Broken Rendering (maybe dependent on mistletoe version?) #73

Closed
opened 2023-07-31 12:34:17 +00:00 by hz · 6 comments
Owner

I'm getting some broken rendering on some of my machines with the same catsoop version and Python version. It looks like a Markdown issue, so maybe something changed in mistletoe that we need to account for?

If that is the issue, maybe we should also specify versions of the packages we need in requirements.txt (but then we need to be careful to keep that file up to date as things are updated).

I'm getting some broken rendering on some of my machines with the same catsoop version and Python version. It looks like a Markdown issue, so maybe something changed in mistletoe that we need to account for? If that is the issue, maybe we should also specify versions of the packages we need in `requirements.txt` (but then we need to be careful to keep that file up to date as things are updated).
Author
Owner

@shensquared, have you noticed any of these kinds of issues in 6.390?

@shensquared, have you noticed any of these kinds of issues in 6.390?
hz added this to the (deleted) milestone 2023-07-31 12:34:31 +00:00
hz changed title from Broken Rendering (maybe dpendent on mistletoe version?) to Broken Rendering (maybe dependent on mistletoe version?) 2023-07-31 12:45:16 +00:00
Author
Owner

Then again, updating mistletoe to 1.1.0 (from 0.9.0) on my other machine did not cause anything to break. And py.mit.edu is running on 1.1.0 as well. So now I'm thinking it's not actually a mistletoe issue... 🤔

Then again, updating `mistletoe` to 1.1.0 (from 0.9.0) on my other machine did not cause anything to break. And py.mit.edu is running on 1.1.0 as well. So now I'm thinking it's not actually a mistletoe issue... 🤔
Contributor

Indeed, mistletoe has caused me (or some TA) compatibility issues on MacOS, with Python 3.9 or 3.10. It was right in a semester beginning and things were too hectic, so I fixed it via a simple rollback

pip3 install mistletoe==0.8.2 --force-reinstall

I did briefly look into it, and recall it was almost surely an isolated mistletoe backwards compatibility issue. It's likely the BaseRenderer.getattr() mentioned in Version 0.9 notes https://github.com/miyuchina/mistletoe/releases (but I didn't get to the bottom of it so not sure)

Also, fwiw, I do find a version-fixed requirement.txt super helpful. But I put it inside the course repo like spring23, not the catsoop source repo. It's easier (less risky) to maintain; --force-reinstall can take care of potential override over upstream catsoop; and I also get the additional benefit of full control over e.g. numpy version controls as well. https://airr.mit.edu/fall23/staff/local_copy actually reflects the most up-to-date of my own personal workflow (and tested just last night on Mac)

Indeed, mistletoe has caused me (or some TA) compatibility issues on MacOS, with Python 3.9 or 3.10. It was right in a semester beginning and things were too hectic, so I fixed it via a simple rollback ``` pip3 install mistletoe==0.8.2 --force-reinstall ``` I did briefly look into it, and recall it was almost surely an isolated mistletoe backwards compatibility issue. It's likely the BaseRenderer.__getattr__() mentioned in Version 0.9 notes https://github.com/miyuchina/mistletoe/releases (but I didn't get to the bottom of it so not sure) Also, fwiw, I do find a version-fixed `requirement.txt` super helpful. But I put it inside the course repo like `spring23`, not the catsoop source repo. It's easier (less risky) to maintain; `--force-reinstall` can take care of potential override over upstream catsoop; and I also get the additional benefit of full control over e.g. `numpy` version controls as well. https://airr.mit.edu/fall23/staff/local_copy actually reflects the most up-to-date of my own personal workflow (and tested just last night on Mac)
Contributor

just found my chat with the TA. the culprit was actually a different/known mistletoe breaking change that seems fixed at the time. dependency rollback was just the easier band-aid.

here's the full context fwiw.

just found my chat with the TA. the culprit was actually a different/known mistletoe breaking change that seems fixed at the time. dependency rollback was just the easier band-aid. here's the full context fwiw.
sss.png
390 KiB
Author
Owner

Ah, OK, thanks. That was actually a separate issue (fixed in fe51a77067 about a year ago 😸) so I think there is something else going on here (though I still haven't been able to figure out quite what's going on).

Ah, OK, thanks. That was actually a separate issue (fixed in fe51a77067df0fc575be8927f3eeb871b189d91e about a year ago 😸) so I think there is something else going on here (though I still haven't been able to figure out quite what's going on).
Author
Owner

My issue is actually looking like it's an issue with 6.101 specifically, so I'm going to close this for now. Sorry for the spam, Shen!

My issue is actually looking like it's an issue with 6.101 specifically, so I'm going to close this for now. Sorry for the spam, Shen!
hz closed this issue 2023-08-03 10:11:14 +00:00
Sign in to join this conversation.
No milestone
No assignees
2 participants
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference
catsoop/catsoop#73
No description provided.