Merge lp:~gz/brz/unmapped into lp:brz
Status: | Merged |
---|---|
Approved by: | Martin Packman |
Approved revision: | no longer in the source branch. |
Merge reported by: | The Breezy Bot |
Merged at revision: | not available |
Proposed branch: | lp:~gz/brz/unmapped |
Merge into: | lp:brz |
Prerequisite: | lp:~gz/brz/i_unzipping |
Diff against target: |
444 lines (+56/-51) 24 files modified
breezy/btree_index.py (+6/-3) breezy/dirstate.py (+2/-2) breezy/export_pot.py (+1/-2) breezy/groupcompress.py (+4/-1) breezy/mutabletree.py (+1/-1) breezy/osutils.py (+2/-1) breezy/plugin.py (+2/-2) breezy/plugins/changelog_merge/changelog_merge.py (+3/-3) breezy/plugins/weave_fmt/bzrdir.py (+1/-1) breezy/repository.py (+2/-2) breezy/smart/repository.py (+2/-2) breezy/tests/per_pack_repository.py (+1/-1) breezy/tests/per_versionedfile.py (+4/-4) breezy/tests/per_workingtree/test_paths2ids.py (+4/-4) breezy/tests/test_diff.py (+4/-4) breezy/tests/test_http.py (+2/-2) breezy/tests/test_rio.py (+2/-2) breezy/transport/memory.py (+2/-2) breezy/transport/sftp.py (+1/-1) breezy/util/simplemapi.py (+3/-4) breezy/versionedfile.py (+3/-3) breezy/vf_repository.py (+2/-2) breezy/weave.py (+1/-1) breezy/weavefile.py (+1/-1) |
To merge this branch: | bzr merge lp:~gz/brz/unmapped |
Related bugs: |
Reviewer | Review Type | Date Requested | Status |
---|---|---|---|
Jelmer Vernooij | Approve | ||
Review via email: mp+324568@code.launchpad.net |
This proposal supersedes a proposal from 2017-05-24.
Commit message
Make use of map Python 3 compatible
Description of the change
Cope with the builtin map function becoming imap in Python 3.
Worse case the fixer throws in an extra pointless list copy, where it looked like that might matter I refactored or pulled in the iterator version from future_builtins.
There are a few changes on somewhat hot paths, where larger refactors/rewrites are probably called for, but avoided doing so for now.
The static tuple intern change is somewhat driveby, we can't use sys.intern on bytestrings in Python 3 anyway, and a better interface would be constructing the tuple with a new method like from_bytes_interned or something to mega-intern the result.
Same as before just with prereq branch set correctly.