summaryrefslogtreecommitdiff
path: root/lib
Commit message (Collapse)AuthorAgeFilesLines
...
* Index: handling an AttributeError exception raised in python 2.6.5 and newer ↵Sebastian Thiel2010-05-041-3/+12
| | | | if an unset slot is being deleted.
* Added additional tz_offset testing in performance test to call it more often.Sebastian Thiel2010-04-301-1/+1
| | | | Fixed test which required to be on master to work - it now uses the HEAD symbolic ref instead.
* Add support for time zone information in tags and commits.Rick Copeland2010-04-273-15/+31
| | | | | | | This commit includes - an update to git.objects.utils:parse_actor_and_date to parse the timezone offset - updates to the git.objects.Commit and git.objects.Tag objects to support *_tz_offset attributes - updates to tests in test.git.test_commit and test.git.test_refs to check for appropriate *_tz_offset attributes
* Converted all tabs to 4 space characters each to comply with pep8Sebastian Thiel2010-03-0419-6350/+6350
|
* Commit.iter_items: Will not restrict comits to the ones containing changes ↵Sebastian Thiel2010-02-281-1/+6
| | | | | | to paths anymore as it will only append '--' if paths are actually given. Added unittest to verify this
* git.Tree: Fixed critical issue when reading trees from binary data. The ↵Sebastian Thiel2010-02-241-8/+6
| | | | previous version was making assumptions that would only be true for old git repositories it sesms. The new version of the algorithm deals with this gracefully.
* IndexFile.write: Added special handling flag allowing to skip TREE extension ↵Sebastian Thiel2010-02-241-17/+27
| | | | data, which becomes important if git-write-tree is supposed to be used
* index: added move method including testSebastian Thiel2010-02-191-11/+88
| | | | test.helpers: temporary rw repository creators now set the working dir of the program, easing working with relative paths a lot
* LockFile: release_lock now checks whether the lockfile to be removed still ↵Sebastian Thiel2010-02-181-1/+6
| | | | exists. Previously it would just fail
* git.commit: Added test to assure we handle the first commit correctly ↵Sebastian Thiel2010-02-181-1/+1
| | | | regarding its parents
* Lockfile now uses low-level commands to interact with the filesystem in ↵Sebastian Thiel2010-02-141-30/+5
| | | | order to make it impossible for threads to overwrite each other if we are talking about miliseconds. This renders plenty of 'special' handling redundant
* Commit.create: now handles empty repositories correctlySebastian Thiel2010-02-131-2/+15
|
* IndexFile: unmerged_blobs lists are now sortedSebastian Thiel2010-02-132-4/+6
| | | | Repo.init: fixed incorrect use of the path which was to optionally specify where to initialize the repository. Update test as well
* Implemented RemoteProgress parsing for git-fetch, which might become ↵Sebastian Thiel2010-01-121-41/+92
| | | | | | available at some point natively, within the git suite Progress parsing now deals properly with Ascii_Escape characters that are meant for the tty - git might stop sending this at some point, but we can deal with it no matter what
* remote.py: improved push-progress parsin to better deal with messages ( i.e. ↵Sebastian Thiel2010-01-121-7/+12
| | | | throughput )
* tree: implemented recursive paths in __div__ and __getitem__ method, ↵Sebastian Thiel2009-12-301-6/+27
| | | | allowing the keys to contain slashes; adjusted test to check for this
* LockFile: id representing the instance that keeps the lock is unique now - ↵Sebastian Thiel2009-12-171-11/+25
| | | | locks will check whether the lock they wrote truly is theirs - in case threads are racing, this might not be the case. Now this issue will be detected and results in a proper failure
* config: GitConfigReader now allows to override its lock-type. By default it ↵Sebastian Thiel2009-12-172-9/+25
| | | | uses a 'failing' lock file, but now its possible to easily put a blocking lock file in its place
* Added imports for standard locking file classes into the git moduleSebastian Thiel2009-12-164-2/+7
|
* index: improved the way stdout is handled as the previous handling rarely ↵Sebastian Thiel2009-12-152-6/+9
| | | | | | caused lockups while waiting for stdout NOTE: This does not have the desired effect, the issue appears to be somewhere within git, possibly, as git simply does not terminate waiting for something, even if stdout is closed.
* refs: Added is_valid method to SymbolicReference base classSebastian Thiel2009-12-151-0/+13
|
* removed nonsense IndexLock implementationSebastian Thiel2009-12-101-32/+1
|
* index: IndexLock added to provide some kind of critical section for index ↵Sebastian Thiel2009-12-101-1/+33
| | | | based git operations in concurrent environments
* index: TempFileMover now handles the case that the file to move does not ↵Sebastian Thiel2009-12-102-4/+51
| | | | | | exist more gracefully Added BlockingLockFile which may be used to wait for the lock according to some criteria
* config: fixed incorrect handling of default value in get_valueSebastian Thiel2009-12-092-4/+6
| | | | remote.config: SectionConstraint now knows about set_value and get_value which are new to the GitConfigParser
* repo: Added alias for refs property, which is called 'references' , to ↵Sebastian Thiel2009-12-091-2/+5
| | | | improve usability. There was no reason that only 'refs' is shortened, whereas there are 'heads' and 'tags' as well
* config.set_value: added more convenient set_value method to ConfigReaderSebastian Thiel2009-12-081-0/+20
|
* config.get_value: Added default argument including testSebastian Thiel2009-12-081-2/+10
|
* index: Checkout exception now contains information about the reason of the ↵Sebastian Thiel2009-12-071-2/+10
| | | | failure as well, one per failed file
* Reference: reading of commit data is now safer and handles non-existing ↵Sebastian Thiel2009-12-072-3/+5
| | | | paths - previously it would run into a code-branch I forgot
* SymbolicRefence base is now fully aware of pack files in all operations.Sebastian Thiel2009-12-042-97/+171
| | | | Ref(anytype) Iteration was improved such that automatic filtering now also works for SymbolicReferences ( which only return symbolic refs)
* refs: Fixed incorrect retrieval of symbolic reference types - previously we ↵Sebastian Thiel2009-12-041-3/+3
| | | | only really knew heads, now we know references as a common base. The adjustment make the ref system as flexible as it was originally meant to be
* SymbolicReference creation now won't fail if the target already exists and ↵Sebastian Thiel2009-12-041-7/+18
| | | | has the same content as the new reference
* refs.rename: improved handling as it will only raise ( if not forced ) if ↵Sebastian Thiel2009-12-031-2/+7
| | | | the destination contains different data. If it contains the same, the move will work as we can do what the user intended
* refs: Implemented low-level (Symbolic)Reference renaming as some references ↵Sebastian Thiel2009-12-032-15/+46
| | | | cannot be reamed using the git-branch command if they are not in refs/heads, that is in a non-standard refs folder
* config: implemented get_value method to have a safe way to make general ↵Sebastian Thiel2009-12-031-0/+33
| | | | queries to the git configuration, returning a value in the proper type. In a way its not supposed to be used as you should know the type of your configuration option or get an exception otherwise
* repo: renamed directories to more descriptive identifiers and made them ↵Sebastian Thiel2009-12-037-43/+86
| | | | safer to use in case of bare repositories
* repo: fixed incorrect path handling thta could lead to issues if the ↵Sebastian Thiel2009-12-032-5/+12
| | | | repository was initialized with a .git repository, although it was not bare actually
* commit.create_from_tree: head will not be advanced anymore as it feels ↵Sebastian Thiel2009-12-011-1/+1
| | | | non-natural when using it
* index: Added BlobFilter utility class to make filtering of blobs from the ↵Sebastian Thiel2009-11-301-2/+27
| | | | index easier
* SymbolicReference.delete: Now takes SymbolicReference instances as well, not ↵Sebastian Thiel2009-11-302-1/+4
| | | | | | only paths diff.__str__: fixed incorrect message generation error
* Generalized custom reference creation down to SymbolicReference level. ↵Sebastian Thiel2009-11-301-35/+76
| | | | 'Reference' implementation now shares all the code from the SymbolicReference base. This implementation allows to create any reference you'd like without requireing git calls
* ref: Added general method to create and delete any reference pointing to any ↵Sebastian Thiel2009-11-301-2/+69
| | | | object. They are only constrained to be located in the refs/ directory
* Traversable.traverse: Added as_edge option allowing to receive the source of ↵Sebastian Thiel2009-11-271-11/+19
| | | | the item as well to enable predicates to do more proper checking
* Fixed issue with commit traversal due to missing visit_once in the traversal ↵Sebastian Thiel2009-11-272-0/+19
| | | | method, improved code performance by disabling the visit-once flag by default for trees, which by nature are not recursive
* ItemTraversal: Predicate and prune functions now provide depth information, ↵Sebastian Thiel2009-11-271-6/+6
| | | | allowing the callee to know more about its environment
* improved performance of item traversal, its nearly as fast as it was with ↵Sebastian Thiel2009-11-271-11/+2
| | | | the first very pure implementation
* tree: added traversal method, adjusted testsSebastian Thiel2009-11-263-41/+19
| | | | Fixed critical bug in object code: IndexObjects now use their path as hashkey, not the data\!
* objects.utils: Added Traversable base and implemented it for commits ↵Sebastian Thiel2009-11-263-1/+91
| | | | including a test
* SymbolicReference is now the base for all references, allowing them to ↵Sebastian Thiel2009-11-261-24/+53
| | | | | | appear litereally anywhere. A 'default' reference is nothing more than a detached symbolic reference. Added method iterate the packed refs file