Did you use prettier (like in superlinter) to find problematic/unparsable html?

Did you use prettier (like in superlinter) to find problematic/unparsable html?

For markdown, I used:

npx prettier --print-width 80 -w ': npx prettier --print-width 80 -w README.md

on the various README.md files but not on HTML.

In the grass-addons, I used prettier and fixed the html parsing errors a while ago, but did not apply the formatting fixes, as it was way too disruptive. But many places the hrefs without quotes were flagged, or tags not in order.

I tried with npx prettier on the four HTML files and the result is pretty ugly (broken </pre> tags and more). So I did not submit these changes.

I tried with npx prettier on the four HTML files and the result is pretty ugly (broken </pre> tags and more). So I did not submit these changes.

I made a quick run to see if there were some parsing errors that might be problematic when converting. Take a look at the 5 files that have red lines in https://github.com/echoix/grass/actions/runs/12001379426/job/33451802353?pr=303#step:4:8310

One is an decimal character entity missing its semicolon;
one is a link that the closing tag doesn't have the slash so the closing </em> is unexpected;
One is a figure not having a closing italic tag, so the closing div is unexpected;
The two others it's hard to understand with the lines shown, the context isn't big enough.

One is an decimal character entity missing its semicolon; one is a link that the closing tag doesn't have the slash so the closing \</em\> is unexpected; One is a figure not having a closing italic tag, so the closing div is unexpected;

This test looks good and useful to me: so it is informative yet it does not modify the entire HTML code.

The identified HTML errors might be fixed along with the submission of a new superlinter HTML linting PR (in case you plan that).

The two others it's hard to understand with the lines shown, the context isn't big enough.

Concerning one of the linting errors:
t.rast.extract.html: here the < in the example code section will need to be converted to < as it seems to confuse superlinter as is.

(other HTML errors addressed in #4757)

One is an decimal character entity missing its semicolon; one is a link that the closing tag doesn't have the slash so the closing \</em\> is unexpected; One is a figure not having a closing italic tag, so the closing div is unexpected;

This test looks good and useful to me: so it is informative yet it does not modify the entire HTML code.

Problem is: they do. That's what all the warnings (in gray) are. I didn't find a way to get what I wanted for grass-addons, so I ended up disabling.

The identified HTML errors might be fixed along with the submission of a new superlinter HTML linting PR (in case you plan that).

Not for now. I'm hoping they get replaced by MD.

The two others it's hard to understand with the lines shown, the context isn't big enough.

Concerning one of the linting errors:

t.rast.extract.html: here the < in the example code section will need to be converted to < as it seems to confuse superlinter as is.

:)