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
TextFormat shouldn't get a ton of use in end user code, but it is very useful in write test cases, configurations, etc. So having better error for when the input is invalid can be pretty useful. Our current values have no context, but for some of them there are likely some pretty easy additions/wins:
.unrecognizedEnumValue getting the enum's full name and the string value
.conflictingOneOf getting the full message name, the already set field, and the field trying to be set.
Also a good number of the errors could actually also use line number (or atleast offset) into the string buffer being parsed to help someone correct their input.
The text was updated successfully, but these errors were encountered:
main has gotten the infrastructure for a new error model that won't be api breaking, so new errors can have more details. However moving existing errors to that new model was counted as api breaking. So I'm closing this off as we won't actually evolve TextFormatDecodingError , but at some future point with 2.0 work, the existing errors could be given more context like this.
TextFormat shouldn't get a ton of use in end user code, but it is very useful in write test cases, configurations, etc. So having better error for when the input is invalid can be pretty useful. Our current values have no context, but for some of them there are likely some pretty easy additions/wins:
.unrecognizedEnumValue
getting the enum's full name and the string value.conflictingOneOf
getting the full message name, the already set field, and the field trying to be set.Also a good number of the errors could actually also use line number (or atleast offset) into the string buffer being parsed to help someone correct their input.
The text was updated successfully, but these errors were encountered: