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

~bzr-pqm/bzr/bzr.dev » Changes from revision 0.20.13

Rev	Summary	Authors	Date
0.20.13	Play around a bit. Play around a bit. 1) Empty texts are no-op inserted, to avoid ever trying to match against their text. 2) If we find a new file-id and the compressor is more than half full, we go ahead and start a new compressor.	John Arbash Meinel	15 years ago
0.20.12	Change the code a little bit. Change the code a little bit. If a given text has not been seen before, insert all lines for that text. At present, we are doing worse than knit compression, because we have so many matching groups from various locations. Which causes us to just have huge swaths of copies. By inserting the full lines, we get more regions that we are able to generate a larger match against. This slows down the processing (10m => 24m), but improves compression (16MB => 12MB).	John Arbash Meinel	15 years ago
0.20.11	start experimenting with gc-optimal ordering. start experimenting with gc-optimal ordering.	John Arbash Meinel	15 years ago
0.20.10	Change the extraction ordering for 'unordered'. Change the extraction ordering for 'unordered'. Instead of using a random ordering, use the ordering defined by the index memos. This should give us the best group-locality. This gives a rather large performance improvement. Like 30s versus 10min.	John Arbash Meinel	15 years ago
0.20.9	Revert previous change. Revert previous change.	John Arbash Meinel	15 years ago
0.20.8	Allow writing negative offsets. Turns out not to actually co... Allow writing negative offsets. Turns out not to actually compress better. After zlib compression, negative offsets are a loss. Presumably because there is redundancy that zlib can factor out from bytes-since-start.	John Arbash Meinel	15 years ago
0.20.7	(ugly hack) autopacking doesn't work, so don't do it. (ugly hack) autopacking doesn't work, so don't do it. Force the fetch order and delta logic to use fulltexts in topological order. It isn't great, but it means things work.	John Arbash Meinel	15 years ago
0.20.6	Merge in the dev5 formats. Merge in the dev5 formats.	John Arbash Meinel	15 years ago
0.20.5	Finish the Fulltext => Chunked conversions so that we work i... Finish the Fulltext => Chunked conversions so that we work in the more-efficient Chunks.	John Arbash Meinel	15 years ago
0.20.4	Simplify the internals. We've already checked 'chk_support' ... Simplify the internals. We've already checked 'chk_support' so we don't need to check again.	John Arbash Meinel	15 years ago
0.20.3	Properly add GCPlainCHK to the pack_incompatible list. Properly add GCPlainCHK to the pack_incompatible list.	John Arbash Meinel	15 years ago
0.20.2	Teach groupcompress about 'chunked' encoding Teach groupcompress about 'chunked' encoding	John Arbash Meinel	15 years ago
0.20.1	Import repo_registry earlier. Import repo_registry earlier.	John Arbash Meinel	15 years ago
0.17.28	Do not register a chk repository format unless chk is availa... Do not register a chk repository format unless chk is available.	Robert Collins	15 years ago
0.17.27	Fix typo. Fix typo.	Robert Collins	15 years ago
0.17.26	Working better --gc-plain-chk. Working better --gc-plain-chk.	Robert Collins	15 years ago
0.17.25	Preliminary --gc-plain-chk support. Preliminary --gc-plain-chk support.	Robert Collins	15 years ago
0.17.24	Add a group cache to decompression, 5 times faster than knit... Add a group cache to decompression, 5 times faster than knit at decompression when accessing everything in a group.	Robert Collins	15 years ago
0.17.23	Only decompress as much of the zlib data as is needed to rea... Only decompress as much of the zlib data as is needed to read the text recipe.	Robert Collins	15 years ago
0.17.22	really get gc working with 1.10 really get gc working with 1.10	Robert Collins	15 years ago

Older »