This is heavily based on Benjamin Tan's fish completions:
https://gist.github.com/bnjmnt4n/9f47082b8b6e6ed2b2a805a1516090c8
Some differences include:
- The end of a `--from`, `--to` ranges is also considered.
- `jj log` is not completed (yet). It has a different `--revisions` argument
that requires some special handling.
I left the "merge-tool-edits-conflict-markers" option unchanged,
since removing it would likely break some existing configurations. It
also seems like it could be useful to have a merge tool use the default
conflict markers instead of requiring the conflict marker style to
always be set for the merge tool (e.g. if a merge tool allows the user
to manually edit the conflicts).
This adds a new `revsets.simplify-parents` configuration option (similar
to `revsets.fix`) which serves as the default revset for `jj
simplify-parents` if no `--source` or `--revisions` arguments are
provided.
Layers are now constructed per file, not per source type. This will allow us
to report precise position where bad configuration variable is set. Because
layers are now created per file, it makes sense to require existence of the
file, instead of ignoring missing files which would leave an empty layer in the
stack. The path existence is tested by ConfigEnv::existing_config_path(), so I
simply made the new load_file/dir() methods stricter. However, we still need
config::File::required(false) flag in order to accept /dev/null as an empty
TOML file.
The lib type is renamed to StackedConfig to avoid name conflicts. The cli
LayeredConfigs will probably be reorganized as an environment object that
builds a StackedConfig.
Adds a new "git" conflict marker style option. This option matches Git's
"diff3" conflict style, allowing these conflicts to be parsed by some
external tools that don't support JJ-style conflicts. If a conflict has
more than 2 sides, then it falls back to the similar "snapshot" conflict
marker style.
The conflict parsing code now supports parsing Git-style conflict
markers in addition to the normal JJ-style conflict markers, regardless
of the conflict marker style setting. This has the benefit of allowing
the user to switch the conflict marker style while they already have
conflicts checked out, and their old conflicts will still be parsed
correctly.
Example of "git" conflict markers:
```
<<<<<<< Side #1 (Conflict 1 of 1)
fn example(word: String) {
println!("word is {word}");
||||||| Base
fn example(w: String) {
println!("word is {w}");
=======
fn example(w: &str) {
println!("word is {w}");
>>>>>>> Side #2 (Conflict 1 of 1 ends)
}
```
There's a subtle difference in error message, but the conversion function to be
called is the same. The error message now includes "for key <key>", which is
nice.
The previous iteration of `jj simplify-parents` would only reparent the
commits in the target set in the `MutableRepo::transform_descendants`
callback. The subsequent `MutableRepo::rebase_descendants` call invoked
when the transaction is committed would then rebase all descendants of
the target set, which includes the commits in the target set again.
This commit updates the `MutableRepo::transform_descendants` callback to
perform rebasing of all descendants within the callback. All descendants
of the target set will only be rebased at most once.
I believe this was an oversight. "jj duplicate" should duplicate commits (=
patches), not trees.
This patch adds a separate test file because test_rewrite.rs is pretty big, and
we'll probably want to migrate CLI tests to jj-lib.
The working-copy revision is usually the latest commit, but it's not always
true. This patch ensures that the wc branch is emitted first so the graph node
order is less dependent on rewrites.
If you have multiple remotes to push to, you might want to keep some changes
(such as security patches) in your private fork. Git CLI has one upstream remote
per branch, but jj supports multiple tracking remotes, and therefore "jj git
push" can start tracking new remotes automatically.
This patch makes new bookmarks not eligible for push by default. I considered
adding a warning, but it's not always possible to interrupt the push shortly
after a warning is emitted.
--all implies --allow-new because otherwise it's equivalent to --tracked. It's
also easier to write a conflict rule with --all/--deleted/--tracked than with
two of them.
-c/--change doesn't require --allow-new because it is the flag to create new
tracking bookmark.
#1278
I hope we'll have support for copies and renames in not too long. It's
good to have as many versions before that as possible without support
for `jj move`, in case we want to later use that to record a moved
file (maybe as an alias for `jj file move`).
There are two known limitations right now:
- Only statically known keys are suggested.
- Keys that the user did not set are still suggested for `jj config get`.
Running that suggestion may result in an error. The error message will be
appropriate though and there is some value in letting the user know that
this config value theoretically exists. Some users may try to explore what
configurations are available via the completions.
This is a workaround for command name completion. On zsh, the complete position
is specified by $_CLAP_COMPLETE_INDEX, and an empty argument isn't padded. So,
for "jj <TAB>", { args = ["jj"], index = 1 } is provided, and our expand_args()
helpfully fills in the default command "log". Since args[index] is now "log",
no other commands are listed.
Perhaps it would be nice to point to the revset docs from every
argument that takes a revset, or perhaps that would be too
verbose. `jj log` is perhaps where most people first run into revset
syntax, so I hope pointing to the docs from there is a good start.
This makes completions suggest aliases from the user or repository
configuration. It's more useful for long aliases that aren't used very often,
but serve the purpose of "executable documentation" of complex and useful jj
commands.
An earlier patch "resolved" aliases, meaning that any arguments following an
alias would be completed as if the normal command had been used. So it only
made sure that using aliases doesn't degrade the rest of the completions.
Commit ID: 325402dc9463ccaa70822069b3e94716d9cc7417
This commit modifies the commit graph used in the `rebase --after
--before` test case to add an additional branch of commits for use in
tests for `rebase -b` along with `--after --before`.
This commit extracts out the common code printing out the
`MoveCommitsStats` information into a shared function. The printed
output was also inconsistent between `-r` and `-s`/`-b` code paths, so I
standardized it to say "Rebased ? commits onto destination" for both
cases.
The destination commits are selected based on annotation, which I think is
basically the same as "hg absorb" (except for handling of consecutive hunks.)
However, we don't compute a full interleaved delta right now, and the hunks are
merged in the same way as "jj squash". This means absorbed hunks might produce
conflicts if no context lines exist. Still I think this is more intuitive than
selecting destination commits based on patch commutativity.
I've left inline comments to the tests where behavior is different from "hg
absorb", but these aren't exhaustively checked.
Closes#170
Unlike other documented aliases, it can be redefined by the user.
However, I think it's important for people to be informed that `b`
exists.
I left `amend`, `unamend`, and `ci` undocumented, since I think we were
considering removing or at least discouraging them.
Previously, attempting to modify an immutable commit only showed the
ID of the commit being modified, which wasn't very helpful when trying
to figure out which immutable commit is being modified at a quick
glance.
This commit prints the commit summary as a hint to make it simpler for
the user to see what the immutable commit is without having to run
`jj show <commit-id>`.
