You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When deserializing a locale Jackson currently uses the underscore character as the separator rather than the dash. Specifically, in FromStringDeserializer.java line 234:
Given the RFC states that only the characters a-z A-Z and - are valid it should be possible to leave the current code in for backward-compatibility but it should also check for '-' as a separator.
The text was updated successfully, but these errors were encountered:
mcdee
changed the title
Deserializing locale uses incorrect separator
Deserializing locale uses non-standard separator
Aug 23, 2016
@mcdee Thank you for reporting this: sounds like a bug indeed.
Do you know of an existing JDK provided Locale value that would exhibit this? Would be great to have a unit test against regression, beyond fixing the issue itself.
Actually I think JDK always uses underscore, so I'll have to just test deserializer with made-up codes.
cowtowncoder
changed the title
Deserializing locale uses non-standard separator
Deserializing locale assumes JDK separator (underscore), does not accept RFC specified (hyphen)
Aug 25, 2016
When deserializing a locale Jackson currently uses the underscore character as the separator rather than the dash. Specifically, in FromStringDeserializer.java line 234:
Many locale implementations use dash as the separator as per https://tools.ietf.org/html/rfc5646
Given the RFC states that only the characters a-z A-Z and - are valid it should be possible to leave the current code in for backward-compatibility but it should also check for '-' as a separator.
The text was updated successfully, but these errors were encountered: