julen / translate forked from egparedes/translate

Toolkit assisting in the localization of software

This URL has Read+Write access

julen (author)
Sat Nov 22 04:54:00 -0800 2008
commit  e05b7b25a7dc733353b233b66ed4789bc6b78000
tree    2f35d4167f20124a704660e993e241f231ced101
parent  24c466f4605dc7f61b1fc4d02b9a312e1dc81aee
translate / tools / TODO
100644 33 lines (30 sloc) 1.759 kb
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
pomerge
* allow status of input PO to override destination ie if input is not fuzzy
    and destination is fuzzy that is you corrected it then make merged message
    not fuzzy
    - A header is currently skipped. When we can handle header merging
      a header should be merged to indicates last translators info, etc but merging
      should be optional with ignoring being the default.
    - msgblock headers ie #: should remain unchanged. There seems to be a bug
      that adds entries on one line when the output templates have them on
separate lines.
    - Allow merge without a template and thus us the output as a template. Or
      at least don't complain when you don't find the templates.
 
New tools
Some ideas for tools that would be helpful
podiff - normal diff show too much spurious data. We need something that
can ignore the location and just shaw the changes in text. So you
could receive an updated PO file but the diff would not show spurious
changes in the header and could match strings in different location.
The output could then be something like diff -u
glossary_creator - take input text or PO file and extract a) the most
frequent words and b) the most frequent word pairs. So things like
"Proxy Server" are extracted. The idea being that you can create a
glossary for a single PO file or for the whole project like KDE.
Translators get to define important words before beginning to translate
pomigrate - need a python based tool that takes an input file the new
template and outputs the new migrated PO files. Initialising as needed
and merging if possible.
 
pocount
- Fix po class so that it handles KDE style plurals there not in pocount
    - Count word\nword and word<tag>word as 2 words