How to install
--------------

0. The sources have been developed and tested under Python 2.5
   and ReportLab 2.1 and 2.2.
   If you are using an older version of Python or ReportLab,
   it will probably not work without some fiddling, since it uses
   at least the boolean constants True and False.
   For users of RL 1.19: Remember that you may have to change your
   code for RL 2.1 (paragraph encoding should be UTF8).

1. Download the file wordaxe-0.3.2.zip from the SourceForge web site
   http://sourceforge.net/project/showfiles.php?group_id=105867

2. Unzip the files in the ZIP archive to the root directory
   (inside the archive, everything is in a directory "wordaxe-0.3.2".
   This will create the following structure:
   C:\
     wordaxe-0.3.2
       htdocs
         css
         examples
         icons
         images
       wordaxe
         dict
         rl
   
   The files have been developed under MS Windows. Perhaps you need
   to convert line-endings on other platforms.
   
3. Run setup.py install

   Alternatively, if your ReportLab version is older than 2.3, you'll have to
   make a backup of the file reportlab\pdfbase\rl_codecs.py
   from your reportlab installation, then overwrite it with the version 
   found in c:\wordaxe-0.3.2\wordaxe\rl\rl_codecs.py.
   Note: This will change only 2 lines that effect the handling of
   the "shy" character.
   
4. If you didn't use setup.py install, then you have to manually add wordaxe
   to your python path, for example by creating a corresponding file 
   wordaxe.pth in your lib/site-packages directory, containing the text
   c:\wordaxe-0.3.2
   Make sure that "import wordaxe" does not produce an error message.
   
5. Unless you have used the "shy" character (0xAD) in your text,
   ReportLab will work exactly like before.
   
6. To see the decomposition of compound words for german language in action,
   navigate to the directory "test" and run the script "test_hyphenation.py".
   It will create two PDF files, test_hyphenation-plain.pdf and test_hyphenation_styled.pdf.
   
7. To see how to add hyphenation support for german text to your own scripts,
   take a look at test_hyphenation.py. 
   Basically, you just add these 4 lines to your script
     from wordaxe import hyphRegistry,DCWHyphenator
     hyphRegistry['DE'] = DCWHyphenator('DE',5)
     cellStyle.language = 'DE'
     cellStyle.hyphenation = True
   and you replace
     from reportlab.platypus.paragraph import Paragraph
     from reportlab.lib.styles import ParagraphStyle, getSampleStyleSheet
   with
     from wordaxe.rl.paragraph import Paragraph
     from wordaxe.rl.styles import ParagraphStyle, getSampleStyleSheet
     

Remarks:

The pyHnj hyphenation algorithm hung on my private computer (on XP)
when using umlauts, whereas it worked on another computer (on Win2000).
If you want to try it, just use
hyphRegistry['DE'] = PyHnjHyphenator('de_DE')

As a workaround, I added a pure Python version that does not hang on umlauts;
you can use it by adding the keyword argument purePython=True in the constructor.

The pattern files for the pyHnj hyphenation algorithm (*.dic) 
are from Open Office (http://www.openoffice.org).
As far as I know, they can be freely distributed.
For details of the Open Office license please visit the Open Office web site.

You can run DCWHyphenator.py -g -f testfile.txt
This will create a file DCWLearn.html.
Unknown words will be marked red in there.
Take a look at this file in your browser and see how hyphenation works 
when you resize your browser window.

You can add your own word stems, prefixes and suffixes 
to the file "DE_hyph.ini", it is far from perfect!
The format of this file will probably change (rename the properties etc.)
An entry can have additional attributes (see at the top of DCWHyphenator.py).