sqoop-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Venkat Ranganathan" <n....@live.com>
Subject Review Request 24586: Fix native character usage for SQL Server object names
Date Tue, 12 Aug 2014 01:03:57 GMT

This is an automatically generated e-mail. To reply, visit:

Review request for Sqoop.

Bugs: SQOOP-1429

Repository: sqoop-trunk


Fixed the SQLServer connection manager get column details query and get primary key query
to properly prefix the string literals with N'...' prefix as the information_schema uses native
character set for table names

Also fixed the Native splitter to use N'...' prefix for the strings generated to be consistent


  src/java/org/apache/sqoop/manager/SQLServerManager.java 625f7cd 
  src/java/org/apache/sqoop/mapreduce/db/DataDrivenDBInputFormat.java 6b30820 
  src/java/org/apache/sqoop/mapreduce/db/NTextSplitter.java PRE-CREATION 
  src/java/org/apache/sqoop/mapreduce/db/TextSplitter.java 6c512aa 
  src/test/org/apache/sqoop/mapreduce/db/TestTextSplitter.java e3d0c0c 

Diff: https://reviews.apache.org/r/24586/diff/


Tested with SQL Server with table names using native chars (characters not in the default
collation) and also tried using import-all-tables which exercises to import all tables from
a database schema with many tables with names containing characters from non-default collation


Venkat Ranganathan

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message