Aggarwal-Raghav commented on code in PR #5909:
URL: https://github.com/apache/hive/pull/5909#discussion_r2173822143
##########
cli/src/java/org/apache/hadoop/hive/cli/CliDriver.java:
##########
@@ -500,6 +500,9 @@ public CommandProcessorResponse
processReader(BufferedReader r) throws IOExcepti
StringBuilder qsb = new StringBuilder();
while ((line = r.readLine()) != null) {
+ // Skipping the proto tables location.
+ if (line.startsWith("LOCATION '_REPLACE_WITH_")) continue;
Review Comment:
Hi @okumin, thanks for the review. Here are few things that I checked before
arriving at this PR:
1. `HiveSchemaTool#replaceLocationForProtoLogTables()` is the one which is
doing the replacement for LOCATION during the schematool initialization at
runtime (cluster setup etc.). But calling that method in qtest (CliDriver) is
"little" troublesome and even after achieving that might not yield expected
results.
2. "Why calling `replaceLocationForProtoLogTables` is troublesome" =>
because there are factors like conf object and adding dependency on tez-api and
hive-beeline modules.
3. "Even after achieving that might not yield expected results" => in
qtests, sysdb tables like `CDS`, `COLUMNS_V2` will have table location
`.../itests/qtest/target/localfs/warehouse/sys.db/<tbl_name>/` but for proto
tables the location will either be picked from config or default location i.e.
`/tmp/` or `/tmp/query_data`.
4. IMO, having a uniform location for all tables for testing purpose is
better.
--
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]
