Skip to content

Latest commit

 

History

History
52 lines (37 loc) · 1.41 KB

README.md

File metadata and controls

52 lines (37 loc) · 1.41 KB

Build Status

hadoop2-csv

Input format for hadoop able to read multiline CSVs

Run BasicTest.java to see it working. Check src/test/resource/test.csv to see a multiline demofile.

The key returned is the file position where the line starts and the value is a List with the column values

Zip files are supported.

More ideas to improve this are welcome.

Example:

If we read this CSV (note that line 2 is multiline):

Joe Demo,"2 Demo Street,
Demoville,
Australia. 2615",[email protected]
Jim Sample,"3 Sample Street, Sampleville, Australia. 2615",[email protected]
Jack Example,"1 Example Street, Exampleville, Australia.
2615",[email protected]

The output is as follows:

==> TestMapper
==> key=0
==> val[0] = Joe Demo
==> val[1] = 2 Demo Street, 
Demoville, 
Australia. 261
==> val[2] = [email protected]

==> TestMapper
==> key=73
==> val[0] = Jim Sample
==> val[1] = 
==> val[2] = [email protected]

==> TestMapper
==> key=10
==> val[0] = Jack Example
==> val[1] = 1 Example Street, Exampleville, Australia. 261
==> val[2] = [email protected]

License

https://www.apache.org/licenses/LICENSE-2.0.html

Credits

Personal fork of CSVInputFormat, but built against hadoop2. Please report the issues to the original fork.