pzhdfy commented on code in PR #68988:
URL: https://github.com/apache/doris/pull/68988#discussion_r4321197428
##########
regression-test/framework/src/main/groovy/org/apache/doris/regression/util/ResultUtils.groovy:
##########
@@ -59,8 +64,8 @@ class ResultUtils {
}
static void assertSparkDorisResultEquals(List<List<Object>> sparkRows,
List<List<Object>> dorisRows) {
- List<List<Object>> normalizedSparkRows = normalizeRows(sparkRows)
- List<List<Object>> normalizedDorisRows = normalizeRows(dorisRows)
+ List<List<Object>> normalizedSparkRows =
decodeNestedBinaryTokens(normalizeRows(sparkRows))
Review Comment:
Confirmed — the decode pass ran after normalizeRows, so scalar byte[]
columns (already 0x-hex) were decoded through new String(bytes, UTF_8) too, and
non-UTF-8 bytes collapsed through U+FFFD, making 0x80 and 0x81 compare equal.
decodeHexString now round-trips the decoded string through UTF-8 and returns
the original 0x-hex token when the bytes do not round-trip: equality is
preserved exactly, and any rendering that cannot round-trip fails loudly
instead of silently passing. Framework jar rebuilt; the suite passes with it.
--
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]
