Deyan Ginev
Deyan Ginev
Hi @bfirsh the reason I haven't followed up further here is that I'm not quite comfortable using the testing harness for engrafo. And I was seeing at least one strange...
Experimenting with raw package interpretation is something Bruce and I have been meaning to do for a while. We got around to trying it out recently, for the arXiv sandbox...
If we obtain a sandbox of arxiv papers that are of high interest to arxiv-vanity, I can set that up as a standalone corpus and do the comparison - but...
Interesting experiment. and it sounds about right. The lack of styling reminds me of the reason binding macro definitions are allowed to be `locked` by latexml. Exactly to protect redefinitions...
Right, the `--includestyles` runs are both less predictable and more prone to errors at the moment, so you may want to be extra conservative in using the switch. The "fallback...
Well, not really... That .sty file may be simple, but a different paper may have a local style file that completely destroys the conversion with a Fatal error, arxiv is...
There is a tricky thing that latexml does to tex annotations that are too long - it introduces a line-break for readability with a tex-style `%\n` comment+break separator. The first...
I'll close here, since #48 is enough for tracking the problem.
Thank you for the report! This is another case showing the need for tcolorbox.sty support in latexml.
Thank you for the report! The submission bundle for this article is exercising to the maximum the file-detection logic and its emulation of arXiv's AutoTeX. I would need to revisit...