The ESA's Python-xmp-toolkit now works with python 2.4 (you'll have to build ctypes yourself obviously). Should anyone run into any problems can you please let me know so I can patch and collect and push upstream if necessary. Lars got me up to date and a late Feb/Early March 1.0 release is expected. A lot of the current implementation for the dev release is missing. Federico Caboni from CRS4 has been working on it however. Hopefully something hits version control soon.
In the meantime this weekend or something I'll try pyexiv2 to see if we can get it working on 2.4 as well. I'm not sure how much 2.5.x ism's it has but we'll try. If this is new to you I'm doing this because Zope is still using Python 2.4 and I was working on something that needed xmp natively and ended up having to use exiftool. Needless to say it was suboptimal.
Update: An initial look at pyexiv2 on a rhel4 box proves to be a non-starter; there are boost and g++ issues which would mean upgrading g++, which would involve other things i'm not prepared to invest in. Also if you are installing pyexiv2 anywhere other than default you'll need this patch. Pyexiv2 uses scons, which is fine but it doesn't take into account custom python installs and locations.
python-xmp-toolkit works with python 2.4
The ESA's Python-xmp-toolkit now works with python 2.4 (you'll have to build ctypes yourself obviously). Should anyone run into any problems can you please let me know so I can patch and collect and push upstream if necessary. Lars got me up to date and a late Feb/Early March 1.0 release is expected. A lot of the current implementation for the dev release is missing. Federico Caboni from CRS4 has been working on it however. Hopefully something hits version control soon.
In the meantime this weekend or something I'll try pyexiv2 to see if we can get it working on 2.4 as well. I'm not sure how much 2.5.x ism's it has but we'll try. If this is new to you I'm doing this because Zope is still using Python 2.4 and I was working on something that needed xmp natively and ended up having to use exiftool. Needless to say it was suboptimal.
Update: An initial look at pyexiv2 on a rhel4 box proves to be a non-starter; there are boost and g++ issues which would mean upgrading g++, which would involve other things i'm not prepared to invest in. Also if you are installing pyexiv2 anywhere other than default you'll need this patch. Pyexiv2 uses scons, which is fine but it doesn't take into account custom python installs and locations.