On Wednesday, March 22, 2017 at 4:32:10 AM UTC-7, Daniel Krenn wrote:
>
> In a Jupyter notebook, when I do
> File --> Download as --> PDF via LaTeX
> then typeset formulas (displayed via show(...) command) are not typeset
> anymore, but displayed as their plain text output in the pdf.
>
> Is this a problem of our SageMath-LaTeX typesetting or a problem of the
> jupyter notebook?
>
I think it's the latter, although perhaps one should blame how we supply
our info. By the time you're selecting that option, I think you're just
running jupyter-nbconvert, which just works with the ipynb file; there's no
kernel/state to query otherwise.
For instance, "show(x^2)" leads to this json fragment:
"outputs": [
{
"data": {
"text/html": [
"<html><script type=\"math/tex;
mode=display\">\\newcommand{\\Bold}[1]{\\mathbf{#1}}x^{2}</script></html>"
],
"text/plain": [
"x^2"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"show(x^2)"
]
As you can see, the output is available in two representations: text/plain
and text/html. It doesn't surprise me that ipynb-to-latex conversion
doesn't try to untangle HTML.
The %%latex special inserts a "text/latex" mimetype entry. This does get
rendered properly in html in my browser and it makes export->PDF produce an
error because "pandoc" is not installed. So this seems to be a route that
may make cells more latex-friendly. I don't know what the protocol is
exactly, but I suspect our kernel can respond with various mimetypes when
reporting back output. Using "text/latex" rather than "text/html" might
lead to better results sometimes?
--
You received this message because you are subscribed to the Google Groups
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.
