GitLab

The UniqueUnion.jar file can run from the bin directory.  Double-clicked
it there and it found the source files, generated new data, and
deposited in the 'data/2014-04-17' directory.

UniqueUnion now finds the 'source' files in the root, reads as UTF-8,
converts from U-label to A-label, and then generates the merged file.
All data is deposited in the 'data' directory and labeled for the day of
creation.

Duplicate entries 

Adding a Java program to merge, sort, normalize, and make unique.

- Moving UniqueUnion.jar into the bin directory.
- Created a data/A-label directory and added A-label conversions for
each section's file.
- Created a data/U-label directory but left this empty for now.  Files
at the root could be moved here, or some other action could be taken.
- Added a final directory to store files with a unique union of Reserved
Names.  Added the unique union of A-labels to this directory.

Updated the toLowerCaseAscii routine to fix a bug.
Also changed the handling of comment characters to prevent duplicates in
certain situations.

As part of normalization, I was using Java's String.toLowerCase()
method.  This changes utf-8 values, corrupting the encoding.

Adding a Java program to merge, sort, normalize, and make unique, all the .txt files in a particular directory.

You need the Net_IDNA2 script from PEAR.