Dimitris Tsirogiannis has posted comments on this change.

Change subject: IMPALA-1903: Add support for partitioning by TIMESTAMP
......................................................................


Patch Set 6:

(5 comments)

http://gerrit.cloudera.org:8080/#/c/1621/6/be/src/runtime/timestamp-parse-util.h
File be/src/runtime/timestamp-parse-util.h:

Line 359: Parse
Sorry, I know I asked you to change the name, but I think something like 
IsValidTimestamp() or something along those lines seems more appropriate. What 
do you think?


http://gerrit.cloudera.org:8080/#/c/1621/6/fe/src/main/java/com/cloudera/impala/analysis/TimestampLiteral.java
File fe/src/main/java/com/cloudera/impala/analysis/TimestampLiteral.java:

Line 29: public class TimestampLiteral
I think it deserves a class comment.


Line 47: public String toSqlImpl() {
       :     return "'" + value_ + "'";
       :   }
single line


Line 59: public String getStringValue() {
       :     return value_;
       :   }
single line


http://gerrit.cloudera.org:8080/#/c/1621/6/fe/src/main/java/com/cloudera/impala/planner/HdfsPartitionPruner.java
File fe/src/main/java/com/cloudera/impala/planner/HdfsPartitionPruner.java:

Line 211: !(child.isLiteral()) || child.getType().isTimestamp()
I don't think this condition is correct. For example, "a IN (timestamp_col)" 
expr, where a is a partition column cannot be evaluated in the FE, but that 
condition will return false. We should also have a test that covers that case.


-- 
To view, visit http://gerrit.cloudera.org:8080/1621
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: Icad7dcdc1b199cce9483dc414072bbe24efd625c
Gerrit-PatchSet: 6
Gerrit-Project: Impala
Gerrit-Branch: cdh5-trunk
Gerrit-Owner: Jim Apple <[email protected]>
Gerrit-Reviewer: Dimitris Tsirogiannis <[email protected]>
Gerrit-Reviewer: Jim Apple <[email protected]>
Gerrit-HasComments: Yes

Reply via email to