[postgis-users] GDAL problems (Kyngesbury GDAL)
Dheeraj Chand
dheeraj at dheerajchand.com
Sat Dec 28 19:53:36 PST 2013
Let me step back and ask a broader question, then. What do you think I can do?
Also, I thought that since the GDAL installer writes the libraries to my system Python, perhaps cloning a virtualenv off of that would result in a VE that has GDAL already installed and usable. However, what I found in the term made me curious.
REINHEIT:lib dheerajchand$ cd /Library/Python/2.7/site-packages/
README shapely-py2.7.pth
gdal-py2.7.pth sqlite-py2.7.pth
numpy-override/ vboxapi/
numpy-override.pth vboxapi-1.0-py2.7.egg-info
REINHEIT:lib dheerajchand$ cd /Library/Python/2.7/site-packages/
REINHEIT:site-packages dheerajchand$ which python
/Library/Frameworks/Python.framework/Versions/2.7/bin/python
REINHEIT:site-packages dheerajchand$ pip freeze
distribute==0.6.28
readline==6.2.2
vboxapi==1.0
virtualenv==1.8.2
wsgiref==0.1.2
REINHEIT:site-packages dheerajchand$
It looks like the installer put some files in site-packages, but that they're not installed. Am I reading this correctly? If so, what do I have to do to install them and test to see if they work?
Best,
-dx
On Dec 28, 2013, at 9:16 PM, William Kyngesburye <woklist at kyngchaos.com> wrote:
> No, readline is present (in the python framework), but something seems to be wrong with it.
>
> On Dec 28, 2013, at 9:14 PM, Dheeraj Chand wrote:
>
>> So, sorry if this is an idiotic question, but is the solution then to install readline?
>>
>> Best,
>>
>> -dx
>>
>>
>> On Dec 28, 2013, at 7:19 PM, William Kyngesburye <woklist at kyngchaos.com> wrote:
>>
>>> Looks like something wrong with readline. The python.org python used in your virtual environment has its own readline library, while something else is loading libedit (a repalcement readline) from the system. Maybe there's confusion about which to use, or something is wrong with readline.
>>>
>>> On Dec 28, 2013, at 6:58 PM, Dheeraj Chand wrote:
>>>
>>>> Sorry for the confusion. I believe that these are what you're looking for.
>>>>
>>>> -dx
>>>> <python_2013-12-27-144518_REINHEIT.crash><python_2013-12-27-144829_REINHEIT.crash><python_2013-12-27-145038_REINHEIT.crash><python_2013-12-27-221320_REINHEIT.crash>
>>>>
>>>>
>>>> On Dec 28, 2013, at 6:53 PM, William Kyngesburye <woklist at kyngchaos.com> wrote:
>>>>
>>>>> It won't show up in the general console log, there's a separate section for crash logs (User Diaognostic Reports section).
>>>>>
>>>>> On Dec 28, 2013, at 6:51 PM, Dheeraj Chand wrote:
>>>>>
>>>>>> Nope, nothing. I am going to try again with 1.9 after dinner and send the results. This is the closest that I could find in Console:
>>>>>>
>>>>>> 12/28/13 6:33:47.063 PM sudo[48980]: dheerajchand : TTY=ttys000 ; PWD=/Users/dheerajchand/Python_Environments/geo ; USER=root ; COMMAND=/usr/bin/find / -name gdal-config
>>>>>>
>>>>>> Best,
>>>>>>
>>>>>> -dx
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> On Dec 28, 2013, at 6:48 PM, William Kyngesburye <woklist at kyngchaos.com> wrote:
>>>>>>
>>>>>>> Is there a crash report in Console.app from when you try to load GDAL python? It will probably be as python.
>>>>>>>
>>>>>>> On Dec 28, 2013, at 6:27 PM, Dheeraj Chand wrote:
>>>>>>>
>>>>>>>> William,
>>>>>>>>
>>>>>>>> Thanks for responding. I was using the 1.10 because I updated when you updated. I got the same kind of error when I used 1.9, though. I'd be more than happy to try again and show you, though. Do you think that it's necessary?
>>>>>>>>
>>>>>>>> Best,
>>>>>>>>
>>>>>>>> -dx
>>>>>>>> On Dec 28, 2013, at 11:46 AM, William Kyngesburye <woklist at kyngchaos.com> wrote:
>>>>>>>>
>>>>>>>>> Sorry, I didn't read to the very end, where you did recompile with the --include-dirs option.
>>>>>>>>>
>>>>>>>>> The missing lib/ warning is no problem, just a warning.
>>>>>>>>>
>>>>>>>>> The segfault though, doesn't say enough to figure out what's wrong. Maybe there is something in the Console.app, like a crash report (User Diaognostic Reports section).
>>>>>>>>>
>>>>>>>>> On Dec 28, 2013, at 11:38 AM, William Kyngesburye wrote:
>>>>>>>>>
>>>>>>>>>> The compile flag to set the GDAL include dir is wrong for the framework. It seems to be assuming that the framework is a 'nix lib/include structure, but really the include dir is named "Headers". Odd that it uses supposedly gdal-config, but doesn't get the -cflags from it that have the correct include dir.
>>>>>>>>>>
>>>>>>>>>> I see that Tim's instructions actually call for using --include-dirs=/Library/Frameworks/GDAL.framework/Versions/1.9/Headers/
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> I wonder: Tim, what is it about the "virtual environment" that makes the existing GDAL python in the framework not work? (I don't know anything about python virtual env)
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> On Dec 27, 2013, at 10:33 PM, Dheeraj Chand wrote:
>>>>>>>>>>
>>>>>>>>>>> Hi, all,
>>>>>>>>>>>
>>>>>>>>>>> I was working on a project, that at a high level, can be described as :
>>>>>>>>>>>
>>>>>>>>>>> 1. Connect Python to PostGIS,
>>>>>>>>>>> 2. Grab spatial tables' data,
>>>>>>>>>>> 3. *PROCESS*,
>>>>>>>>>>> 4. Spit out shapefiles.
>>>>>>>>>>>
>>>>>>>>>>> In the process of doing this, I was trying to use the Python GDAL library in a new virtualenv called geo, but I got these errors. I am using William Kyngesbury's GDAL 1.10 for OS X, and I'm running into crazy errors. I am following the instructions from here (http://linfiniti.com/2013/02/installing-python-gdal-into-a-python-virtualenv-in-osx/), and this is what happened in my shell.
>>>>>>>>>>>
>>>>>>>>>>> Can anyone help me out?
>>>>>>>>>>>
>>>>>>>>>>> Best,
>>>>>>>>>>>
>>>>>>>>>>> -dx
>>>>>>>
>>>>>>>>>>> (geo)REINHEIT:GDAL dheerajchand$ python
>>>>>>>>>>> Python 2.7.3 (v2.7.3:70274d53c1dd, Apr 9 2012, 20:52:43)
>>>>>>>>>>> [GCC 4.2.1 (Apple Inc. build 5666) (dot 3)] on darwin
>>>>>>>>>>> Type "help", "copyright", "credits" or "license" for more information.
>>>>>>>>>>>>>> from osgeo import gdal
>>>>>>>>>>>>>> gdal.__version__
>>>>>>>>>>> Segmentation fault: 11
>>>>>>>>>>> (geo)REINHEIT:GDAL dheerajchand$
>
> -----
> William Kyngesburye <kyngchaos*at*kyngchaos*dot*com>
> http://www.kyngchaos.com/
>
> "Those people who most want to rule people are, ipso-facto, those least suited to do it."
>
> - A rule of the universe, from the HitchHiker's Guide to the Galaxy
>
>
More information about the postgis-users
mailing list