Small
Tasks related to gathering information and/or turning that information into little scripts.
Convert 1.10 mods into 2.1 mods
Even if the sources for an 1.10 mod exist, it is a cumbersome task to make it load in 2.1. The task is to create a set of shellscripts (or similar) to automate things like:
- Rename files to lowercase
- Fix .lev and .wrf files (lowercase filenames, '\' -> '/', etc)
- ...
This doesn't have to be very comfortable or the ultimate automation solution. Any set of scripts automating even just parts of this process is better than what we currently have.
This also does not include fixing problems in maps, or similar. That can be done after a basic conversion.
Convert .wdg mods into .wz mods
Possibly utilising the above mentioned 1.10-to-2.1 converter, extract .wdg archives and repack them as .wz (zipfile).
This would possibly include mapping filename hashes back to the correct strings. (Lookup tables sound like an idea.)
The article on Converting Old Maps ? may be of help.
RodZilla is said to have at least parts of this done. Maybe if you contact him, he can send you what he has.
Translation statistics BuildStep for Buildbot
Write a BuildStep for Buildbot, which runs "msgfmt --statistics $file" on every .po file listed in po/LINGUAS, and outputs the result in an easily parsable (CSV, XML, ...) form to be displayed by a webpage.
The script (Python, PHP, ...) to create that webpage could also be written, which should display a ranging of languages. Possibly list of bars, seperated into "translated/green", "fuzzy/yellow", "untranslated/red".
Big
Tasks requiring knowledge of the topic and requiring C coding skills.
Pathfinding - Moving other units
Tell units a prefered direction when ordering to "go out of the way" in pathfinding. If there is no free space, the other unit should try to respect that and even make others move into the same direction. This should help to solve the crowded-area problem.
Pathfinding - Threat avoidance
Described here: WarzoneForum:137#p785 Maybe Troman still has the code/patch. Would likely need to be ported to current trunk. Would perhaps have to be an optional feature, if the performance issues cannot be solved.
