rust: fix build errors on darwin I'm not all _that_ versed in Rust, but I think the root cause is that some constants are u16 rather than u32 on Darwin. I checked that the code still compiles on the latest Ubuntu. Differential Revision: https://phab.mercurial-scm.org/D11955

merge: with stable

help: properly document the default pattern The default isn't to glob. It's relpath, which does no globbing at all. Thus most commands and options like -X/-I and hg files require `glob:` in front of them to actually do any globbing. I think long ago the default was to glob, but that hasn't been true for a long time.

hg-core: dedup LazyAncestors Iterator impl … Which has been moved to the vcsgraph crate. AncestorsIterator cannot yet be removed because it is still being used in hg-core. Differential Revision: https://phab.mercurial-scm.org/D11948

hg-cpython: use ancestor iterator impls from vcsgraph Differential Revision: https://phab.mercurial-scm.org/D11947

hg-cpython: implement vcsgraph::Graph for our Index Differential Revision: https://phab.mercurial-scm.org/D11946

rust: add vcsgraph crate as dependency Preparing the integration of the vcsgraph library, which will provide generic algorithms for computations on version control graphs. Differential Revision: https://phab.mercurial-scm.org/D11945

simplemerge: don't calculate conflict labels when resolving automatically Differential Revision: https://phab.mercurial-scm.org/D11903

simplemerge: remove now-unused `localorother` argument from `merge_lines()` Differential Revision: https://phab.mercurial-scm.org/D11902

simplemerge: add a specialized function for "union", "local", "other" Differential Revision: https://phab.mercurial-scm.org/D11901

rhg: Fall back to Python if verbose status is requested by config Differential Revision: https://phab.mercurial-scm.org/D11943

rhg: Accept different "invalid ignore pattern" error message formatting At the moment rhg compiles all patterns into a single big regular expression, so it’s not practical to find out which file the invalid bit of syntax came from. Differential Revision: https://phab.mercurial-scm.org/D11942

rhg: Properly format warnings related to ignore patterns Differential Revision: https://phab.mercurial-scm.org/D11941

rhg: Sub-repositories are not supported Differential Revision: https://phab.mercurial-scm.org/D11940

simplemerge: make `localorother` a "mode" instead of a separate thing `simplemerge()` takes a `mode` argument, which can be "union", "merge" or "mergediff", and a `localorother` argument, which can be `None`, "local", or "other". The two options are not at all orthogonal -- most combinations don't make sense. Also, at least "union", "local", and "other" are very closely related. Therefore, it makes sense to combine them into one. It probably makes sense to split the `mode` argument into `resolve` and `marker_style`, where the former can be `None`, "union", "local", or "other", and the latter can be "merge", "merge3", "mergediff", or "minimize". This is a good step in that direction whether or not we end up doing that. Differential Revision: https://phab.mercurial-scm.org/D11887

simplemerge: avoid a call to `pycompat.strkwargs()` Differential Revision: https://phab.mercurial-scm.org/D11886

simplemerge: stop merging file flags As 384df4db6520 (merge: merge file flags together with file content, 2013-01-09) explains, we shouldn't do a 3-way merge of the symlink. However, since 84614212ae39 (flags: actually merge flags in simplemerge, 2020-05-16), we do that in `simplemerge.simplemerge()`. What's more, the merging of the executable flag there isn't actually necessary; it was made a no-op by the very next commit, i.e. 4234c9af515d (flags: read flag from dirstate/disk for workingcopyctx (issue5743), 2020-05-16). I found the overall flag-merging code (not the bit in `simplemerge.py`) very hard to follow, but I think I now finally understand how it works. `mergestate.resolve()` calculates the merged file flags and sets them on the local side of the merge (confusingly by calling `_restore_backup()`). Then it calls `filemerge.filemerge()`, which in turn calls `simplemerge.simplemerge()` (if premerge is enabled). That means that the flags on the local side `fcs.flags()` are already correct when the flag-merging code in `simplemerge.simplemerge()` runs. Interestingly, that code still works when the local side already has the merged value, it just doesn't change the value. Here's a truth table to explain why: ``` BLOMCAR 0000000 0011111 0101011 0111111 1000000 1010000 1100000 1111101 ``` B: Base L: Local O: Other M: Merged flags from `mergestate.resolve()`, i.e. what's called "local" when we get to `simplemerge.simplemerge()` C: `commonflags` in `simplemerge.simplemerge()`, i.e. `M & O` A: `addedflags` in `simplemerge.simplemerge()`, i.e. `(M ^ O) - B` R: Re-merged flags `simplemerge.simplemerge()`, i.e. `C | A` As you can see, the re-merged flags are always unchanged compared to the initial merged flags (R equals M). Therefore, this patch effectively backs out 84614212ae39. (I might later refactor this code to have the flags explicitly passed in.) `simplemerge.simplemerge()` is also called from `contrib/simplemerge.py`, but that code never passes any flags. Differential Revision: https://phab.mercurial-scm.org/D11879

filemerge: stop returning always-`True` value Now that we've removed some more leftovers from "merge driver", it's clear that the first element of `filemerge()`'s return value is always `True`. Differential Revision: https://phab.mercurial-scm.org/D11885

filemerge: remove unused `orig` argument from tool functions Differential Revision: https://phab.mercurial-scm.org/D11878

filemerge: stop passing around 3 unused `None` values in `files` argument Since 67cfffbfb6a0 (filemerge: eliminate most uses of tempfiles, 2017-08-31), we have used only the last entry in the `files` tuple that gets passed around to various functions. That commit said "Emphasize that they're unused so we can more easily remove them later.". It's now "later". Differential Revision: https://phab.mercurial-scm.org/D11877

filemerge: rename backup variables from `back` to `backup` `backup` seems like an obviously clearer name for something containing something related to a backup. Differential Revision: https://phab.mercurial-scm.org/D11876

setup.py: fix some documentation typos Differential Revision: https://phab.mercurial-scm.org/D11944

dirstate-v2: Apply SECOND_AMBIGUOUS to directory mtimes too This would only be relevant in contrived scenarios such as a dirstate file being written with a libc that supports sub-second precision in mtimes, then transfered (at the filesystem level, not `hg clone`) to another system where libc *doesn’t* have sub-second precision and truncates the stored mtime of a directory to integer seconds. Differential Revision: https://phab.mercurial-scm.org/D11939

rank: actually persist revision's rank in changelog-v2 The changelog v2 format is now persisting whatever "rank" value is recorded. However keep in mind that for we do not record any value. Differential Revision: https://phab.mercurial-scm.org/D11937

rank: add a "rank" value to the revlog-entry tuple The rank of a revision is the size of sub-graph it defines as a head. In other words, the rank of X is the size of `ancestors(X)` (X included). This is a property that can help various algorithm and we intend to store it in changelog-v2. We start with adding this new information to the "entry tuple", with a default value. We will start to compute and persist the rank later. Differential Revision: https://phab.mercurial-scm.org/D11936

changelog-v2: fix an assertion error to display the right data We were checking on thing and displaying the other. Differential Revision: https://phab.mercurial-scm.org/D11935

changelog-v2: use helper constant in the code to pack/unpack entries This makes the code simpler to read and modify. This is probably less efficient, but we are talking about the Python implementation here, so performance is concerne secondary to readability. Differential Revision: https://phab.mercurial-scm.org/D11934

revlog: add some information about the revision we cannot find Parameter to Exception are good, use them. Differential Revision: https://phab.mercurial-scm.org/D11933

rhg: Use binary search in manifest lookup … instead of linear scan, when looking for a single entry based on its path. Manifest entries are sorted by path, but are variable-size so we can’t use the standard library’s `[T]::binary_search`. We can still jump to a byte index and then look around for entry boundaries. Differential Revision: https://phab.mercurial-scm.org/D11932

rust: Upgrade to format-bytes 0.3 This removes use of the proc-macro-hack crate, which is possible now that we don’t support Rust 1.41 to 1.44 anymore. This in turn fixes spurious errors reported by rust-analyser: https://github.com/rust-analyzer/rust-analyzer/issues/9606#issuecomment-919240134 Differential Revision: https://phab.mercurial-scm.org/D11938