Minted filter for beamer and latex, safe for all writers #40
Conversation
There was a problem hiding this comment.
These are the things I'm a little concerned about, but overall I think this works pretty well!
If you would rather not have such a complicated filter that's OK too. But I think this would be a really helpful filter for anybody who wants to change the listings for beamer / tex 🙂
|
Stephen McDowell <notifications@github.com> writes:
+ -- NOTE: using "latex" here because using FORMAT as a parameter does
+ -- not seem to work. It should not cause problems for beamer.
+ return pandoc.RawBlock("latex", raw_minted)
Sorry, this one occurs in two places and I missed both in the last review. Why can't I do `return pandoc.RawBlock(FORMAT, raw_minted)`? If I do that then nothing shows up (aka the code blocks just disappear).
That's a bit of a rough edge in pandoc. "beamer"
doesn't currently work as a raw format. I've just
pushed a commit fixing that, but for now use "latex".
|
|
Ok the proposed final documentation is rendered here. I just need to add some tests later tonight and then it should be good (I'll add a new comment and remove When we think this is ready for merge, is it ok if I squash these all into one commit? The commit messages aren't very good, and the diff stack is unnecessary (I had some back and forth ...). I hesitate to do this because when I do I think GitHub removes the review comments with the questions I have above. But if they aren't actually problems, then I'll force push the badness away :) |
|
Great! Feel free to squash, but we can also squash this when merging (we generally do this for all PRs introducing new filters). |
|
ok i think this looks good 🙂 i hope the tests being written in python is ok, i had to do a lot of string manipulation and that's definitely my go-to language for that. grepping for the latex was annoying because latex syntax definitely conflicts with regex syntax. sorry for multiple force pushes :D but now it's just one commit, unless things need to change in which case future commits will be structured better and can just be squash merged on your end. |
svenevs
changed the title
- elem.attributes preferred to elem.attr[3] - use `kind` as parameter name, overloading `type` makes debugging more difficult
update tests to make sure that `fragile` is not present in non-beamer output
make sure tests assert that `nil` is not present to ensure that metadata keys are not being processed incorrectly
svenevs
changed the title
|
Ok round two complete, thanks for the really helpful review on making this much cleaner @tarleb! All of your feedback has been incorporated and a few updates to the tests to actually validate the IIRC when squashed the title of the PR is what the main commit message is. I don't know what the main message should be, but feel free to edit it to be whatever you want. |
|
Thanks again! This is probably the best documented and tested filter in the collection. |
|
Woohoo, you're very welcome! Hehe yes ... I got a little carried a way xD |
|
Am impressed by how well this filter is written and documented! Dedication to detail at its best :-)! I hope this is the best place to leave this info for others: For the ArXiv preprints workflow, there are some limitations on using minted. In essence, it is important to use a non-hidden cache directory, like so
The important thing is that the Full details are here: https://arxiv.org/help/faq/mistakes#minted Disclaimer: I did not test this yet, but thought it is worth logging it for the future. |
Hey! I made a new years resolution to get this all nice and shiny and in
lua, and I think it's looking pretty good so far.