nbconvert does not manage to take the final step to transform SVGs into PDFs and hence breaks figure inclusion in the final pdflatex call to generate PDF files using, for example, the --format="latex" and --format="sphinx_howto" options.
This is despite the technology being ready and waiting in IPython/nbconvert/transformers/svg2pdf.py. (Requires inkscape to do the conversion.)
Crucial for 1.0.
By the way, I am on Mac OS but installed inkscape with brew, so the logic to choose the Mac OS version of inkscape will fail in my case. It should check if the command line inkscape is not already available.
I can confirm this on linux as well. The SVGs get embedded like "ordinary" images without any conversion.
Maybe @jdfreder can bring some light into this issue?
Thanks guys, I'll check this out right now
I've got this just about figured out - the main issue is that SVG2PDF is before extract figures in the transformer list, so it doesn't do anything.
That could be a problem... haha
Also, another possibility is the data priority (in the nbconvert base class). It may need to be overwriten for the Latex case, svg excluded.
Lastly, the templates themselves may current have 'svg' instead of 'pdf'
Should be fixed by #3695.