From: Alan W. Irwin <irwin@be...> - 2004-12-01 01:36:49
I just discovered the hard way that swig-1.3.22 changed the style of the
returned tuples compared to swig-1.3.21. I don't know how this escaped
testing, but when I did a clean start this morning (for another purpose),
the python interface got generated by swig-1.3.22, and example 8 and 11 quit
working due to the "new" default hierarchical returned tuple style.
I have fixed this problem by using our own version of fragments.i so that
simple tuples are returned. That is, (1,2,3,4,5) is returned rather than
the hierarchical style of tuple ((((1,2),3),4),5). (See recent commit.)
Rafael, I just checked, and your python-plplot package escaped running afoul
of this inconsistency because it was generated with swig-1.3.21. Anyhow,
when you make your next set of Debian packages using swig-1.3.22, be sure to
include this fragments.i fix that I just put into cvs.
Alan W. Irwin
Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).
Programming affiliations with the FreeEOS equation-of-state implementation
for stellar interiors (freeeos.sf.net); PLplot scientific plotting software
package (plplot.org); the Yorick front-end to PLplot (yplot.sf.net); the
Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project
From: Rafael Laboissiere <rlaboiss@us...> - 2004-12-01 09:21:20
* Alan W. Irwin <irwin@...> [2004-11-30 17:36]:
> Rafael, I just checked, and your python-plplot package escaped running
> afoul of this inconsistency because it was generated with swig-1.3.21.
> Anyhow, when you make your next set of Debian packages using swig-1.3.22,
> be sure to include this fragments.i fix that I just put into cvs.
Thanks for the remainder but, since you included fragments.i in the
EXTRA_DIST variable, I think that no particular action will be required for
the next Debian packages. Indeed, new packages will be built from cvs
snapshot tarballs, which will include the file you are mentioning.