An implication of this current approach is that `NVIM_API_LEVEL` should be
bumped when a new Lua function is added.
TODO(future): add a lint check which requires `@since` on all new functions.
ref #25416
**Problem:** The documentation for `TSNode` and `TSTree` methods is
incomplete from the LSP perspective. This is because they are written
directly to the vimdoc, rather than in Lua and generated to vimdoc.
**Solution:** Migrate the docs to Lua and generate them into the vimdoc.
This requires breaking up the `treesitter/_meta.lua` file into a
directory with a few different modules.
This commit also makes the vimdoc generator slightly more robust with
regard to sections that have multiple help tags (e.g. `*one* *two*`)
Some composite/compound types even as basic as `(string|number)[]` are
not currently supported by the luacats LPEG grammar used by gen_vimdoc.
It would be parsed & rendered as just `string|number`.
Changeset adds better support for these types.
Problem:
The <br> hack in a0c64fe816 causes weird layout if a "h4 pseudo-heading"
is not the only tag on the line. For example in the help text below, the
"*'buflisted'*" tag was treated as h4 and followed by <br>, which is
obviously wrong:
*'buflisted'* *'bl'* *'nobuflisted'* *'nobl'* *E85*
'buflisted' 'bl' boolean (default on)
Solution:
Only treat a tag as "h4 pseudo-heading" if it is the only tag in the
line. This is fragile, but in practice seems to get the right balance.
Problem:
The right-aligned tag "pseudo-heading" layout mushes together with the
left-aligned text. This is especially messy in a narrow viewport.
Solution:
Put a `<br>` on it. This is a hack until tree-sitter-vimdoc recognizes
these pseudo-headings.
In the api_info() output:
:new|put =map(filter(api_info().functions, '!has_key(v:val,''deprecated_since'')'), 'v:val')
...
{'return_type': 'ArrayOf(Integer, 2)', 'name': 'nvim_win_get_position', 'method': v:true, 'parameters': [['Window', 'window']], 'since': 1}
The `ArrayOf(Integer, 2)` return type didn't break clients when we added
it, which is evidence that clients don't use the `return_type` field,
thus renaming Dictionary => Dict in api_info() is not (in practice)
a breaking change.
Problem:
It has long been a convention that references to the builtin terminal UI
should mention "tui", not "term", in order to avoid ambiguity vs the
builtin `:terminal` feature. The final step was to rename term.txt;
let's that step.
Solution:
- rename term.txt => tui.txt
- rename nvim_terminal_emulator.txt => terminal.txt
- `gen_help_html.lua`: generate redirects for renamed pages.
**Problem:** `vim.treesitter.get_parser` will throw an error if no parser
can be found.
- This means the caller is responsible for wrapping it in a `pcall`,
which is easy to forget
- It also makes it slightly harder to potentially memoize `get_parser`
in the future
- It's a bit unintuitive since many other `get_*` style functions
conventionally return `nil` if no object is found (e.g. `get_node`,
`get_lang`, `query.get`, etc.)
**Solution:** Return `nil` if no parser can be found or created
- This requires a function signature change, and some new assertions in
places where the parser will always (or should always) be found.
- This commit starts by making this change internally, since it is
breaking. Eventually it will be rolled out to the public API.
Problem:
The headings and help tags overlap when browsing the docs in neovim.io/doc/user/ from a mobile phone.
Solution:
Apply the correct CSS rules so that the headings and help tags wrap
nicely below one another.
This also makes shada reading slightly faster due to avoiding
some copying and allocation.
Use keysets to drive decoding of msgpack maps for shada entries.
Problem: No way to get the arity of a Vim function
(Austin Ziegler)
Solution: Enhance get() Vim script function to return the function
argument info using get(func, "arity") (LemonBoy)
fixes: vim/vim#15097closes: vim/vim#1510948b7d05a4f
Co-authored-by: LemonBoy <thatlemon@gmail.com>
Suggest adding them to news.txt instead.
Also don't ignore changes to intro.txt and sponsor.txt, as they don't
change much these days, and it's necessary to consider whether to
include their changes in Nvim's intro.txt.
This reverts 2875d45e79.
Allowing lintcommit to ignore "fixup" makes it too easy to fixup commits
to be merged on master as the CI won't give any indications that
something is wrong. Contributors can always squash their pull requests
if it annoys them too much.
Problem:
Not using minified version of bootstrap.
Don't need to load normalize with new version of bootstrap.
See https://github.com/neovim/neovim.github.io/pull/350
Solution:
Update link to bootstrap file.
Remove link to normalize.
`vim.health` is not a "plugin" but part of our Lua API and the
documentation should reflect that. This also helps make the
documentation maintenance easier as it is now generated.
Problem: vim-patch commits lack an informative title and summary in the
very first line of the commit message when the vim-revision is a Git SHA
hash, unlike when is a Vim version. This makes it difficult to discern
at a glance what changes are introduced by such vim-patch commits (in
git log, PR title, changelog generated by git-cliff, etc.).
BEFORE:
vim-patch:abcdef123456
runtime(vim): improve performance
<some details>
...
Solution: Repeat the title of the upstream commit message, to improve
the clarity and visibility of the commit message.
AFTER:
vim-patch:abcdef123456: runtime(vim): improve performance
<some details>
...
Note: the `vim-patch:<hash>` token is still needed by `vim-patch.sh`
(but not necessarily in the very first line of the commit message) to
determine which vim patches have been applied. `<hash>` is internally
normalized to 7 hex digits.
Problem:
'modeline' and 'note' are unhandled in the online HTML documentation.
Some (not all) modelines are parsed by the vimdoc parser as a node of
type 'modeline'.
Solution:
- Ignore 'modeline' in HTML rendering.
- Render 'note' text in boldface.
