view hgext/git/TODO.md @ 48587:3c8cc987672e

simplemerge: take over formatting of label from `filemerge`

The padding we do of conflict labels depends on which conflict marker
style is used. For two-way conflict markers (the default), the length
of the base label shouldn't matter. It does before this patch,
however. This patch moves the formatting from `filemerge` to
`simplemerge`. The latter knows which conflict marker style to use, so
it can easily decide about the padding.

This change will allow us to use more descriptive "base" labels
without causing illogical padding in 2-way markers. I'll do that next.

One wrinkle is that we pass the same labels to external merge tools. I
decided to change that in this patch to be simpler: no padding, and no
ellipsis to fit within 80 columns. My reasoning is that the typical
external, 3-or-4-panel merge tool doesn't show the labels on top of
each others, so the padding doesn't make sense there. The ellipsis is
probably not necessary because the external tools probably have their
own way of dealing with long labels. Also, we limit them to "80 - 8"
to fit the "<<<<<<< " before, which is almost definitely not what an
external tool would put there.

Differential Revision: https://phab.mercurial-scm.org/D12019

Octopus Merge Support
=====================

This will be moderately complicated, as we'll need to synthesize phony
changeset entries to explode the octopus into "revisions" that only
have two parents each. For today, we can probably just do something like

    aaaaaaaaaaaaaaaaaaXX{20 bytes of exploded node's hex sha}

where XX is a counter (so we could have as many as 255 parents in a
git commit - more than I think we'd ever see.) That means that we can
install some check in this extension to disallow checking out or
otherwise interacting with the `aaaaaaaaaaaaaaaaaa` revisions.


Interface Creation
====================

We at least need an interface definition for `changelog` in core that
this extension can satisfy, and again for `basicstore`.


Reason About Locking
====================

We should spend some time thinking hard about locking, especially on
.git/index etc. We're probably adequately locking the _git_
repository, but may not have enough locking correctness in places
where hg does locking that git isn't aware of (notably the working
copy, which I believe Git does not lock.)