Firstly: thanks @eteq for the script - very useful when my Authorea PDF build stopped working, and I also wanted a full .tex version and control over the final PDF (including Bibtex citations) for submission versions and such.
It's still mostly working in August 2022, although I did make a few modifications (albeit not very well), which might be useful for you or others.
Specifically, there were two key additions/mods that are possibly required for the current Authorea file structure, although may also be due to the way my document was put together.
- Deal with HTML files for figure captions (basic implementation with tag-stripping via BeautifulSoup only), see function
get_input_string_HTML(). In my case, I only had caption.html for figures. This would probably be best done with pypandoc (or similar), but I couldn't work out how to deal with custom in-line maths tags.
- Added read figure config.yml files to
get_figure_string() to determine correct figure from fig subdirs when multiple versions present. In my case, some figure dirs had multiple versions when, e.g., a figure had been replaced, or was originally an interactive version. There are likely other useful things in there too, but that was all I needed.
See https://github.com/phockett/Extracting-Molecular-Frame-Photoionization-Dynamics-from-Experimental-Data/tree/submission050922/authorea_scripts for full details and notes.
Some of this could easily be back-ported, although I definitely broke some of the existing functionality by messing about with the paths too much, and in a rush, to get a quick-and-dirty build for my needs that I could push to Overleaf. But, if there is interest, let me know and I can do a more careful job to pull into this repo.
Firstly: thanks @eteq for the script - very useful when my Authorea PDF build stopped working, and I also wanted a full .tex version and control over the final PDF (including Bibtex citations) for submission versions and such.
It's still mostly working in August 2022, although I did make a few modifications (albeit not very well), which might be useful for you or others.
Specifically, there were two key additions/mods that are possibly required for the current Authorea file structure, although may also be due to the way my document was put together.
get_input_string_HTML(). In my case, I only hadcaption.htmlfor figures. This would probably be best done with pypandoc (or similar), but I couldn't work out how to deal with custom in-line maths tags.get_figure_string()to determine correct figure from fig subdirs when multiple versions present. In my case, some figure dirs had multiple versions when, e.g., a figure had been replaced, or was originally an interactive version. There are likely other useful things in there too, but that was all I needed.See https://github.com/phockett/Extracting-Molecular-Frame-Photoionization-Dynamics-from-Experimental-Data/tree/submission050922/authorea_scripts for full details and notes.
Some of this could easily be back-ported, although I definitely broke some of the existing functionality by messing about with the paths too much, and in a rush, to get a quick-and-dirty build for my needs that I could push to Overleaf. But, if there is interest, let me know and I can do a more careful job to pull into this repo.