PDF export html validation
I run the html to generate the PDF on the html validator, and i found a couple of issues.
Most of them are ok, but with paged.js i think we should be only taking valid html to make sure a bug is not coming from there.
-
Thee
section.content
includes a full html page withhtml
,head
andbody
element. It should only include the content. On export, we should only keep the innerHTML of the body element and throw the rest away -
author-list ul
inside anauthors
paragraph is prohibited. -
element
footnote
not allowed as a child of p (and pretty sure not allowed at all). footnote should be replaced by a span with a footnote class. (should be done in wax imho) -
multiple mathjax with the same ID, or nested svg with the same ID. I guess that’s because each math is done before pagedjs actually runs: I think we definitely want to use some kind of pagedjs plugin to render the math instead of making before end, but there may be good reasons for not doing that.
-
Warning: Text run is not in Unicode Normalization Form C. it’s a warning but im curious how this can break the pdf rendering
Ω(<em>x))
: Ω(x) -
link can’t have empty target or empty rel
-
img doesnt need a
/
to be a self closing element in the html world (but it needs in xml right?)