1742 commits

Author	SHA1	Message	Date
Yuya Nishihara	f568bac7f6	tests: rename test_annotate_command.rs to test_file_annotate_command.rs	2024-11-05 14:49:57 +09:00
Yuya Nishihara	987bb5e0bd	cli: util: don't use formatter to print raw data ... I don't think escape sequences in those contents should be escaped depending on the output device.	2024-11-04 13:33:37 +09:00
Yuya Nishihara	8f9d1a955f	cli: util: sort sub commands lexicographically	2024-11-04 13:33:37 +09:00
Yuya Nishihara	7d327d482a	cli: util: insert underscore to markdown-help command function name	2024-11-04 13:33:37 +09:00
Yuya Nishihara	09f9443417	cli: util: move command function closer to enum definition	2024-11-04 13:33:37 +09:00
Yuya Nishihara	0f7a4bf564	cli: util: split sub commands into modules	2024-11-04 13:33:37 +09:00
Yuya Nishihara	ce601efa3e	cli: bench: hyphenate command and bench names ... I don't think the names (without dashes) matter here.	2024-11-04 13:33:28 +09:00
Yuya Nishihara	541376b944	cli: bench: place command dispatcher function closer to enum definition	2024-11-04 13:33:28 +09:00
Yuya Nishihara	0ae282180a	cli: bench: split sub commands into modules	2024-11-04 13:33:28 +09:00
Yuya Nishihara	1c3a988371	cli: bench: extract sub commands to functions	2024-11-04 13:33:28 +09:00
Yuya Nishihara	0a73245b82	revset: move RevsetCommitRef::Root to RevsetExpression ... For the same reason as the previous patch. It's nice if root() is considered a "resolved" expression. With this change, most of the evaluate_programmatic() callers won't have to do symbol resolution at all.	2024-11-04 09:20:46 +09:00
Yuya Nishihara	a740eaeb86	revset: add convenient method that extracts symbol name from expression	2024-11-04 09:20:46 +09:00
Yuya Nishihara	15da697616	cli: insert dummy -h/--help flag when parsing early args ... Fixes #4746	2024-11-03 13:32:12 +09:00
Yuya Nishihara	8e1eb23246	cli: remove special case for clap's default "help" subcommand ... Since 536c629d "cli: Explicitly add a Help command to accept the early args after it", "jj help" is a normal subcommand.	2024-11-03 13:32:12 +09:00
Arthur Grillo	f2acb98ef2	help: Add a keyword feature ... It would be nice to not need to go the documentation website. This aims to solve that by introducing the concept of keyword to the help command. Basically, keywords are things that we want to add help messages to, but they don't necessarily have an associated subcommand. For now we only have two keywords: - `revsets`: Shows the docs for revsets - `tutorial`: Shows the tutorial that is on the documentation You get the keyword content by tipping `jj help --keyword revsets` or `jj help -k revsets`. You can also list the available keywords with `jj help --help`. It would be nice to have all the documentation on the keywords, maybe a next commit could do it.	2024-10-31 16:55:18 -03:00
dploch	0fa7df935c	rebase: fix typo	2024-10-30 14:40:54 -04:00
Martin von Zweigbergk	30ab71d340	bookmarks: add support for git.auto-local-bookmark (to match docs) ... We had documented that we support `git.auto-local-bookmark` but we don't. The documentation has been incorrect since d9c68e08b1. This patch fixes it by adding support for `git.auto-local-bookmark` with fallback to the old/current `git.auto-local-branch`. .	2024-10-30 08:01:02 -07:00
pylbrecht	8b5b4c19ea	cli: add jj config unset ... Allow unsetting config values similar to `git config unset`. ```bash $ jj config set --user some-key some-val $ jj config get some-key some-val $ jj config unset --user some-key $ jj config get some-key Config error: configuration property "some-key" not found For help, see https://martinvonz.github.io/jj/latest/config/. ``` Unsetting a key, which is part of a table, might leave that table empty. For now we do not delete such empty tables, as there may be cases where an empty table is semantically meaningful (https://github.com/martinvonz/jj/issues/4458#issuecomment-2407109269). For example: ```toml [table] key = "value" [another-table] key = "value" ``` Running `jj config unset --user table.key` will leave us with `table` being empty: ```toml [table] [another-table] key = "value" ```	2024-10-30 06:50:15 +01:00
pylbrecht	4575ea8794	cli: extract read/write config helpers ... This is a preparatory refactoring. We will introduce another function to remove config values, in which we will reuse these two helper functions.	2024-10-30 06:50:15 +01:00
Yuya Nishihara	e464c0e607	annotate: rename AnnotateResults to FileAnnotation ... The name "Results" was a bit misleading because Result<T, E> aliases are often called FooResult.	2024-10-29 23:33:46 +09:00
Yuya Nishihara	ab10b7c0a0	annotate: do not collect result lines into Vec, return Iterator instead ... We might want to calculate (commit_id, range) pairs of consecutive lines in order to "absorb" changes, for example. This should also be cheaper since Vec<u8> doesn't have to be allocated per line.	2024-10-29 23:33:46 +09:00
Arthur Grillo	bfaa5b8890	templates: Add new template alias builtin_log_compact_full_description ... For a new user, it is not clear how to view the full commit message/description of a change with `jj log`. This fix this, add a new template alias `builtin_log_compact_full_description` to display the commit like `builtin_log_compact` does but with a full description. The user can set it to true on the config like this: ``` templates.log = builtin_log_compact_full_description ``` Fixes: #3688	2024-10-26 14:01:18 -03:00
Dave Townsend	891fa88be0	diff: Allow setting the default level of context in config.	2024-10-25 17:56:56 +01:00
Yuya Nishihara	3268409d2f	cli: bookmark list: exclude local-only refs if --remote is specified ... The doc says --remote "shows all tracking and non-tracking remote bookmarks belonging to this remote." Suppose --remote=<remote> is a name pattern like "*@<remote>", it doesn't make sense that "jj bookmark list --remote=<remote>" includes local-only refs (whose "<name>" doesn't match the pattern.) The "--remote=<remote> --tracked" combination is still valid to exclude remote-only (i.e. untracked) refs.	2024-10-25 22:24:46 +09:00
Yuya Nishihara	c93d97d00b	cli: bookmark list: fix --tracked --remote to not include untracked refs ... Different filter options should intersect. The doc also says --tracked "shows remote tracked bookmarks only."	2024-10-25 22:24:46 +09:00
Yuya Nishihara	fc3e99d0e9	cli: bookmark list: extend --remote filter to string patterns ... I don't have a practical use case, but I think this is more consistent with NAMES arguments.	2024-10-25 22:24:46 +09:00
Yuya Nishihara	63785a24da	cli: bookmark list: simply apply --remote filter to source remote_refs	2024-10-25 22:24:46 +09:00
Yuya Nishihara	d40e06395e	cli: bookmark list: use Option<Vec<_>> to encode all-by-default filter types ... I feel .is_some() is easier to follow than !.is_empty().	2024-10-25 22:24:46 +09:00
Samuel Tardieu	5895df7d02	bookmark list: add --remote to filter remote bookmarks listing	2024-10-24 18:34:43 +02:00
Samuel Tardieu	8032ffdc46	git fetch: capitalize <REMOTE> in help message	2024-10-24 18:32:51 +02:00
Vincent Ging Ho Yim	5a8d5f9f5a	bookmark: fix broken docs URLs that lead to 404 errors ... Found a few more broken links beyond those in 9b5cdf2.	2024-10-24 18:02:16 +11:00
Yuya Nishihara	153873a093	diff: remove duplicated source message from AccessDenied error	2024-10-23 23:51:21 +09:00
Marco Neumann	9b5cdf25e6	bookmark: fix doc links ... The current links result in a 404. Replace them with the actual location of the respective pages.	2024-10-23 22:55:42 +09:00
Benjamin Tan	0a38fdc9d3	rewrite: move_commits: add MoveCommitsTarget enum to specify roots or commits to move ... This also allows some minor optimizations to be performed, such as avoiding recomputation of the connected target set when `MoveCommitsTarget::Roots` is used since the connected target set is identical to the target set (all descendants of the roots).	2024-10-22 20:39:50 +08:00
Yuya Nishihara	a493913000	revset: propagate evaluation errors from other Revset methods ... is_empty() could also return Result<bool, _>, but I think the current definition is also good. If an error occurred, revset.iter() would return at least one item, so it's not empty.	2024-10-22 09:03:53 +09:00
Yuya Nishihara	9d98ca491e	cli: propagate error from find_bookmarks_with() callback ... An error may be returned from revset containing_fn.	2024-10-22 09:03:53 +09:00
Yuya Nishihara	825b6670b3	revset: move containing_fn() type alias to lib ... If the return type were changed to Result<bool, _>, clippy would complain about the type complexity.	2024-10-22 09:03:53 +09:00
Yuya Nishihara	5dd6185fbc	templater: make "git_head" return boolean, render it as "git_head()" ... For the same reason as the root revision is indicated by "root()".	2024-10-21 09:21:34 +09:00
Yuya Nishihara	3d31928dac	revset: drop support for HEAD@git symbol resolution ... This was added at f5f61f6bfe "revset: resolve 'HEAD@git' just like other pseudo @git branches." As I said in this patch, there was no practical use case of the HEAD@git symbol. Suppose we implement colocated workspaces/worktrees #4436, there may be multiple Git HEAD revisions. This means HEAD can no longer be abstracted as a symbol of the "git" remote.	2024-10-21 09:21:34 +09:00
Yuya Nishihara	009284736d	templater: add pad/truncate functions ... As I said in the preceding patch, I settled on separate pad/truncate functions instead of a function taking multiple optional parameters. It's less efficient to process truncation and padding independently, but I don't think that would matter. The order of arguments follows the current f(..., content) convention. We can also add a method syntax, but I'm not sure if it's useful. In order to call a method of Template type, we'll need to coerce printable object to Template: concat(author.email()).truncate_end(10).pad_end(10) ^^^^^^ String -> Template FWIW, String type could provide more efficient truncate/pad methods. Closes #3183	2024-10-21 08:47:08 +09:00
Yuya Nishihara	6e06a79cfd	templater: add support for keyword arguments	2024-10-21 08:47:08 +09:00
Yuya Nishihara	ca5d119c10	text_util: add function to pad labeled text ... This could be inlined in templater implementation, but I decided to add text_util functions for ease of testing.	2024-10-21 08:47:08 +09:00
dploch	49e9003c4e	revset: allow iterators to return evaluation errors ... Custom backends may rely on networking or other unreliable implementations to support revsets, this change allows them to return errors cleanly instead of panicking. For simplicity, only the public-facing Revset and RevsetGraph types are changed in this commit; the internal revset engine remains mostly unchanged and error-free since it cannot generally produce errors.	2024-10-18 17:09:35 -04:00
dploch	c970181ae2	revset: change the commits() iterator to return RevsetEvaluationError ... This facilatates a change to the Revset trait in the next commit.	2024-10-18 17:09:35 -04:00
Yuya Nishihara	0eb4fe8389	text_util: add functions to truncate labeled text ... This will be used by truncate_start/end() template functions. I considered adding a template function that supports both padding and truncation, but the function interface looked a bit messy. There may be (max_width, ellipsis, left|middle|right) parameters for truncation, and (min_width, fill_char, left|center|right) for padding. I'm not going to add ellipsis and centering support, but it's weird if pad(center) implied truncate(middle).	2024-10-18 18:26:36 +09:00
Yuya Nishihara	8a1cdd9215	text_util: extract generic truncation helpers to support &[u8]	2024-10-18 18:26:36 +09:00
Yuya Nishihara	6a1db0dc21	text_util: add elide_end() function ... This function mirrors elide_start(), literally. We don't have any callers for the moment, but it helps write tests of inner truncation helpers. I'm going to add bytes version of these functions to implement "truncate" template functions.	2024-10-18 18:26:36 +09:00
Yuya Nishihara	9294507a1a	cli: leverage clap's default ArgGroup to handle config --user|--repo ... https://docs.rs/clap/latest/clap/_derive/index.html#arggroup-attributes	2024-10-17 22:57:47 +09:00
Samuel Tardieu	2112584973	style: remove unneeded lifetimes from trait implementations ... Clippy 1.83 (currently in beta) detects more cases of unneeded lifetimes, namely in trait implementation declarations. Since this lint is warn by default, we need to fix those instances to get a clean CI.	2024-10-17 13:35:06 +02:00
Ilya Grigoriev	61429739b3	cli branch move: document that we don't want to support -f for --from ... Reasoning: we don't want it confused with --force	2024-10-16 10:54:28 -07:00