[reportlab-users] Python 3.x progress

Peter Cock p.j.a.cock at googlemail.com
Thu Jan 9 09:22:30 EST 2014


On Thu, Jan 9, 2014 at 1:34 PM, Robin Becker <robin at reportlab.com> wrote:

> Peter,

>

> I'm a bit confused here about which version of python you're using.


Two different Pythons, which give different behaviour.

Self-compiled Python 3.3,

$ which python3
/Users/peterjc/bin/python3
$ python3 --version
Python 3.3.0

Apple provided Python 2.7,

$ which python
/usr/bin/python
$ python --version
Python 2.7.5



> The error in your previous email

>

>

>> $ python3

>> Python 3.3.0 (default, Sep 29 2012, 19:41:44)

>> [GCC 4.2.1 Compatible Apple Clang 4.0 ((tags/Apple/clang-421.0.60))] on

>> darwin

>> Type "help", "copyright", "credits" or "license" for more information.

>>>>>

>>>>> import reportlab

>>>>> reportlab.__version__

>>

>> Segmentation fault: 11

>

> is certainly very odd. This should only involve python and python scripts;

> according to Andy he's run builds on the mac with some success with both 2.7

> and 3.3


OK, that's informative - I wasn't sure if you'd tried this under Mac OS X.


>> The second problem is curious - a TypeError under Python 3.3

>> (self compiled) but a segmentation fault in _rl_accel.so under

>> Python 2.7 (provided by Apple):

>>

>> $ python3 test_GraphicsChromosome.py

>> test_add_count (__main__.ChromosomeCountTest)

>

> ......

> can you abstract the problem to something simple that I can test elsewhere?

> Andy & Ricardo both have macs that could be used. The _rl_accel.c code was

> changed significantly to make things work with both 3.3 & 2.7 so I would

> like to get at any seg faults in there asap.


Try https://gist.github.com/peterjc/8334631 which extracts the
failing Biopython unit test as a short script. I'm working on
further simplifying it - currently in addition to ReportLab it
requires Biopython (using Biopython 1.63 should be fine).

I've not yet tried this under Linux, but could do so as well...

Peter


More information about the reportlab-users mailing list