refactor: refactor code tab split #1369
Conversation
|
Deploy preview for docusaurus-2 ready! Built with commit 1a6897e |
zachbadgett
force-pushed
the
code-tabs-comments-fix
branch
from
a1d7b9d to
1cbbef3
Compare
|
Deploy preview for docusaurus-preview ready! Built with commit a1d7b9d |
|
Deploy preview for docusaurus-preview ready! Built with commit 1a6897e |
There was a problem hiding this comment.
This seems to be a fix that changes the prev implementation to be more bug-less.
I am currently away so wont be able to test this out and review it carefully, but in the meantime do you mind to add comment on several places so that future contributor can understand your code intention ?
Thank you for the attempt, by the way 😊
zachbadgett
force-pushed
the
code-tabs-comments-fix
branch
from
1cbbef3 to
dc3a94e
Compare
|
@endiliey Added comments |
There was a problem hiding this comment.
Sorry for the delayed review.
LGTM. Awesome 🎉. However, there might be some edge cases that we might miss.
Would like to get the logic approved before getting too far into testing.
Use cases involving whitespace still need to be tested.
Are you going to write some unit test ? We can create fixtures to test out splitTabsToTitleAndContent
|
Yeah, I'll create unit tests. |
zachbadgett
force-pushed
the
code-tabs-comments-fix
branch
from
372a1fe to
371805c
Compare
| @@ -9,3 +9,4 @@ website/ | |||
| scripts | |||
| packages/docusaurus-1.x/lib/core/metadata.js | |||
| packages/docusaurus-1.x/lib/core/MetadataBlog.js | |||
| packages/docusaurus-1.x/lib/core/__tests__/split-tab.test.js | |||
There was a problem hiding this comment.
Added because the lint fixer keeps removing @jest-environment jsdom from my test file. I'm sure there's a workaround so this is just temporary.
| it('renders content correctly', () => { | ||
| const node = docPage().getDOMNode(); | ||
| const firstContent = node.querySelector('[id$="-content-7"]').textContent; | ||
| expect("console.log('Hello, world!');\n").toEqual(firstContent); |
There was a problem hiding this comment.
It might be okay to match the generated html from the highlighter.
| }; | ||
| beforeEach(() => { | ||
| props = { | ||
| content: ` |
There was a problem hiding this comment.
can we do content: fs.readFile('/some/path/to/fixtures/doc1.md') instead ?
zachbadgett
force-pushed
the
code-tabs-comments-fix
branch
3 times, most recently
from
20ba2f2 to
4996207
Compare
zachbadgett
force-pushed
the
code-tabs-comments-fix
branch
from
4996207 to
1a6897e
Compare
| const contents = m.split('\n').filter(c => { | ||
| if (!indents) { | ||
| indents = ( | ||
| c.match(/((\t|\s{4})+)<!--DOCUSAURUS_CODE_TABS-->/) || [] |
There was a problem hiding this comment.
Can remove redundant parentheses.
|
I like the fact that you put test on it. So far LGTM, I will test it out again locally tonight. cc @yangshun |
|
I just tried this out. This is awesome. Great job on fixing #1260 🎉 . However, the #1279 will only works if it's written with indentation For example, this does not work
What do you think ? |
|
@endiliey that might be an issue with remarkable. Adding a code block in between the list items on the online demo (https://jonschlinkert.github.io/remarkable/demo/) shows the same result. |
Motivation
Fix #1260 to allow html/xml comments in code tabs.
Unsure if this would be considered a
refactoror afixsince it changes the way the tabs are parsed.Have you read the Contributing Guidelines on pull requests?
Yes
Test Plan
Only tested it locally, ensuring the code tabs on page
/docs/en/doc-markdownstill worked.Would like to get the logic approved before getting too far into testing.
Use cases involving whitespace still need to be tested.