aboutsummaryrefslogtreecommitdiff
path: root/git
AgeCommit message (Collapse)Author
2011-07-07Added store support. Now the basic object IO is implemented, which shall be ↵Sebastian Thiel
enough for the first batch of work
2011-07-07Removed cache in PureCompoundDB as it had the tendency to slow things down ↵Sebastian Thiel
actually
2011-07-07Added trivial implementation for info and stream methods - info is very ↵Sebastian Thiel
inefficient, but can't help it. Basic repo tests don't work as dulwich ignores alternate files
2011-07-07Fixed submodule related issues once again - is it ever going to end ?Sebastian Thiel
2011-07-07Added default performance tests - these should help to measure something at ↵Sebastian Thiel
least, which implicitly includes pack handling. For the pack specific tests to work, one would need a pack interface though, which is currently not planned to be specifically exposed
2011-07-07Initial version of the DulwichType inheritance. For now, it inherits ↵Sebastian Thiel
everything from the existing implementation, but one by one things can be reimplmented to use dulwich. It also shows that py 2.6 is quite plagued from its new feature, which is actually a bug, as objects inability to accept any args makes mixins hard to use ...
2011-07-07removed now superfluous InstanceDecorator, as it was just complicating ↵Sebastian Thiel
things after all
2011-07-07Added auto-skip mixin metacls, some serious brainfuck, if the required ↵Sebastian Thiel
module was not found. Its actually a nice mix between decorators which are class types, and a mixin as a metaclass, which applies said decorator. The InstanceDecorator wouldn't actually be needed, but it adds flexibility. Maybe it should be removed ...
2011-07-06Added basis for initial dulwich integration. Many basic issues should ↵Sebastian Thiel
surface while integrating this
2011-07-06cmd: on windows, we will now fallback to using git.cmd, as the default ↵Sebastian Thiel
installation puts it into the path, instead of git.exe. Of course we warn about the performance penalty that goes with it. Fortunately, using a wrapped git.exe works when creating pipes to the process, admittedly I was a bit surprised by this. fixes #26
2011-07-05Implemented GIT_PYTHON_GIT_EXECUTABLE including test and docsSebastian Thiel
2011-07-05fixes python 2.6 compatibility issuesSebastian Thiel
2011-07-05util: pick the type of memory manager based on the python version, to have ↵Sebastian Thiel
optimal results in all cases (at least the ones I can test) pack: now works properly with a sliding memory manager test_packedodb_pure: fixed very memory hungry implementation by using an iterator. This will of course reduce the measured performance a bit, but 750MB of memory is just a little bit too much for an ordinary test. Maybe it would be alright to just reduce the number of items ... but performance isn't a strength of python after all
2011-07-05util: added global sliding memory managerSebastian Thiel
pack: now using the global sliding memory manager. The current implementation uses assumes that packs are small enough to fit into memory right away, where the window size will be about 1 GB, as it never calls use_window() to assure the required offset actually exists. It will need to change to set the window appropriately.
2011-07-05Added smmap as submodule, package initialization adjusted to assure it is ↵Sebastian Thiel
available
2011-07-04Fetch info can now deal much better with non-default ref specs, see #23, ↵Sebastian Thiel
#24, #25
2011-07-04Basic remote functionality moved to Reference type, as it can in fact be ↵Sebastian Thiel
useful for tags as well, which might end up somewhere in the refs/remotes space. Its not likely that it will ever be used on a pure Reference instance though, but it is the smallest common base
2011-07-04refs: added constructor flag to allow refs to be instatiated from any path, ↵Sebastian Thiel
including simple test
2011-06-13Fixed implementation of _version_info which didn't work for python 2.4 ↵Sebastian Thiel
thanks to the rpartition method. omg
2011-06-13fixed submodule test to work on windowsSebastian Thiel
2011-06-13Submodule now uses a specialized method to remove its trees to allow ↵Sebastian Thiel
read-only files to be removed on windows as well
2011-06-13Made remote line parsing more stable. On windows it can be that we encounter ↵Sebastian Thiel
partial or intermixed lines from the pipe. This really shouldn't be, but its windows so it happens
2011-06-13test_submodule: Will now load everything into memory if smmap is present, to ↵Sebastian Thiel
help the submodule test to work even on windows
2011-06-13root submodule: Added parent_commit parameter for convenience. Fixed ↵Sebastian Thiel
submodule tests to run
2011-06-13Fixed submodule tests to go with either the old or the new version of gitdbSebastian Thiel
2011-06-08Fixed critical issue that would cause a string to be passed to methods that ↵Sebastian Thiel
expect a stream
2011-06-08symbolic reference handles different types for comparison more gracefully. ↵Sebastian Thiel
Fixed possible issue in test_refs, which occurred in 0.3 previously
2011-06-08util: Added test for iterable list, and implemented __contains__ and __del__ ↵Sebastian Thiel
functionality
2011-06-08pre-emptively moved all reference related tests into own package, to ↵Sebastian Thiel
resemble their actual location a little more. Renamed test_git to test_cmd as it tests the git cmd located in git.cmd (not in git.git)
2011-06-08Improved debug assertion to provide proper path on which to run a git daemonSebastian Thiel
2011-06-08git.version_info now returns exactly 4 numbersSebastian Thiel
2011-06-08log: non-existing logs no longer throw an exception, but are ignored. Fixed ↵Sebastian Thiel
critical bug which caused packed-ref files to be written with native line endings, which made git fail to parse it. I wonder why I never noticed this before, or ignored it. Unbelievable \!
2011-06-07Added additional opcodes to remote progress to make it compatible to newer ↵Sebastian Thiel
git versions. This bug existed for quite a while but didn't show up as progress wasn't sent most of the time. All methods that could use a progress will only activate it if a progress is actually given
2011-06-07Greatly improved robustness of config parser - it can now take pretty much ↵Sebastian Thiel
everything. Includes an updated config file which includes all the new additions
2011-06-07Match any number of leading spaces in config valuesJesse Keating
The regex comments state that any number of leading tabs or spaces should be allowed, however the regex was only catching zero or one space. This allows multiple spaces.
2011-06-07Making comparisons with non-GitPython objects more tolerant.Julien Miotte
2011-06-07The --progress flag will now automatically be used if possible when doing ↵Sebastian Thiel
any push or fetch operation
2011-06-07submodule now doesn't use hardcoded repository implementations anymore. ↵Sebastian Thiel
Instead it allows the user to override the type in the classmethod he calls. Otherwise the type of the own repo will be respected
2011-06-07Submodule tests are fully back and workingSebastian Thiel
2011-06-07Submodule tests are nearly working. Only root module needs more attentionSebastian Thiel
2011-06-07Fixed index checkout return code parsing to work with latest git versionSebastian Thiel
2011-06-07First run in order to fix the remote handling. Cleaned up interfaces and ↵Sebastian Thiel
figured out that the implementation really should be specific to the git command. This leaves the interface open for other implemntations which use a different way to provide feedback (as we do not make assumptions about the format of a feedback line)
2011-06-07clone and clone_from methods now support the RemoteProgress interface, using ↵Sebastian Thiel
similar functionality as used by the fetch, push and pull methods
2011-06-07Added version_info property to git commandSebastian Thiel
2011-06-07Fixed fetch/push/pull implementation. Next up is to integrate the ↵Sebastian Thiel
consolidation changes from master to make clone use the same facilities
2011-06-06Improved pack writing test to show that the pack generation can be lightning ↵Sebastian Thiel
fast with nearly now overhead if the data streams in fast enough (~30 MB/s when writing a pack). This shows that there is huge potential for sending packs, considering that we are actually recompressing them (without deltification). To be faster in future, we could probably just send ref-deltas or full objects as found in the pack without doing any recompression.
2011-06-06Streams returned by git cmd db are now containing all the data right away. ↵Sebastian Thiel
This could cause several copies to exist, and makes the cmd implementation a bad choice if big files are involved
2011-06-06Fixed performance pack writing tests. As they are actually depent on the ↵Sebastian Thiel
database (as streams have to be decompressed, it should be redesigned to have multiple database implementations)
2011-06-06Moved performance commit object test into the respective subfolder to ↵Sebastian Thiel
indicate what it tests
2011-06-06Fixed performance commit object testingSebastian Thiel