Bugs translating Unicode Vulgar Fractions

[jkoshy]

% ebookmaker --version
EbookMaker 0.9.7

Processing VULGAR FRACTION ONE HALF

File: fractions.rst ::

.. meta::
   :DC.Title: Unicode Fractions 1

*   ½ (U+00BD, VULGAR FRACTION ONE HALF)

Run ebookmaker:

% ebookmaker --output-dir=/tmp/PG --make txt fractions.rst
% cat /tmp/PG/Unicode\ Fractions\ 1.txt 

  - A1/2 (U+00BD, VULGAR FRACTION ONE HALF)

Please note the extra A generated in front of the 1/2.

Processing VULGAR FRACTION FIVE EIGHTHS

File: fractions.rst:

.. meta::
   :DC.Title: Unicode Fractions 2

*   ⅝ (U+215D, VULGAR FRACTION FIVE EIGHTHS)
*   ½ (U+00BD, VULGAR FRACTION ONE HALF)

Run ebookmaker:

% ebookmaker --output-dir=/tmp/PG --make txt fractions.rst
ERROR:root:unitame: unhandled chars: {~VULGAR FRACTION FIVE EIGHTHS U+215d~}
ERROR:root:unitame: unhandled chars: {~VULGAR FRACTION FIVE EIGHTHS U+215d~}
% cat /tmp/PG/Unicode\ Fractions\ 2.txt

  - {~VULGAR FRACTION FIVE EIGHTHS U+215d~} (U+215D, VULGAR FRACTION
    FIVE EIGHTHS)
  - 1/2 (U+00BD, VULGAR FRACTION ONE HALF)

Apart from the ⅝ character not being handled, the extra A in front of the 1/2 is no longer generated.

[eshellman]

you might look at Unitame.py and UnitameData.py to see where the ascii-ifier is going wrong. We might also consider just not "taming" the Unicode.