[WIP] Add message parsing #2
Conversation
The old RawMessage implementation effectively brute forced the initial processing of a comment message by breaking it down into lines, and grouping them into paragraphs. This is useful, but, we actually only need the first paragraph, the last paragraph, and then everything between. So there's no need to break down the message into each paragraph. In theory, the Buffer implementation is more performant than RawMessage was, but most importantly I think it will be easier to work with it.
There was a problem hiding this comment.
Well, this seems to be what happens when I can't sleep. I end up messing around with the GitHub iPadOS app and review one of my own PRs as a form of note keeping about what to fix, cause getting my laptop out and just doing the fixes is more effort than I'm willing to commit to at 1am... lol
| // | ||
| // It is intentionally VERY forgiving so as to be able to extract the various | ||
| // parts even when things aren't quite right. | ||
| var HeaderToken = regexp.MustCompile( |
There was a problem hiding this comment.
I think it would be better to expose helper functions for checking if a string/[]byte matches a header/footer token, and also functions that extract the component parts in a standardized way.
Otherwise the capture groups of these regexp patterns effectively become part of the package's public API, making it difficult to change. Hence this public interface should be provided as helper functions instead, meaning the underlying implementation can be safely fixed and/or improved without breaking backwards compatibility.
jimeh
force-pushed
the
add-message-parsing
branch
from
2b7fe45
to
3fc20eb
Compare
jimeh
force-pushed
the
add-message-parsing
branch
from
3fc20eb
to
bf44c4a
Compare
|
Code Climate has analyzed commit bf44c4a and detected 0 issues on this pull request. The test coverage on the diff in this pull request is 100.0% (90% is the threshold). This pull request will bring the total coverage in the repository to 100.0% (0.0% change). View more on Code Climate. |
This is nearly a complete rewrite of #1 from last weekend :P
I opted for a slightly different approach to breaking down and accessing data out of the a commit message, which I believe is a bit more performant, and yields less and simpler code further down the parsing chain.
Main message parsing does seem to work, but more test cases are needed, and some cleanup too I'm sure.