Intent to archive

[simoncozens]

This repository contains a lot of tools which were once used in the Noto production process; however, the process has changed and simplified over time, and the only parts of the repo which are now used (by the Noto team) are noto_lint.py and test_vertical_extents.py.

noto_lint.py should really be replaced by the Noto profile in fontbakery; test_vertical_extents.py is a handy script which could find a home in gftools.

Having the other tools around is misleading, as it implies they are used by Noto when they aren't. However, if there are tools out there which people are using, I'd like to hear about it, so we can think how those uses can be best supported. If I don't hear anything, I'm planning to convert two tools we do use and then the archive this repo by the end of April.

[rsheeter]

I believe some parts of this are actively used by https://github.com/googlefonts/noto-emoji and therefore we should not deprecate. Pruning away long-unused parts is fine, maybe we can rescope to that?

[simoncozens]

Thanks - this is the sort of thing I was hoping to hear!

It looks like noto-emoji uses three parts of nototools: tool_utils, add_vs_cmap.py, and unicode_data. The first two could be moved into the noto-emoji repo.

unicode_data is useful enough to be spun off to its own module; another option is that it gets merged with youseedee - which automatically downloads and caches the latest version of UCD from the Unicode web site, rather than bundling it.

[davelab6]

Moving noto-emoji specific parts to noto-emoji sounds good to me.

Youseedee also sounds good to me, if you add an Apache license and the other typical G-OSPO stuff and transfer to me then I will transfer to /googlefonts and you can take it from there?

[rsheeter]

I'd prefer we start by just pruning. Please send any PRs to me and @anthrotype .

[rsheeter]

Wrt unicode data, @anthrotype maintains https://github.com/fonttools/unicodedata2. If we want to invest in improving a unicode data library I'd prefer we look to that.

[davelab6]

I defer to Rod on these choices :)

[anthrotype]

(technically I am off until next week - just wanted to point out unicodedata2 is simply a backport of the latest cpython unicodedata and as such it is bound by the upstream)

[simoncozens]

Yeah, youseedee and unicode_data both provide a much richer interface to the whole of the UCD, not just the limited bits that the Python API exposes. unicodedata2 would need a lot of work to come up to the same level.

[marekjez86]

@simoncozens : please add me to PRs

[simoncozens]

@rsheeter @marekjez86 @anthrotype please see

#637
#638
#639
#640
#641

(for starters)

[rsheeter]

Merged the first three, defer to @marekjez86 on #640, approved #641 but did not merge as it appeared to be reporting anger from CI.

Lets keep unicode_data here for now.