Korean language support on WebAssembly regression #69787
Description
Description
The dotnet/aspnetcore repo is getting consistent build failures when trying to ingest the latest runtime update: dotnet/aspnetcore#41801
The failure (e.g., in this build) comes from WebAssemblyICUShardingTest.cs LoadingApp_KoreanLanguage_Works:
Assert.Equal() Failure
↓ (pos 12)
Expected: 2020. 9. 2. 오전 12:00:00
Actual: 2020. 9. 2. 00:00:00
↑ (pos 12)
This failure never occurs on our main branch and never occurred before this PR, so it seems pretty definite that it's a change in the latest runtime update.
However, is this an intentional change? The new output is more consistent with the output we use in other languages, so maybe this was deliberate. 오전 refers to "morning (AM)", but I don't know whether 오전 12:00:00 or 00:00:00 is the more idiomatic way to express midnight.
Temporary resolution
In case this is deliberate, and to unblock the runtime ingestion, I'm going to change the aspnetcore E2E test assertion to look for the new pattern instead of the old one. Then, if it turns out this is actually a regression, I trust the runtime folks will revert it and then aspnetcore will change its test back. And if it's not a regression you can just close this as by design.
Reproduction Steps
See this build
Expected behavior
I don't know. Asking for clarification.
Actual behavior
See above
Regression?
No response
Known Workarounds
No response
Configuration
No response
Other information
No response