add link identifiers into emitted tags #436
Conversation
|
Any feedback on this ? |
|
Sorry about the long radio silence. I acknowledge that your use case is a valid one, but I have my doubts about widespread its use is. Another way you could achieve this goal without having to modify pulldown itself is to use the links source map to read the tag. |
Break html scan on eof
I do not know. However reference labels are internally parsed and saved to do resolution so it not really introducing complexity merely just exposing information that is already there.
Perhaps once #430 is fixed that would be an option. |
|
@marcusklaas I think it is also worth considering issue #423 in this context. And specifically:
It appears to me that it would make a lot more sense to have link / image reference be similar to footnote references. Instead of resolving those and concealing details like reference identifiers. In such design exposing reference identifiers is natural. It also would imply that second pass that I believe is there to do reference resolutions would be unnecessary (at least for use cases that don't require serializing markdown to html). |
Will let us publish a release for pulldown-cmark#438
Bump version for 0.7.1 release
Also updates some deps (bumpalo and quote were on yanked versions).
Update Cargo.lock with version number
We want to say "lets us" here, not "let is us".
Fix typo: incorrect apostrophe.
I have a use-case for it. |
Addresses GitHub issue pulldown-cmark#430.
We would previously close lists when a line wasn't completely empty but didn't have the required indentation. Now lines that only contain whitespace are also correctly identified as empty.
It was previously only present in the *generated* test file.
This makes it easier for new contributors to understand the code by reducing the number of abstractions that need to be learned.
This is only possible because things are now Option
Remove TreePointer in favor of Option<TreeIndex>
Refactor link parsing
|
+1 would be great to have this or provide a way in pulldown-cmark to lazily get this information out of the node when providing a range for a link. |
The update didn't include the table header, it was right at start!
Fix whitespace issue in lists
YAML-style and pluses-delimited metadata blocks are supported, each one in a different `Options` flag. This is a breaking change because a new `Tag` enum variant is added. Fixes: pulldown-cmark#580
…attrs feat!: allow custom headings attributes with optional value. Fixes pulldown-cmark#634
…/bumpalo-3.12.0 dependencies: bump bumpalo from 3.10.0 to 3.12.0
feat!: metadata blocks. Fixes pulldown-cmark#580
Fix parser bug where link label gets broken by ] in code span
Make `html` and `escape` modules optional
…-lists Add task list with a list item's first paragraph
|
I think that enum variants get too much complicated to have unnamed fields instead of named fields, so a new pull request replicating the functionality will be implemented if this is not changed. Thanks. |
fix: strikethrough can be intra-word
Allow dynamically generated header ids and classes
fix: shift right overflow panic in debug mode in rare cases
This is implementation for #434. I do not like the fact that
idis added for both inline and reference links, however different approaches were way more intrusive (and more difficult for newbie to implement). I have proposed multiple options in the issues maybe we can discuss to identify one that would be most appropriate.