Conversation
|
The latest updates on your projects. Learn more about Vercel for Git ↗︎
|
There was a problem hiding this comment.
Pull Request Overview
This PR addresses issue #356 by refactoring and organizing regex and Unist query utilities into separate, focused modules. The changes improve code organization by extracting regex patterns into a reusable regex module and creating dedicated Unist query functions.
Key Changes
- Extracted all regex patterns from
queries/index.mjsinto a newqueries/regex.mjsmodule for better reusability - Created a new
queries/unist.mjsmodule containing node type checking functions that were previously inline - Updated all import statements across the codebase to reference the new modular structure
Reviewed Changes
Copilot reviewed 14 out of 14 changed files in this pull request and generated 2 comments.
Show a summary per file
| File | Description |
|---|---|
src/utils/queries/regex.mjs |
New module containing all regex patterns with comprehensive documentation |
src/utils/queries/unist.mjs |
New module containing Unist node type checking functions |
src/utils/queries/index.mjs |
Refactored to import and use patterns from new modules, removed inline definitions |
src/utils/parser/index.mjs |
Updated import to use YAML_INNER_CONTENT from regex module |
src/utils/parser/constants.mjs |
Updated to import heading regex patterns from regex module |
| Multiple generator and linter files | Updated imports to use functions from the new unist module |
avivkeller
force-pushed
the
organize-regex
branch
from
2693be0 to
6bc3f1e
Compare
AugustinMauroy
left a comment
AugustinMauroy
left a comment
There was a problem hiding this comment.
LGTM ! but could you provide a before after for the issue
After
Before
|
|
Thanks 👍 |
|
Bump @nodejs/web-infra, this is a bug fix |
ovflowd
left a comment
ovflowd
left a comment
There was a problem hiding this comment.
Bump @nodejs/web-infra, this is a bug fix
How is such a huge PR a bug fix? This seems a refactor per PR-description. What are you fixing?
| } from '../parser/index.mjs'; | ||
| import { getRemark } from '../remark.mjs'; | ||
| import { transformNodesToString } from '../unist.mjs'; | ||
| import { |
There was a problem hiding this comment.
Cleaner in this situation, these are a bunch on non prefixed constants, doing a wildcard import here better explains where they belong to.
|
What I fail to understand is why you arbitrarily decided to redesign the structure of the whole queries. Please revert these changes and only include your bug fix. If you want to separate certain things, that's fine, but at least give an explanation why. Arbitrary refactors aren't something I support. |
This PR refactors our RegExp queries, and, in the process, fixes #356, I should not have classified this as a "bug fix", it's a change that also fixes a bug.
I figured, while I'm adding a RegExp file, why not move the query RegExps there as well, and at that point, also move the Unist functions. But I'm happy to revert, since that was quite the tangent. |
I do insist. Not against the refactor, but make a dedicated PR just for the refactor, without the bug fix. This allows us to discuss the bug fix on this PR and the refactor on another :) |
|
Blocked by #379. TODO Add test cases |
avivkeller
force-pushed
the
organize-regex
branch
from
ee35904 to
67904ed
Compare
avivkeller
force-pushed
the
organize-regex
branch
from
67904ed to
2c75bcf
Compare
Codecov Report❌ Patch coverage is Additional details and impacted files@@ Coverage Diff @@
## main #366 +/- ##
==========================================
+ Coverage 74.27% 76.27% +1.99%
==========================================
Files 118 120 +2
Lines 11041 11906 +865
Branches 695 718 +23
==========================================
+ Hits 8201 9081 +880
+ Misses 2837 2822 -15
Partials 3 3 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
|
The RegExps have 100% test coverage, with around 600 lines of tests (but I'm open to adding more). Ideally, these RegExps will almost never change, lest we break something |
| exports.identifiers.push(value.name); | ||
|
|
||
| if (CONSTRUCTOR_EXPRESSION.test(value.name[0])) { | ||
| if (/^[A-Z]/.test(value.name)) { |
There was a problem hiding this comment.
Let's keep regexes in constants for easily finding regexes in one place
There was a problem hiding this comment.
Even for these, single-character ones? Or, are you referring to all the RegExps?
| @@ -0,0 +1,296 @@ | |||
| // ============================================================================ | |||
There was a problem hiding this comment.
Instead of doing this, why not different files? These section blocks aren't a common thing in JavaScript.
|
I think my main issue with this PR is... What's the motivation behind it? Why is this refactoring needed? What's the problems with the current approach? |
|
Okay, perhaps refactoring the RegExps wasn't really needed. I'm closing this. |
Oh! I didn't even see this comment, I suppose we reached the same conclusion :) |
4 participants
The queries and RegExp's we use should be extensively tested to prevent anything slipping through. This PR moves all of our RegExps and queries into their own file, and adds unit tests. This way, we can re-use the same testing structure across our RegExps, to prevent de-syncs.