Thougts on Skellington and submission
Well due to my own ineptitude (and being separated from my linux box so having to do it the windows way) I messed up the md5 sum for my team's final entry and so it was (temporarily) rejected. Thankfully I'd uploaded a less polished version earlier in the day, so all would not have been lost if I had not been allowed to correct my error.
From what I've been reading though I'm not the only one who had a bit of difficulty creating a correct md5. It got me thinking: Skellington is a great package and the create-upload script made it really easy to package our game; it would, I suspect, be fairly trivial to have that script optionally (but on by default) produce md5 sum along with the zip file. I think that could save some stress when it comes down to the wire.(log in to comment)
Comments
Traceback (most recent call last): File "run_game.py", line 16, in main.main() File "/home/james/src/misc/pyweek6/BotBuilder2000-1.1/lib/main.py", line 41, in main stateManager = StateManager() File "/home/james/src/misc/pyweek6/BotBuilder2000-1.1/lib/main.py", line 62, in __init__ self.credits = Credits() File "/home/james/src/misc/pyweek6/BotBuilder2000-1.1/lib/credits.py", line 20, in __init__ self.textList.append(pygletfont.Text(self.font, line, x = 800 / 2, y = sy, halign = pygletfont.Text.CENTER)) File "/usr/lib/python2.5/site-packages/pyglet/font/__init__.py", line 357, in __init__ self.halign = halign File "/usr/lib/python2.5/site-packages/pyglet/font/__init__.py", line 504, in _set_halign self._layout.halign = halign File "/usr/lib/python2.5/site-packages/pyglet/text/layout.py", line 1409, in _set_halign self._update() File "/usr/lib/python2.5/site-packages/pyglet/text/layout.py", line 836, in _update left = self._get_left() File "/usr/lib/python2.5/site-packages/pyglet/text/layout.py", line 854, in _get_left return self._x - width // 2 TypeError: unsupported operand type(s) for //: 'NoneType' and 'int'I tried with both pyglet 1.0 and pyglet 1.1 installed.
saluk on 2008/04/06 07:16:
Yeah, I think the main reason many md5's didn't work was we used the code on the help page and opened the files in text mode rather than binary mode - producing improper hashes on windows machines. If the md5 code were build into the create script, this would be a simple matter. I think Richard mentioned looking into doing this, I'm sure it will be added by next time.