search

unified-font-object / ufoNormalizer

A tool that will normalize the XML and other data inside of a UFO.
Other
51 stars 19 forks source link

Release on PyPI? Integrate into ufoLib? #42

Closed madig closed 6 years ago

madig commented 6 years ago

It would be nice to have this on PyPI, or even better, inside ufoLib and turned on automatically or even a mandatory part of the spec. It would make requirements.txt look nicer for font development repositories.

anthrotype commented 6 years ago

I'll do that

anthrotype commented 6 years ago

@typesupply I'm preparing the package to be published and it looks like we have no explicit LICENSE file.

Would it be ok to use the same as ufoLib, i.e. BSD-3-Clause?

And what should the copyright notice be?

Copyright (c) 2017, Tal Leming
anthrotype commented 6 years ago

@typesupply let me know if this looks fine to you, thanks

https://github.com/unified-font-object/ufoNormalizer/blob/master/LICENSE.txt

anthrotype commented 6 years ago

ok, everything should be ready for a release, just let me know and I'll push a new git tag and Travis will do the rest

typesupply commented 6 years ago

Would it be ok to use the same as ufoLib, i.e. BSD-3-Clause?

Sure. That's fine.

And what should the copyright notice be? Copyright (c) 2017, Tal Leming

Eh, I guess. I'm not particularly possessive of it.

anthrotype commented 6 years ago

ok done! you can pip install ufonormalizer now

https://github.com/unified-font-object/ufoNormalizer/releases/tag/0.3.2 https://pypi.python.org/pypi/ufonormalizer

typesupply commented 6 years ago

Regarding incorporating this into ufoLib... ufoLib seems to be going through some pretty massive changes so I'm reluctant to let it go over there at this time. The dust needs to settle. Maybe later.

Regarding making this a default behavior in ufoLib... I wouldn't do this because the normalizer is not necessarily something everyone needs. In fact, I wrote this a few years ago and only had a need for it a week ago. It's great for files going into a repository but an unnecessary step for most uses. This particular implementation is also inefficient because it's a post-processor, not something that happens during the ufoLib write. ufoLib's output could be written to follow the normalization recommendation, but that would be different from incorporating this version as is.

Regarding making it part of the spec... We discussed this a good bit, but it complicates things pretty deeply. The discussions are out there, but one of the big reasons I remember is that not all XML toolkits allow customizable whitespace formatting.

madig commented 6 years ago

ufoLib's output could be written to follow the normalization recommendation, but that would be different from incorporating this version as is.

I'd be content with that ;)

And thanks Cosimo :)