Regression: Fix int and float compatibility in union types #559
Conversation
NanoSector
force-pushed
the
bugfix/fix-float-and-int-compatibility-in-union
branch
from
6f7a9cb
to
db1d88d
Compare
NanoSector
changed the title
|
Hi @NanoSector, thanks for the contribution. I don't have rights to merge this PR as it is owned by @socialdeal (thanks for the fix sponsoring btw). This is a GitHub limitation. Could you: a. Temporarily add rights to me on the repository fork so I can rebase it on master and merge it or… Thanks! |
|
@romm Thanks for the heads-up! This is interesting, can you share what GitHub is showing you and why this would differ with a personal repository? We are experimenting with workflows in our organisation. In the meantime I'll manually rebase this PR on master. |
This was broken in commit CuyZ@0479532 as withType may be called in union type context. First calling withValue() to set an int value and then updating the type with withType() does not trigger the float cast, in turn causing a mapping error. Developed by: Social Deal (@socialdeal)
NanoSector
force-pushed
the
bugfix/fix-float-and-int-compatibility-in-union
branch
from
db1d88d
to
9dd83fc
Compare
|
Thanks! FYI, this is a long-known issue from GitHub, external organization developers cannot push on branches owned by this very organization, see https://github.com/orgs/community/discussions/5634 It is irritating when I have to do further development on someone's work, luckily I had no modification to do on your branch, so a rebase on |
NanoSector
deleted the
bugfix/fix-float-and-int-compatibility-in-union
branch
|
I see, thank you, that is really good to know. I'll share this with the rest of the team, this sounds highly impractical... |
This was broken in commit
0479532 as withType may be called in union type context.
First calling withValue() to set an int value and then updating the type with withType() does not trigger the float cast, in turn causing a mapping error.
Developed by: Social Deal (@socialdeal)