[threeten-dev] ZoneIdPrinterParser #262

[roger riggs]

Hi,

Updated the webrev to use a new ZoneId.ofOffset(prefix, zoneOffset) method.
Corrected some tests that still allowed UT0, UTC0, GMT0 and the prior forms
of ZoneIds.  This change also affects the default parser of ZoneDateTime.
The results do match ZoneId.of.

http://cr.openjdk.java.net/~rriggs/webrev-zoneid-parser-262/

Please review and comment.

Thanks, Roger



On 6/10/2013 11:35 AM, Stephen Colebourne wrote:
> On 8 June 2013 18:34, roger riggs <roger.riggs at oracle.com> wrote:
>> Please review the changes and comment whether the test cases are correct.
>> There are a number of input values that were previously valid for parsing
>> of ZoneIds that are no longer valid (to match ZoneId).  For example,  "UTC0"
>> is no longer valid.
> Looks right to me. It may be worth a cross-test with ZoneId.of to
> ensure both work the same.
>
>> One inconvenience in the code after a prefix and the offset value is parsed
>> there is no public method to create the ZoneId.  The ZoneId.of method as to
>> re-parse the input.
>> The method   static ZoneRegion ofPrefixedOffset(String zoneId, ZoneOffset
>> offset)
>> is use for this purpose but it is package private and can't be used from the
>> format package.
>> Is it reasonable to add a method to ZoneId for a similar purpose?
> I think adding a ZoneId.ofOffset(String prefix, ZoneOffset offset)
> method would be OK. I don't think it needs a longer name than
> "ofOffset" because the arguments and javadoc should be clear enough. I
> think that as well as "UTC, "UT" and "GMT" prefixes, it should accept
> a blank string, which just returns the offset.
>
> thanks
> Stephen
>
>
>> http://cr.openjdk.java.net/~rriggs/webrev-zoneid-parser-262/
>>
>> Thanks, Roger
>>
>>
>>