Python script to convert read tables from xls(x) files and output LaTeX tables. Based on excel2latexviapython by Michael Kirker.
To be considered alpha quality! It works for my use case but testing has been very limited, so YMMV. Feel free to make improvements!
Outputs valid LaTeX code for tables from Excel/LibreOffice workbooks (one table per worksheet) while
honoring the following properties in the worksheet:
- colours (untested)
- number format
- alignment (even for individual cells!)
- borders.
Additionally, for multi-line cells needing wrapped text, you can now set the column widths on the
command line (since openpyxl seems to incorrectly report column widths).
Needs (at least) the following LaTeX packages:
longtablearrayxcolorcaptionmultirowmulticolumn
You can use xls2latex.py stand alone to read myfile.xlsx and pipe the output to mytable.tex for inclusion
in a LaTeX document.
/path/to/xls2latex.py -f myfile.xlsx > mytable.tex
Get the possible command line options by executing without arguments xls2latex.py.
A more convenient use, however, is to use call the script from LaTeX by using piped input:
\input{|"xls2latex.py -f myfile.xlsx"}
You can do this either from your LaTeX workflow or from your pandoc workflow. This requires activation
of shell command execution (off by default for security reasons), in the pandoc case e.g.:
pandoc --filter=pandoc-citeproc --filter=pandoc-crossref --filter=pandoc-xnos \
--pdf-engine=xelatex --pdf-engine-opt=-shell-escape \
-i testdoc.md -o testdoc.pdf
where testdoc.md contains the above \input command.
No user-friendly method yet, just copy/clone.
I used a poetry environment, so you could do the same to handle dependencies.
The main reason for developing this script is to get around the severe limitation of markdown tables in terms of formatting and to automate inclusion of an effective workflow to author tables. (Let's face it, making tables is still by far the easiest using office-like software.)
But, if, like me, you replaced your documentation workflow to make e.g. letters, reports, quotations, etc. from well-known office software to markdown, you may be faced with the challenge of including these external tables.
Since for this workflow my output format is still the "traditional" pdf, it makes sense to use LaTeX as an intermediate format.
Enter Pandoc.
If you want to leverage the markdown - pandoc LaTeX - pdf workflow to the fullest, it is advisable to fine-tune the workflow
in such a way, that you only need to configure each document type once (customised templates). The objective being that you can
write any document in markdown completely without leaving your editor, and generate a pdf with a generic one-line command.
Enter pandocomatic.
After setting up pandoc and pandocomatic (and your custom templates, which are called from the pandocomatic config file
depending on which document type you choose in your document's YAML header), you can use one single invocation of
pandocomatic with your md file to automatically generate the desired output.