring the caller
to use a try:, except: block.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Rather than using `git ls-remote` every time to see if there are changes
at the remote end, use a local cache.
Git already solves this problem with its refs/remotes/$foo branch
namespace, so we can just use that instead.
In addition, the detection of which upstream branch to use has been
improved; it now uses git's notion of which the upstream branch of your
current branch is, which saves effort in the implementation, and allows
the name of the local branch to differ from that of the remote branch.
This now won't notice if the branch you currently have checked out had
commits pushed from another source, but for some use-cases this is
preferable, as the result equivalent to if you had built before the
other push.
It may make sense to further extend this logic to check that the local
branch is not ahead of the remote branch, instead of requiring them to
be equal.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We assumed that the sha1 of the tree of the commit of the ref we care
about was sufficient to cache, but `git replace` means that you need to
know the state of other branches too, since anything in refs/replace can
completely change what the tree you check-out is.
This behaviour can be disabled globally by setting
GIT_NO_REPLACE_OBJECTS, so we're going to do that.
If we need to integrate a project that uses git-replace to change the
contents of their git trees then we could support that by:
if any(refs/replace/*):
potentially_replacable_objects = [
`git rev-parse HEAD`,
`git rev-parse HEAD^{commit}`,
`git rev-parse HEAD^{tree}`]
potentially_replacable_objects.extend(
`git ls-tree -r HEAD | awk '{print $3}'`)
# NOTE: doesn't handle submodules, and you'd need to expand this
# set whenever you process a replacement
for object in refs/replace/*:
if basename(object) not in potentially_replacable_objects: continue
cache_key['replacements'][basename(object)] = `git rev-parse $object`
If we were to support this would need to traverse the tree anyway, doing
replacements, so we may as well use libgit to do the checkout anyway,
and list which replacements were used.
However, since the expected use-case of `git replace` is as a better way
to do history grafting, we're unlikely to need it, as it would only have
any effect if it replaced the commit we were using with a different one.
Rubber-stamped-by: Daniel Silverstone
|
| |
|
|
|
|
|
|
| |
This reduces the vast number of 'git config -z core.bare' which
we used to get a lot of, to one or two per run. It doesn't
save much time, but it does make logs less full of confusion.
Reviewed-By: Richard maw <richard.maw@codethink.co.uk> +2
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
There was a check in it to see whether it needed to do the git branch
and git checkout based on whether the name of the branch matched that in
the morphology.
This had a couple of problems:
1. Now that we aren't always building from HEAD, we need to be able to
roll its commit back, so using the existing branch isn't always the
best idea.
2. It only checks the "ref" field, not "unpetrify-ref", so even though
we clone the right ref in there, it's checking the commit id against
the system branch name, so would always try to re-create the branch,
and fail when it already exists.
So now, we remove the original ref and re-create it with our preferred
HEAD.
A better solution might be to change the clone logic to not
automatically checkout HEAD, and instead require an explicit branch then
checkout, but the initial clone logic is shared with build, and I didn't
feel like tracking down all the different places that it was used.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Sorry about the big lump, I can split it into a nicer set of changes,
but they didn't naturally emerge in a nice series.
This creates a pushed_build_branch context manager, to eliminate code
duplication between build and deploy.
Rather than the build branch being constructed knowing whether it needs
to push the branch, it infers that from the state of the repositories,
and whether a local build would be possible.
If there are no uncommitted changes and all local branches are pushed,
then it doesn't create temporary branches or push them, and instead uses
what it already has.
It will currently create and use temporary build branches even for
chunks that have no local changes, but it's pretty cheap, and doesn't
require re-working the build-ref injection code to check whether there
are local changes.
|
| |
|
|
|
|
|
|
|
|
|
| |
This was noticed because our definitions.git had a dangling symlink,
so it failed to construct the temporary build branch.
We shouldn't process symlinks as morphologies either, since either we
make symlinked morphologies a first-class behaviour, and validate that
the link points inside the repository, which is a lot of work for
something we don't and probably won't need; or we can just assume that
we deal with the morphology this is linked to correctly anyway.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The default Python __repr__() is useless for both developers and users,
so avoid using it in error messages.
The following message:
ERROR: Push to remote <morphlib.gitdir.Remote object at 0x905096c>,
push url None with refspecs (<morphlib.gitdir.RefSpec object at 0xf72fec2c>,)
failed with exit code 128
Becomes:
ERROR: Push to remote "origin", push url None failed with exit code 128
|
| |
|
|
|
| |
This function causes a UnicodeDecodeError for some repositories
when building. Use os.path.isfile() when looking for .gitfat instead.
|
| |
|
|
|
|
| |
When cloning a repository, the files stored using git-fat need to
be pulled. This situation occurs in `morph branch`, `morph edit`,
and `morph checkout`.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
These commands are:
* git fat <init|push|pull>
* git pull
They are required for the morph add-binary and push/pull plugins. Also
make sure that GitDirectory is working in the root directory of the
specified git repository, and add some helper functions for handling
paths of files in the working tree.
|
| |
|
|
|
|
|
|
|
| |
gitdir._list_work_tree_files() needs to use os.relpath() instead of
direct string manipulation to avoid chopping off the first line of
every filename in cases where the base gitdir path string includes the
trailing /.
Unit test updated to catch this.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This commit introduces a new requirement: USERS MUST NOT HAVE SENSITIVE
DATA IN THEIR ENVIRONMENT. Otherwise it will be leaked into the system.
Note that configuration fields with 'PASSWORD' in their name are
stripped before writing the /baserock/deployment.meta file, so the
OpenStack OS_PASSWORD field is not leaked.
We want this so that we can run hooks at upgrade-time in the future.
These hooks might need to know how the system was configured and what
releaseuu it was. I'm not quite sure how we will define 'release' yet,
but by using `git tag` and `git describe` we are able to textually label
a time period in the history of the system's source code. We already
have the specific SHA1 of definitions.git stored in the system metadata,
so this should give us enough to be able to implement specific hooks
that work around any awkward upgrade complications we encounter in the
future.
|
| |
|
|
|
|
|
|
| |
Remotes have a push method, which takes multiple RefSpecs, runs git push
using arguments derived from the set of refspecs, then returns the
push's result.
If it fails the push, it will return the result in the exception.
|
| | |
|
| |
|
|
| |
Operations on remotes are now accessed through this proxy object.
|
| | |
|
| |
|
|
|
| |
This is used to create commit objects. This is used by build without
commit to provide the behind-the-scenes history.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This represents the state of the index of a GitDirectory.
Methods that use the index are now used via the GitIndex class, rather
than using the default index, as previously used when the methods were
in GitDirectory.
GitIndex may be constructed with an alternative path, which can be used
to manipulate a git checkout without altering a developer's view of the
repository i.e. The working tree and default index.
This is needed for `morph build` and `morph deploy` to handle the build
without commit logic.
|
| |
|
|
| |
This is needed for making commits without touching the workspace.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Some APIs should take SHA1 object IDs, there needs to be a way to get
one from a ref. To handle this, we add APIs for getting either the
commit or the tree pointed to by the ref.
Commits are provided because we need to create commits from
existing commits and update refs from existing values.
Trees are provided because we need trees to create commits, and we can
read trees into the index to eventually create another tree.
|
| |
|
|
|
|
|
| |
We need to use cat-file for files by SHA1, commits by SHA1 and files by
ref and path, so provide access in separate methods, since while it's
all the same thing "under the hood", it avoids the user needing to know
the command-line syntax.
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
| |
get_uncommitted_changes() is needed for morph status to tell if git
repositories have changes in them.
_get_status() is private, since it does not currently have a user, the
small amount of code in get_uncommitted_changes() wrapping _get_status()
is in the GitDirectory class instead of the branch and merge plugin,
since `morph build` will also need to know about uncommitted changes.
|
| |
|
|
|
| |
This is needed for status to tell if a repo is checked out in an
unexpected branch.
|
| |
|
|
|
|
| |
This adds methods to list and read files. The difference between doing
this to a commit and the currrent working tree is abstracted over by
whether the passed ref is None or omitted.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is needed when we load morphologies in "morph edit": it needs
to load things differently depending on whether we're loading from
the currently checked out ref or not.
The new method does not have a unit test, because I'm betting that
the test would not find a real bug. The test would just set up a
git repo and check whether the right ref is checked out in a couple
of ways (it is, and it isn't), and the code is simple enough that
it's unlikely to break the test. Creating the test, however, is a
fair bit of work. What the test would not catch is when we need the
method to do something quite different, but then we'll want a new
method anyway.
|
| |
|
