Nerd Font data and extractor

[reeseovine]

I've written an extractor for Nerd Fonts so users can easily insert icons from various icon sets into their projects. Closes #54 (sort of).

How it works

1. Download the cheat sheet page: https://www.nerdfonts.com/cheat-sheet
2. Parse the data with beautifulsoup
3. Generate list of icon/classname pairs
4. Write to file

Concerns

• Contrary to my previous claim, Font Awesome codepoints do appear to be the same in Nerd Fonts and their source font. However some of the other icon sets are moved to avoid conflicts within the font, meaning if someone wants to use the source icon font rather than a Nerd Font this data set may not work for them.
  • Nerd Fonts also don't include any of FA's pro icons, for obvious reasons. But also, NF has ~736 FA icons whereas FA 5 free has 1,609. Might be using an older version.
• Icon fonts aren't Unicode standard (they reside in private use blocks) and may be confusing to people who see them when they specify --files all. I think one way to work around this is to have a non-standard/ directory inside data/ which is not indexed by --files all but it is when specifying a filename directly. This proposed solution can decrease clutter for when others decide to contribute data sets like this one.

[polyzen]

But also, NF has ~736 FA icons whereas FA 5 free has 1,609. Might be using an older version.

ryanoasis/nerd-fonts#221

[fdw]

Looks good so far, thanks.

[fdw]

Looks good, thank you!

[polyzen]

For some reason a large number of the nf-mdi icons appear on the right-hand side of rofi.

[fdw]

I think that is because the code point used for these symbols is reserved/used in a RTL language. I don't think there's a better solution than #62. Or do you have an idea?

[polyzen]

Oh, I see. Perhaps the extractor should automatically apply the left-to-right mark to the affected characters in nerd_fonts.csv.