Hello there--new TextMate user, enjoying it a fair bit, save for one major problem: it seems unable to use the Python installed from Fink. Whenever PyMate runs, it always uses Python 2.3.5 installed in /usr/bin, as opposed to Python 2.4.2 in /sw/bin.
"/usr/bin/env python" calls /sw/bin/python; /sw/bin is listed before /usr/bin in my PATH.
My suspicion is that there's some preference somewhere in TextMate that tells it the path to Python. However, I've been unable to find it by browsing through Preferences.
Can anyone give me a hint as to how to get TextMate to call the proper Python?
Thanks for any help you can give. :)
Can anyone give me a hint as to how to get TextMate to call the proper Python?
you should define the TM_PYTHON environment variable within textmate and point it to your python setup. I recommend to use the pythonw executable, otherwise some scripts cannot run.
TM_PYTHON=/sw/bin/pythonw
ciao,
domenico
Domenico Carbotta wrote:
Can anyone give me a hint as to how to get TextMate to call the proper Python?
you should define the TM_PYTHON environment variable within textmate and point it to your python setup. I recommend to use the pythonw executable, otherwise some scripts cannot run.
TM_PYTHON=/sw/bin/pythonw
Actually, you should probably get the [universal python][1], instead of the one from Fink. It is most up-to-date, most optimized, and has the most testing of external packages, etc. (several of which are available from that same page). If you do get it, then AFAIK there is no reason to use pythonw, as it is a symlink to python.
-Jacob
Sorry for a slightly off topic post to this list, but it is relevant in the grand scheme of things...
On Aug 19, 2006, at 2:28 PM, Jacob Rus wrote:
Actually, you should probably get the [universal python][1], instead of the one from Fink. It is most up-to-date, most optimized, and has the most testing of external packages, etc. (several of which are available from that same page). If you do get it, then AFAIK there is no reason to use pythonw, as it is a symlink to python.
Does that MacPython package still Seg Fault when navigating through a "prefixed" command history?
For instance, say you're in the python shell and you type:
matrix =
then hit up-arrow to get commands from your history that started with that "matrix =" prefix. When I would find one, then hit down arrow I'd get a seg fault. Maybe it happened when I didn't find a completion ... perhaps both, sorry, it's been a while.
After some googling, I found others had problems with it in their compiled pythons and the culprit was the readline package -- which needed to be patched. Installing python + readline via darwinports has worked out well for me so far.
Just curious if this is still (ever?) happening to those people using the MacPython package?
-steve
Steve Lianoglou wrote:
Does that MacPython package still Seg Fault when navigating through a "prefixed" command history?
For instance, say you're in the python shell and you type:
matrix =
then hit up-arrow to get commands from your history that started with that "matrix =" prefix. When I would find one, then hit down arrow I'd get a seg fault. Maybe it happened when I didn't find a completion ... perhaps both, sorry, it's been a while.
I can't duplicate this, so it may have been fixed since you had problems with it. I usually use iPython when I need an interactive shell, however, and have never had this problem.
-Jacob
On Aug 19, 2006, at 3:26 PM, Jacob Rus wrote:
Steve Lianoglou wrote:
Does that MacPython package still Seg Fault when navigating through a "prefixed" command history? For instance, say you're in the python shell and you type: matrix = then hit up-arrow to get commands from your history that started with that "matrix =" prefix. When I would find one, then hit down arrow I'd get a seg fault. Maybe it happened when I didn't find a completion ... perhaps both, sorry, it's been a while.
I can't duplicate this, so it may have been fixed since you had problems with it. I usually use iPython when I need an interactive shell, however, and have never had this problem.
Hmm .. interesting ... I actually only noticed it when I started to use ipython since that's when I really started using the interactive shell in any serious way.
We even had a recent email to the ipython list about this very issue (it causes a Bus Error, not a Seg Fault, sorry).
I'll try installing it again to see what's cooking.
Thanks for taking the time to check it out. -steve
-
Domenico Carbotta -
Jacob Rus -
Robert J. Hansen -
Steve Lianoglou