github-actions[bot] commented on code in PR #63281:
URL: https://github.com/apache/doris/pull/63281#discussion_r3255969615
##########
fe/fe-core/src/main/java/org/apache/doris/catalog/PartitionKey.java:
##########
@@ -147,11 +148,15 @@ private static Literal getDateTimeLiteral(String value,
Type type) throws Analys
} else if (type.isDatetimeV2()) {
return new DateTimeV2Literal(value);
} else if (type.isTimeStampTz()) {
+ if (hasExplicitTimeZone(value)) {
+ return
Literal.fromLegacyLiteral(DateLiteralUtils.createDateLiteral(
+ normalizeTimestampTzOffset(value), type), type);
Review Comment:
This new explicit-time-zone branch rejects valid TIMESTAMPTZ literals that
use named or lowercase zones. `hasExplicitTimeZone()` returns true for values
like `2024-01-15 12:00:00Asia/Shanghai` or `2024-01-15 12:00:00 utc`, so they
are routed here, but `DateLiteralUtils` only detects uppercase letters as
timezone names and `getTimeZoneSplitPos()` only walks uppercase letters and
`/`. For `Asia/Shanghai` it computes an empty `tzString`, and for lowercase
`utc` it does not strip the zone at all, so table creation fails even though
the normal Nereids TIMESTAMPTZ parser accepts these forms (see the existing
`DateTimeLiteralTest`/`TimestampTzLiteralTest` cases around `Asia/Shanghai` and
`uTc`). Please parse explicit TIMESTAMPTZ partition values with the same
TIMESTAMPTZ literal parser, or extend this path and its tests to cover named
and lowercase zones.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
