github-actions[bot] commented on code in PR #69591:
URL: https://github.com/apache/doris/pull/69591#discussion_r4374322962
##########
regression-test/suites/datatype_p0/uuid/test_uuid_csv_schema.groovy:
##########
@@ -19,27 +19,17 @@
suite("test_uuid_csv_schema", "p0,external") {
sql "SET enable_sql_cache = false"
sql "SET enable_query_cache = false"
- String localPath = context.config.otherConfigs.get('uuidLocalExportPath')
String token = UUID.randomUUID().toString()
def sources = [:]
for (String kind : ['values', 'invalid']) {
File fixture = new File(context.dataPath,
"test_data/uuid_external_${kind}.csv")
- if (localPath != null) {
- File target = new File("${localPath}_${token}_${kind}.csv")
- java.nio.file.Files.copy(fixture.toPath(), target.toPath())
- String path =
"${context.config.otherConfigs.get('uuidLocalTvfPath')}_${token}_${kind}.csv"
- sources[kind] = """LOCAL('file_path'='${path}',
- 'backend_id'='${(sql 'SHOW
BACKENDS')[0][0]}','format'='csv',
- 'column_separator'=',','csv_schema'='id:int;u:uuid')"""
- } else {
- String bucket = context.config.otherConfigs.get('s3BucketName')
- String key = "uuid_csv_schema/${token}/${kind}.csv"
- getS3Client().putObject(bucket, key, fixture)
- sources[kind] = """S3('uri'='s3://${bucket}/${key}','format'='csv',
- 'column_separator'=',','csv_schema'='id:int;u:uuid',
-
's3.endpoint'='${getS3Endpoint()}','s3.region'='${getS3Region()}',
-
's3.access_key'='${getS3AK()}','s3.secret_key'='${getS3SK()}')"""
- }
+ String bucket = context.config.otherConfigs.get('s3BucketName')
+ String key = "uuid_csv_schema/${token}/${kind}.csv"
+ getS3Client().putObject(bucket, key, fixture)
Review Comment:
[P2] Preserve Azure uploads in the migrated UUID fixture suites
For `s3Source="azure"` with the default Azure endpoint and previously valid
`uuidLocalExportPath`/`uuidLocalTvfPath` mappings, removing the local branch
now forces this AWS upload. `Config.fillDefaultConfig()` supplies
`s3Endpoint="azure-endpoint"`, while `Suite.getS3Client()` always builds an
`AmazonS3Client` using that raw endpoint, so `putObject` fails before the
schema/value assertions. The same newly forced path exists in
`test_uuid_parquet_native.groovy:32` and
`test_uuid_parquet_predicate.groovy:29`. These are separate UUID fixture
loaders from the two ORC loaders addressed in the existing Azure thread;
without the local overrides, the old AWS fallback was already broken and is not
the new regression here. Apply provider-aware upload and matching TVF
provider/endpoint properties to these loaders as well.
--
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]
