<i18n dev> RFR: 8276220: Reduce excessive allocations in DateTimeFormatter [v6]

[Stephen Colebourne]

On Wed, 3 Nov 2021 13:14:42 GMT, Claes Redestad <redestad at openjdk.org> wrote:

>> Prompted by a request from Volkan Yazıcı I took a look at why the java.time formatters are less efficient for some common patterns than custom formatters in apache-commons and log4j. This patch reduces the gap, without having looked at the third party implementations. 
>> 
>> When printing times:
>> - Avoid turning integral values into `String`s before appending them to the buffer 
>> - Specialize `appendFraction` for `NANO_OF_SECOND` to avoid use of `BigDecimal`
>> 
>> This means a speed-up and reduction in allocations when formatting almost any date or time pattern, and especially so when including sub-second parts (`S-SSSSSSSSS`).
>> 
>> Much of the remaining overhead can be traced to the need to create a `DateTimePrintContext` and adjusting `Instant`s into a `ZonedDateTime` internally. We could likely also win performance by specializing some common patterns.
>> 
>> Testing: tier1-3
>
> Claes Redestad has updated the pull request incrementally with one additional commit since the last revision:
> 
>   Add test verifying OOB values throw exception

Thanks for adding the new tests, and finding that fraction formatting is more constrained than I thought it was.

I think there is a case for a spec update in `DateTimeFormatterBuilder` to clarify the constraint on the input value (at this point, that seems better than changing the behaviour of fraction formatting). As things stand, the exception that is thrown is not described by the spec. (The spec change could be part of this PR or a separate one).

-------------

Marked as reviewed by scolebourne (Author).

PR: https://git.openjdk.java.net/jdk/pull/6188