Bugs in libdmtx
-----------------------------------------------------------------

1. libdtmx - Core Library
::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::

While regular encoder works fine, the optimizer feature (--best)
still occasionally generates codeword sequences that are not 100%
legal according to the ISO specification.  Everything still
appears to decode properly, but until I have time to go through
every corner case and validate the behavior this will be treated
as an experimental feature.  For now dmtxwrite will encode using
a straight ASCII scheme by default.

Data Mosaic encoding doesn't produce output for certain sizes:

   $ echo -n foo | dmtxwrite -M    <-- works
   $ echo -n fooo | dmtxwrite -M   <-- doesn't work
   $ echo -n foooo | dmtxwrite -M  <-- works


2. Test Programs
::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::

multi_test:
  * No known issues (not included in general download)

rotate_test:
  * No known issues (not included in general download)

simple_test:
  * No known issues

unit_test:
  * Missing files   (not included in general download)


3. Scripts in the script directory
::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::

The check_headers.pl script verifies that every function has
a correctly-formed header comment.  But the test condition is
currently pretty simple, and does not test the first function
appearing in each file.