00:06 <_whitenotifier-f> [YoWASP/nextpnr] whitequark pushed 1 commit to develop [+0/-0/±1] https://git.io/JkCRS

00:06 <_whitenotifier-f> [YoWASP/nextpnr] whitequark f5c03d3 - Update dependencies.

00:34 <_whitenotifier-f> [YoWASP/yosys] whitequark pushed 1 commit to develop [+0/-0/±1] https://git.io/JkCua

00:34 <_whitenotifier-f> [YoWASP/yosys] whitequark acb5cf7 - Update dependencies.

18:15 <awygle> whitequark: k. to be clear, "UserValue only had shape() because it subclassed Value" implicitly ends with "and Value is an ABC with shape as @abstractmethod so UserValue had to fill it in", yes? it's not clear to me whether leaving it to Record to implement shape() would have failed the ABCMeta contract

18:16 <awygle> shape() actually seems like implementing it on ValueCastable might be the right thing to do just because i could see nmigen getting confused if Subclass.shape() and Value.cast(Subclass).shape() don't match, but maybe it's fine

18:17 <awygle> (sorry i sent you a question and then fell asleep for ~18 hours)

18:18 <whitequark> uh, nmigen doesn't care about ValueCastable.shape() at all

18:18 <whitequark> there is no conditions in which it will call that method

18:19 <whitequark> (because ValueCastable is not a Value)

18:20 <awygle> mm, i spose that's true. user code would call it, not nmigen core

18:21 <awygle> well, whatever, shape() in record is the least-disruptive option and we can always move it back into ValueCastable without functional change later if people complain loudly enough about always doing the same implementation of it

18:22 <whitequark> i don't think it should ever be implemented on ValueCastable in nmigen

18:22 <whitequark> in fact, once we ship ValueCastable, the interface will just never change period

18:22 <whitequark> that's *the whole point*

18:23 <awygle> i expect significant pushback on that from users

18:24 <awygle> not that that means we have to change anything ofc

18:24 <whitequark> well, UserValue/ValueCastable is a compromise in first place

18:27 <_whitenotifier-f> [nmigen] awygle opened pull request #545: Implement shape() for Record - https://git.io/JklP4

18:30 <_whitenotifier-f> [nmigen] codecov[bot] commented on pull request #545: Implement shape() for Record - https://git.io/JklP1

18:31 <_whitenotifier-f> [nmigen] codecov[bot] edited a comment on pull request #545: Implement shape() for Record - https://git.io/JklP1

18:31 <_whitenotifier-f> [nmigen] codecov[bot] edited a comment on pull request #545: Implement shape() for Record - https://git.io/JklP1

18:32 <_whitenotifier-f> [nmigen] codecov[bot] edited a comment on pull request #545: Implement shape() for Record - https://git.io/JklP1

19:37 <_whitenotifier-f> [nmigen] whitequark closed pull request #545: Implement shape() for Record - https://git.io/JklP4

19:37 <_whitenotifier-f> [nmigen/nmigen] whitequark pushed 1 commit to master [+0/-0/±2] https://git.io/JklSr

19:37 <_whitenotifier-f> [nmigen/nmigen] awygle c1ed908 - nmigen.hdl.rec: restore Record.shape().

19:37 <_whitenotifier-f> [nmigen] whitequark commented on pull request #545: Implement shape() for Record - https://git.io/JklSo

19:37 <_whitenotifier-f> [nmigen/nmigen] github-actions[bot] pushed 1 commit to gh-pages [+0/-0/±13] https://git.io/JklS1

19:37 <_whitenotifier-f> [nmigen/nmigen] whitequark 0cb5f44 - Deploying to gh-pages from @ c1ed90807be2d4973338b951413f636836a5dfce 🚀