search

lbryio / lbrycrd

The blockchain that provides the digital content namespace for the LBRY protocol
https://lbry.com
MIT License
2.57k stars 178 forks source link

Normalization - Do we care about UCA versions #255

Closed tiger5226 closed 5 years ago

tiger5226 commented 5 years ago

I just wanted to get this conversation out of slack and have documented for historical purposes.

The unicode standards are always updated

screen shot 2019-02-13 at 12 10 28 pm

LBRYcrd is use the ICU library for unicode support v57.1. This uses UCA 8.0

screen shot 2019-02-13 at 12 43 27 am

Among our different services, there are 3 that standout in requiring unicode support also.

LBRYNet - The base lbry uses UCA 11.0 Chainquery(MySQL) - MySQL 8.0 users UCA 9.0 Lighthouse(ElasticSearch) - as of yet unknown

@lyoshenka Has already made the call that for Chainquery we will wait and see if there will be any issues between UCA 8.0 and UCA 9.0.

The list of migration issues are below

https://www.unicode.org/versions/Unicode9.0.0/#Migration https://www.unicode.org/versions/Unicode10.0.0/#Migration https://www.unicode.org/versions/Unicode11.0.0/#Migration

So the question is do we care? If not lets close this issue.

BrannonKing commented 5 years ago

After upstream merge, update Boost and ICU versions to match Unicode 11. In fact, can we ditch the lbrycrd_dependencies repo and use the "depends" folder instead?

BrannonKing commented 5 years ago

Upstream merge now includes ICU 62.

lyoshenka commented 5 years ago

doing some cleaning. reopen if this is still active.

BrannonKing commented 5 years ago

To clarify one thing: I'm planning to ship the next release with ICU 63.2 (which is still Unicode 11).