pdfmerge 1.0.0 is now published.
Following the principle of "make it easy", I dropped requirements for Windows and upgraded everything to python 3.
In the process of rebuilding
pdfmerge, I discovered that they type-checker really doesn't like
AttrDict properties because they can be
None, but my functions typically need a specific type. I had to resort to calling
cast() which felt a little gross.
I also realized that while
docopt-style is great for CLI argument parsing, it's less pretty as markdown documentation, so I had to add some markdown into the help text which, again, felt a little gross, but it's not the worst thing in the world. Maybe in some future version I'll use
rich to render the help text so that it gets a nice markdown treatment.