~bzr-pqm/bzr/bzr.dev : changes

~bzr-pqm/bzr/bzr.dev » Changes from revision 2255.2.170

Rev	Summary	Authors	Date
2255.2.150	Unbreak status --short. Unbreak status --short.	Robert Collins	17 years ago
2255.2.149	Crufty but existing _iter_changes implementation for Crufty but existing _iter_changes implementation for WorkingTreeFormat4.	Robert Collins	17 years ago
2255.2.148	lock a basis tree during a commit test. lock a basis tree during a commit test.	John Arbash Meinel	17 years ago
2255.2.147	Move fast id => path lookups down into DirState Move fast id => path lookups down into DirState It belongs there anyway, and it means all _get_entry(file_id) lookups will be improved, rather than just id2path.	John Arbash Meinel	17 years ago
2255.2.146	Implement move_directory by factoring out move_one Implement move_directory by factoring out move_one as a helper function, and then implementing a recursive update for all moved dirblocks.	John Arbash Meinel	17 years ago
2255.2.145	Support unbreakable locks for trees. Support unbreakable locks for trees.	Robert Collins	17 years ago
2255.2.144	Simplify update_minimal a bit more, by making id_index a Simplify update_minimal a bit more, by making id_index a cached variable, which is updated on mutating operations also use this to implement a faster id2path, and add direct tests since id2path now uses a more complex implementation.	John Arbash Meinel	17 years ago
2255.2.143	Update break-lock blackbox test to not break with dirstate a... Update break-lock blackbox test to not break with dirstate as the default tree format. Unfortunately this slightly reduces test coverage of the UI.	Robert Collins	17 years ago
2255.2.142	Simplify the update_minimal api a little bit, and document t... Simplify the update_minimal api a little bit, and document the parameters. Switch from kind => minikind, since we frequently already have that anyway, and update_minimal is a thin wrapper on changing the dirblocks in memory, not converting from an Inventory.	John Arbash Meinel	17 years ago
2255.2.141	Some small changes to move tests. Some small changes to move tests.	John Arbash Meinel	17 years ago
2255.2.140	Update tests to ensure basis tree is not modified Update tests to ensure basis tree is not modified Small bugfixes for things that would only be triggered if update_minimal failed.	John Arbash Meinel	17 years ago
2255.2.139	test cases for moving after a file has already been moved. test cases for moving after a file has already been moved. Fix some more WorkingTree4.move() bugs.	John Arbash Meinel	17 years ago
2255.2.138	implement several new WorkingTree.move() tests implement several new WorkingTree.move() tests and fix the bugs in WorkingTree4.move() that it uncovers.	John Arbash Meinel	17 years ago
2255.2.137	Move the WorkingTree.move() tests into their own module Move the WorkingTree.move() tests into their own module and remove the requirement for maintaining api compatibility for the new WorkingTreeFormat4, its ugly and deprecated anyway.	John Arbash Meinel	17 years ago
2255.2.136	(James Westby) add read locks around read_working_inventory(... (James Westby) add read locks around read_working_inventory() in test_commit_move	John Arbash Meinel	17 years ago
2255.2.135	Add locking in the test_commit_kind_changes test. Add locking in the test_commit_kind_changes test.	John Arbash Meinel	17 years ago
2255.2.134	Add a tree-test for get_symlink_target Add a tree-test for get_symlink_target and implement it in DirStateRevisionTree.	John Arbash Meinel	17 years ago
2255.2.133	Implement _paths2ids using bisect recursive rather than load... Implement _paths2ids using bisect recursive rather than loading the whole dataset from disk. But leave it disabled until it can actually benefit us by having other code paths not require the whole dir-state in memory.	John Arbash Meinel	17 years ago
2255.2.132	Implement _bisect_recursive, which uses multiple bisect call... Implement _bisect_recursive, which uses multiple bisect calls to handle renames and finding entries in subdirs. As is, this could be hooked into paths2ids() if the dirstate has not been loaded yet. However, it doesn't quite provide enough, since the parsed entries would probably not be saved. Further, the multiple bisect calls are less efficient then they could be, because they do not remember the last bisect call. We should explore switching to a caching structure, which maintains all records that have been processed, in a structure that can be in-memory searched before going back to disk.	John Arbash Meinel	17 years ago
2255.2.131	Change the return values for bisect functions so they just r... Change the return values for bisect functions so they just return the found dictionaries. This saves processing, and is more useful for a future _bisect_recursive function.	John Arbash Meinel	17 years ago