There were a couple of reports of "Buffer X newer than edits" problems.
We first assumed that it is incorrect for a server to send 0 as a
version - and stated that they should send a `null` instead, given that
in the specification the `textDocument` of a `TextDocumentEdit` is a
`OptionalVersionedTextDocumentIdentifier`.
But it turns out that this was a change in 3.16, and in 3.15 and earlier
versions of the specification it was a `VersionedTextDocumentIdentifier`
and language servers didn't have a better option than sending `0` if
they don't keep track of the version numbers.
So this changes the version check to always accept `0` values.
See
- https://github.com/neovim/neovim/issues/12970
- https://github.com/neovim/neovim/issues/14256
- https://github.com/haskell/haskell-language-server/pull/1727
The current guidance for install Python packages is to use
python -m pip install <package_name>
Instead of
pip install <package_name>
This ensures that one is using the version of pip that is tied to the
environment's interpreter (and, thusly, its packages). This has [been
endorsed by a core
maintainer](https://snarky.ca/why-you-should-use-python-m-pip/) as
being the recommended way to invoke pip.
As there currently are a few places where the old invocation was used,
attempt to bring them in line.
Fixes#14234
[skip ci]
Changes from original include:
- "See |terminal-window|" -> "See |terminal|".
- Remove mention of using CTRL-W window commands in insert mode.
- Converted usage example to use the Nvim job and channel API.
- Removed logging from usage example, as ch_logfile() has no direct Nvim
counterpart.
- Fixed some small grammar/spelling mistakes.
I build Vim and Nvim from source and have both repos cloned locally. To prevent
vim-patch.sh from downloading the Vim source once again to /.vim-src/, I usually
use a symlink pointing to the already existing Vim repo.
The pattern "/.vim-src" works for both, a directory or a symlink.
It looks solid with the default `FloatBorder` group.
If you set the bgcolor of FloatBorder to the same color as for FloatNormal, you
effectively get an "1-cell padding".
Because borders add up to 2 to the height of a float, we need to subtract that
from the anchor position, when opening a float in the lower half of the window.
Problem: No test for Ex shift commands.
Solution: Add a test. (Dominique Pelle, closesvim/vim#4801)
d5e3cc11d3
Comment out some tests because of Neovim inccomand.
Problem: Search test leaves file behind.
Solution: Delete the file. Also use Check commands.
b68df220c5
Reorder tests based on test_search.vim in patch v8.2.0409.
Problem: Expression type is used inconsistently.
Solution: Add "ETYPE_IS" and "ETYPE_ISNOT" as separate enum values. Rename
"TYPE_" to "ETYPE_" to avoid confusion.
07a3db89b8
Problem: Extending a list with itself can give wrong result.
Solution: Remember the item before where the insertion happens and skip to
after the already inserted items. (closesvim/vim#1112)
dcae51facc
Originated from Neovim commit 7ceebacb3f.
Problem: List test doesn't fail.
Solution: Adjust the test for NULL list handling.
f57497276b
Comment out test cases that modify null lists
because Neovim throws error messages instead of silently failing.
Null lists should be read-only and constant.
https://github.com/neovim/neovim/issues/4615
Replacement for Vim's test_null_string().
Vim uses it to verify that its codebase handles null strings.
Preparation for the Test_null_list() in patch v8.2.1822.
Use v:_null_string, not non-existent env var, for null string tests.
Mention v:_null_string in id() because id(v:_null_string) returns (nil).
Problem: Typo in test comment. (Christ van Willegen)
Solution: Avoid mixing up a data structure with a body part.
f7b398c6a9
Cherry-pick Test_dict_lock_operator() from patch v8.2.0619.
