relnotes: add 6.0.1 and 6.0.2 notes Differential Revision: https://phab.mercurial-scm.org/D12130

branchmap: stop writing cache for uncommitted data If we are about to write the branch while a transaction is active. we delay that write. After the transaction is closed, we flush all the write we delayed (unless they have been written in between). Differential Revision: https://phab.mercurial-scm.org/D12128

transaction: add a way to know a transaction has been finalized This will be useful to fix the timing of the branchmap on disk caching. Differential Revision: https://phab.mercurial-scm.org/D12127

branchmap: Add a test about writing branchmap and aborted transaction It turn out that we update the branchmap on disk whenever we recompute it… including in a middle of a transaction. This means writing the new branchmap too early (the changeset are not visible yet) and worse… it mean writing branchmap for transaction we rollback. so we introduce a test to highlight the issue (and prevent it to regress in the future). Differential Revision: https://phab.mercurial-scm.org/D12126

dirstate-v2: rename the configuration to enable the format The rename of the old experimental name was overlooked before the 6.0 release. We rename everything to use the new name (and keep the released name as an alias for compatibility). Differential Revision: https://phab.mercurial-scm.org/D12129

arbitraryfilectx: use our existing helpers for reading and writing files Differential Revision: https://phab.mercurial-scm.org/D12090

fix: remove unnecessary and overly strict check for divergence `rewriteutil.precheck()` checks for divergence these days, so we can remove the redundant check in `hg fix`. Differential Revision: https://phab.mercurial-scm.org/D12088

encoding: fix trim() to be O(n) instead of O(n^2) `encoding.trim()` iterated over the possible lengths smaller than the input and created a slice for each. It then calculated the column width of the result, which is of course O(n), so the overall algorithm was O(n). This patch rewrites it to iterate over the unicode characters, keeping track of the length so far. Also, the old algorithm started from the end of the string, which made it much worse when the input is large and the limit is small (such as the typical 72 we pass to it). You can time it by running something like this: ``` time python3 -c 'from mercurial.utils import stringutil; print(stringutil.ellipsis(b"0123456789" * 1000, 5))' ``` That drops from 4.05 s to 83 ms with this patch (and most of that is of course startup time). Differential Revision: https://phab.mercurial-scm.org/D12089

stream-clone: filter possible missing requirements using all supported one The `supportedformat` requirements is missing some important requirements and it seems better to filter out with all requirements we know, not just an "arbitrary" subset. The `supportedformat` set is lacking some important requirements (for example `revlog-compression-zstd`). This is getting fixed on default (for Mercurial 6.1) However, fixing that in 6.1 means the stream requirements sent over the wire will contains more items. And if we don't apply this fix on older version, they might end up complaining about lacking support for feature they actually support for years. This patch does not fix the deeper problem (advertised stream requirement lacking some of them), but focus on the trivial part : Lets use the full set of supported requirement for looking for unsupported ones. This patch should be simple to backport to older version of Mercurial and packager should be encouraged to do so. This is a graft of d9017df70135 from default. Differential Revision: https://phab.mercurial-scm.org/D12091

share-safe: enable by default (BC) The feature have been around for a year (4 version) and is quite important. Lets make it enabled by default. Differential Revision: https://phab.mercurial-scm.org/D11997