[vim/vim] Matchfuzzy: Prefer complete matches over separator matches (#7140)
Christian Brabandt
this is a fix for the recently discussed issue of matchfuzzy at vim-dev: https://groups.google.com/forum/#!topic/vim_dev/QQrHWIwI8jM
What this does is, give sequential matches a higher priority than separator matches, so that
echo matchfuzzy(['.vim/v_i_m_r_c', '.vim/vimrc', '.vim/vimrc_colors'], 'vimrc')
will return matches in the order of ['.vim/vimrc', '.vim/vimrc_colors', '.vim/v_i_m_r_c'].
For now this PR still contains the old changes to give an extra bonus for a omplete match. But this is commented out. I just left it in, just in case if we agree that this would be a good idea in general :)
If all agree that this is not needed, I'll remove that part and push up and merge everything into one single commit.
You can view, comment on, or merge this pull request online at:
https://github.com/vim/vim/pull/7140
Commit Summary
- Make matchfuzzy prefer complete matches over separator matches
- matchfuzzy Fix tests and prefer complete matches
File Changes
- M src/search.c (12)
- M src/testdir/test_matchfuzzy.vim (10)
Patch Links:
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or unsubscribe.
Christian Brabandt
those github actions failures are unrelated. Not sure why it fails in tests, that should run cleanly and which I haven't touched
Yegappan Lakshmanan
@yegappan commented on this pull request.
In src/search.c:
> @@ -4358,6 +4364,10 @@ fuzzy_match_compute_score(
score += FIRST_LETTER_BONUS;
}
}
+ #if 0
Hi Christian,
Can you add a comment here about why this code is commented out?
Thanks,
Yegappan
Yegappan Lakshmanan
@yegappan commented on this pull request.
In src/search.c:
> @@ -4358,6 +4364,10 @@ fuzzy_match_compute_score( score += FIRST_LETTER_BONUS; } } + #if 0Hi Christian,
Can you add a comment here about why this code is commented out?
vim-dev ML
> ------------------------------
>
> In src/search.c
> <https://github.com/vim/vim/pull/7140#discussion_r506517165>:
>
> > @@ -4358,6 +4364,10 @@ fuzzy_match_compute_score(
> score += FIRST_LETTER_BONUS;
> }
> }
> + #if 0
>
> Hi Christian,
>
> Can you add a comment here about why this code is commented out?
>
>
>
BTW, the test in FreeBSD (Cirrus/CI) is failing.
- Yegappan
Bram Moolenaar
I think we can just drop ENDMATCH_SEQUENTIAL_BONUS and the "fullMatch" variable. At least for now.
—
You are receiving this because you commented.
codecov[bot]
Codecov Report
Merging #7140 into master will increase coverage by
0.02%.
The diff coverage isn/a.
@@ Coverage Diff @@ ## master #7140 +/- ## ========================================== + Coverage 88.71% 88.73% +0.02% ========================================== Files 148 148 Lines 162018 162113 +95 ========================================== + Hits 143732 143851 +119 + Misses 18286 18262 -24
| Impacted Files | Coverage Δ | |
|---|---|---|
| src/search.c | 92.51% <ø> (ø) |
|
| src/vim9compile.c | 93.03% <0.00%> (-0.11%) |
⬇️ |
| src/list.c | 95.02% <0.00%> (ø) |
|
| src/version.c | 92.13% <0.00%> (ø) |
|
| src/terminal.c | 91.00% <0.00%> (ø) |
|
| src/evalfunc.c | 96.28% <0.00%> (+<0.01%) |
⬆️ |
| src/eval.c | 96.77% <0.00%> (+<0.01%) |
⬆️ |
| src/vim9execute.c | 93.87% <0.00%> (+<0.01%) |
⬆️ |
| src/evalvars.c | 95.74% <0.00%> (+0.01%) |
⬆️ |
| src/userfunc.c | 93.52% <0.00%> (+0.01%) |
⬆️ |
| ... and 12 more |
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact),ø = not affected,? = missing data
Powered by Codecov. Last update 8d739de...ffbdce2. Read the comment docs.
—
You are receiving this because you commented.
Yegappan Lakshmanan
@yegappan approved this pull request.
—
You are receiving this because you commented.