| Age | Commit message (Collapse) | Author |
|---|
|
possible input types, DiffIndex still requires implementation though
|
|
diff module, which probably is the right thing to do anyway
|
|
open vim to get a message though which makes the program appear to freeze - using git-status now
|
|
create objects, whose import would create a dependency cycle
|
|
is missing - these tests have to be redone in test-diff module accordingly
|
|
Diff class has been prepared to process raw input, but its not yet more than a frame
|
|
this was assumed - now there is a test for it as well
repo: removed diff and commit_diff methods, added 'head' property returning the current head as Reference object
|
|
|
|
stream - previously it would cache everything to memory and try to provide zipping functionality itself
gitcmd: allows the output stream to be set explicitly which is mainly useful for archiving operations
|
|
do it 'barely' only. New method names closely follow the default git command names
|
|
|
|
create an object directly - this feature is used in several places now, allowing for additional type-checking
|
|
achieved using the iter_commits method; reorganized methods within the type as a start for more interface changes
|
|
Renamed Commit.commits to iter_commits
repo: assured proper use of the terms revision ( rev ) and reference ( ref )
|
|
obsolete once the interface overhaul is finished. This commit is just intermediate
|
|
memory
time imports cleaned up and mostly removed as they were not required (anymore)
|
|
achieved throughput
repo.commits: max_count is None by default
moved benchmark-like test from test_commit to test_performance
|
|
safe possibly hundreds of command calls
|
|
work, but assures refs behave as expected
|
|
possible input ref - the objects pointed to by refs are now baked on demand - perhaps it should change to always be re-retrieved using a property as it is relatively fast - this way refs can always be cached
|
|
cannot handle the complexity of the command backend
All objects but Tree now use the persistent command to read their object information - Trees get binary data and would need their own pretty-printing or they need to parse the data themselves which is my favorite
|
|
information retrieval
|
|
currently it cause hundreds of command invocations which is slow
Fixed issue with trees not properly initialized with their default mode
_set_cache_: some objects checked whether the attribute was within their __slots__ although it should have been accessed through its class
|
|
output directly from the output stream
commit: now reads commit information directly from the output stream of the process by implementing its iterator method
repo: removed log method as it was redundant ( equal to the commits method )
|
|
|
|
preparation for command changes
|
|
|
|
different packages
|
|
Removed tests that were testing that method
|
|
dict as cache is a problem as the tree is ordered, added blobs, trees and traverse method
repo: remove blob function as blobs are created directly or iterated - primitve types should not clutter the repo interface
|
|
|
|
case someone tries to access an unset path or mode - this information cannot be retrieved afterwards as IndexObject information is kept in the object that pointed at them. To find this information, one would have to search all objects which is not feasible
|
|
to dynamically retrieve their objects
Improved way commits are returned by refs as they now use the path to be sure they always point to the ref even if it changes - previously it would use the sha intead so it would not update after being cached on the ref object
|
|
_list_from_string to indicate their new status as private method, adjusted all callers respectively
|
|
|
|
existance was doubtful or unsafe
|
|
mode is now generally an int compatible to the stat module
|
|
return the same thing which does not work anymore - re-implemented it in a more dynamic manner, but in the end tests will have to be revised anyway
Added slots to Diff and Stats type respectively
|
|
properly
|
|
whether it bakes more. This also leads to more efficient use of memory as values are only cached and set when required - the baking system does not require an own tracking variable anymore, and values are only to be cached once - then python will natively find the cache without involving any additional overhead. This works by using __getattr__ instead of __get_attribute__ which would always be called
|
|
|
|
|
|
editor to work with the files properly. Can convert it back for releaes
|
|
dyamically checks the existance of tags within the repository - it basically tests the interface and checks that expected return types are actually returned
|
|
test is in place which still fails on purpose. Need to integrate tags which can be objects or just a special form of a ref
|
|
changed to allow proper rename tracking
|
|
|
|
a|b_blob_id as it better reflects the actual value
actor regex now precompiled on class level
blob regex now precompiled on class level; made blame method more readable and faster although it can still be improved by making assumptions about the blame format and by reading the git command stream directly ( which is a general issue right now )
|
|
|
|
it would make existing code incompatible in some places
|
