chrisjsewell/ipypublish

sphinx html output has all figures filled in with the same single png

Closed this issue · 3 comments

parmentelat commented

Bug Report

Describe the bug

I have a rather long notebook based on the calypso-bash kernel (not that it should matter)

when I run this:

nbpublish -f sphinx_ipypublish_main.run -o /Users/tparment/git/flotpython-gittutorial/notebooks/exporting exporting/ALL.ipynb

the output html exhibit a very odd behaviour, which is that all the markdown cells that include an image with s/t like ![](media/imagename.png), well all these figures show the same image, which is indeed one of the ones that is inserted, but it should not be the same one over and over again.

To Reproduce

Steps to reproduce the behavior:

  1. step 1
  2. step 2

(for extended explanation, use the additional context section)

Minimal Notebook Example

If possible, please create a minimal Jupyter Notebook to replicate the issue.
Then rename its file extension to .txt (since Github rejects .ipynb),
and drag and drop it here.

Expected Behaviour

A clear and concise description of what you expected to happen.

Runtime Information

(please complete the following information)

  • IPyPublish: [e.g. 0.8.3]

  • Python: [e.g. 3.6.1]

  • OS: [e.g. OSX 10.5.1]

  • Pandoc: [run pandoc --version]

  • (optional for pdf issues) texlive: [run tex]

  • (optional for pdf issues) latexmk: [run latexmk --version]

  • Versions of any other relevant python packages

Additional context

Add any other context about the problem here.

Logging

Drag and drop the log file here, or copy and paste the output from the terminal

chrisjsewell commented

@parmentelat good spot :) This should be fixed via #89, but feel free to reopen if you still have any issues. I'll release a new version shortly

parmentelat commented

thanks for fixing it :)
do you have any estimate for that release ? I'd rather wait for it than install from sources, if that's likely to happen shortly; what I do in these cases is to issue a dev release so that others can test by installing with pip install --pre, maybe this scheme could work here too

thanks !

chrisjsewell commented

Already released v0.10.4 yesterday :) FYI, if you didn't know, ipypublish is also available on conda-forge: https://anaconda.org/conda-forge/ipypublish, which is the recommended way of installing