(not sure when I'll get to the changes, but removing from Needs Review, sorry I delayed so long doing that)

(Please queue this on stable, thanks!)

Note: This is meant for 'stable'

In D5094#77146, @yuja wrote:

> We might want to structure these variables as `base.node|short` instead of
>  `base_node` for example, but that would require more work.

[...]

Unfortunately, this doesn't work super well, when using the following flags:

There isn't a building block for a mapping dict holding mapping dicts, and
the formatter API doesn't support such structure either.

Fortunately, we don't need a formatter here since we just want to apply
a template to a single item. So if we had a wrapper for a single mapping
dict, the templater can be rendered as follows:

props = {
    'base': templateutil.mappingdict({'ctx': fca.changectx(), 'fctx': fca,
                                      'label': baselabel})
    ...
}
cmdutil.rendertemplate(fcd.changectx(), tmpl, props)

And base.path, base.node, etc. should just work.

In D5094#76457, @yuja wrote:

+def _describemerge(ui, repo, env, toolpath, args):
+ template = ui.config('ui', 'pre-merge-tool-output-template')
+ if not template:
+ return
+
+ # Remove HG_ prefix from entries in env and lowercase them
+ def sanitizeenv(k):
+ if k.startswith('HG_'):
+ return k[3:].lower()
+ return k

Nit: HG_FILE should be called as path per
https://www.mercurial-scm.org/wiki/GenericTemplatingPlan#Dictionary

hooper and I talked at the sprint, and they wrote up the following for why they feel we should keep this out of graphmod/templatekw for now:

My primary concern is migrating from hg-experimental remotefilelog to this. Making one of them compatible with the other at a wire protocol level would assist with the migration immensely, so I think that avoiding complex changes to the wire protocol would make sense. I think we should make changes to hg-experimental's remotefilelog to support any capabilities/command we implement here, so that the migration path is:

In D4606#70323, @yuja wrote:

Looks good to me. One nit.

• • a/mercurial/manifest.py +++ b/mercurial/manifest.py @@ -1203,7 +1203,7 @@ s._dirty = False self._loadfunc = _load_for_read
• def writesubtrees(self, m1, m2, writesubtree): + def writesubtrees(self, m1, m2, writesubtree, match=None):

Can you remove the unsupported default value match=None from inner functions?

In D4370#69050, @yuja wrote:

Queued, thanks.

+ def _loadchildrensetlazy(self, visit):
+ if not visit:
+ return None
+ if visit == 'all' or visit == 'this':
+ self._loadalllazy()
+ return None
+
+ todel = []
+ for k in visit:
+ kslash = k + '/'
+ ld = self._lazydirs.get(kslash)
+ if ld:
+ path, node, readsubtree = ld
+ self._dirs[kslash] = readsubtree(path, node)
+ todel.append(kslash)
+ for kslash in todel:
+ del self._lazydirs[kslash]

Any reason not to use self._loadlazy(kslash)?