#development
Title
# development
h
happy-kitchen-89482
08/09/2019, 5:34 PMRandom question: has there been any thought given to using standard Python3.7 dataclasses and type annotations instead of our custom
datatypea
aloof-angle-91616
08/09/2019, 5:34 PMyes!
we should do it and it is planned! let me find the ticket
https://github.com/pantsbuild/pants/issues/7074 and https://github.com/pantsbuild/pants/issues/7354 are the relevant tickets
w
witty-crayon-22786
08/09/2019, 5:35 PMi think that we might still end up with some customizations to avoid mutability and prevent overriding
eqa
aloof-angle-91616
08/09/2019, 5:36 PMi was trying to start moving towards that when i made
DatatypeMixinChoicesMixinh
happy-kitchen-89482
08/09/2019, 5:37 PMYou can mark dataclasses as
frozen=Trueš 1
Which emulates immutability to a good enough degree
š 1
w
witty-crayon-22786
08/09/2019, 6:34 PMhow do you enforce that they are marked? and/or make that the default.
(anywho: it feels to me like swapping from extending
namedtupleh
hundreds-father-404
08/09/2019, 6:45 PMYes Iām hoping to see this. For enforcing that people make it frozen, I think code review should catch most cases
w
witty-crayon-22786
08/09/2019, 6:46 PMWell, that's why superclass. To avoid allowing them to do the wrong
h
hundreds-father-404
08/09/2019, 6:46 PMWe can superclass dataclass then. Sgtm
h
happy-kitchen-89482
08/09/2019, 6:48 PMPresumably you can check that they're marked when validating the product graph
extending
dataclassš 1
a
aloof-angle-91616
08/09/2019, 6:48 PMnot necessarily
or at least, pants will accept exactly one type, be it a
datatypedataclassw
witty-crayon-22786
08/09/2019, 7:00 PMAnd there is nothing enforcing use of datatype with the engine (...yet?). From a sandbox perspective that may eventually be a thing, but it isn't now
a
aloof-angle-91616
08/09/2019, 7:00 PMretweet