graft: introduce --base option for using custom base revision while merging The graft command usually performs an internal merge of the current parent revision with the graft revision, using p1 of the grafted revision as base for the merge. As a trivial extension of this, we introduce the --base option to allow for using another base revision. This can be used as a building block for grafting and collapsing multiple changesets at once, or for grafting the resulting change from a merge as a single simple change. (This is kind of similar to backout --parent ... only different: this graft base must be an ancestor, but is usually *not* a parent.) This is probably an advanced use case, and we do thus not show it in the non-verbose help.

changegroup: add a option to create bundle with full snapshot only This is easy to implement now and can be useful for benchmarking.

changegroup: allow to force delta to be against p1 This new developer option is useful to general more "generic" bundle. Without this option, a bundle generated from the repository use deltas similar to the one stored in the specific repository it was generated from. This makes performance testing a bit tricky. Using deltas similar to the final result means all delta stored in the bundle can be applied to the target repository without any further processing (except for the rare case of a full snapshot). The application of such bundles (almost) never exercises the (slower) path of searching for a new valid delta. This result in unrealistic and too favorable timing and profile. Instead, we introduce an option to make sure all revisions are stored as a delta against p1. It might not be the best generation option, but it guarantees that the content will be "generic", not favoring a specific target.

fix: disable use of thread-based worker getfixes() accesses to repo, changectx, filectx, etc., so I believe there are code paths triggering data race. Mercurial API isn't thread safe in general.

storage: also use `deltamode argument` for ifiledata Now that lower level uses such argument, we can propagate the change to higher layers.

configitems: rename the config to prevent adding an alias in future Right now the config option looks like: [experimental.server] stream-narrow-clones= which does not match how config options are generally defined in core. So let's rename this to: [experimental] server.stream-narrow-clones= before the new release so that we don't have to add an alias in future for this. Differential Revision: https://phab.mercurial-scm.org/D5198

sparse-revlog: only refine delta candidates in the sparse case (issue6006) Starting with 5aef5afa8654, a valid delta parent might be "refined". This allows repository using sparse-revlog to produce better delta chain by using better intermediate snapshot base. However, this refining step was performed in all cases, including for repository not using sparse-revlog. This could produce a strange chain in the general delta case and corrupted repository in the non-general delta case. We now skip this step unless sparse-revlog is in use. In issue 6006, Yuya Nishihara provided a test case using an external repository, so we did not include it. Finding "laboratory" condition to reproduce this case and implementing an efficient test reproducing it is a bit tricky. We do not foresee to have the time to provide one by the release date. Differential Revision: https://phab.mercurial-scm.org/D5197

changegroup: refactor emitrevision to use a `deltamode` argument This new argument gathers the semantic of `sendfulltext` and `deltaprevious` in a single value. We are about to introduce a new type of constraints. Avoiding yet another argument sounds like a plus.

http: work around custom http client classes that refuse extra attrs I have no idea what is going on with our custom http client code at Google, but it chokes on these extra attributes we're tucking on http clients. Since it feels more than a little wrong to just stuff extra data on a client, let's degrade gracefully when the client class refuses the attributes.

branchmap: do not specify changelog as an argument Since (unfiltered)repo.changelog lookup gets as fast as __dict__ lookup, there's no point to pass in changelog instance. $ hg perfbranchmap --clear-revbranch -R mozilla-central ! base (orig) wall 20.593091 comb 20.600000 user 20.520000 sys 0.080000 (best of 3) (this) wall 20.129126 comb 20.130000 user 20.020000 sys 0.110000 (best of 3) This backs out most of the changes in 76d4272bd57b and 47c03042cd1d.