skeinfox vs. powerpc osx solution

1 view
Skip to first unread message

jet

unread,
Nov 9, 2009, 6:57:50 PM11/9/09
to skei...@googlegroups.com
It looks like python on powerpc/10.5.8 is not compatible with
skeinforge. I just downloaded python 2.6.4 for OSX from python.org,
installed it in the default location (/usr/local) and updated
skeinforge.pl to call /usr/local/bin/python. After that, SkeinFox
parsed my STLs just fine.

--jet


--
J. E. 'jet' Townsend, IDSA
Design, Fabrication, Hacking
design: www.allartburns.org; hacking: www.flatline.net; HF: KG6ZVQ
PGP: 0xD0D8C2E8 AC9B 0A23 C61A 1B4A 27C5 F799 A681 3C11 D0D8 C2E8

jet

unread,
Nov 9, 2009, 7:04:44 PM11/9/09
to skei...@googlegroups.com
now other things are not working, grumble. will continue to investigate.

Jordan Miller

unread,
Nov 9, 2009, 10:10:53 PM11/9/09
to skei...@googlegroups.com
wow, that's an excellent solution, thanks! what is not working now?

Jordan

jet

unread,
Nov 10, 2009, 12:05:27 PM11/10/09
to skei...@googlegroups.com
Some of my other .stl files stopped being valid after I upgraded python.
skeinforge.py generates the "is being chain exported." message then
exits without any sort of error or output.

I'll look at it later this week, I'd really like to have a PowerPC
solution so I can "retire" this box as a makerbot/file server.

--jet

TeamTeamUSA

unread,
Nov 10, 2009, 6:28:30 PM11/10/09
to SkeinFox
A more future-proof solution might be to create a symbolic link from
where skeinforge expects python to be to where it is on your machine:

sudo ln -s <actual_python> <skeinforge_python>

As far as skeinforge ceasing to work, it's likely caused by a well-
known bug that is triggered by moving/renaming the previous file
processed by skeinforge.

To fix, delete the filename in the ~/.skeinforge/skeinforge.csv file
[typically on the second line].

Go!

=ml=

jet

unread,
Nov 10, 2009, 6:39:35 PM11/10/09
to skei...@googlegroups.com
Yeah, I've run into the missing file problem, but this is a small set of
test files that never change. Basically, I'm trying to put together a
calibration suite of models ranging from simple to complex, so they
should always be there.
Reply all
Reply to author
Forward
0 new messages