Ralf Gommers wrote: > Yes I got one response at http://projects.scipy.org/numpy/ticket/1322, a > negative one unfortunately.
For what it's worth, the posted binary seems to work fine on my OS-X 10.5 PPC system (python2.6), so maybe the PPC part works, the Intel part not. > My conclusion is that binaries built on 10.6 > do not work on 10.5 even when using the correct SDK and deployment > target. Well, I still think if you do everything right, it works, but yes, it is a pain to get right, and a pain to test. Building on the oldest system you want to support is a lot easier. > Sage distributes separate binaries for each OS X version, presumably > compiled on those same versions. If we want to have a 10.5 binary we > should just do the same. I don't think so -- that's a lot to maintain, and the older versions work on the newer OS -- so why build the newer? Also a lot (OK -- some, I have no idea how many) of us re-distribute, via py2app, and so need Universal binaries that work on all the systems we need to support. 10.4 or greater would be great -- basically matching the python.org builds. > The binary naming scheme should also change to include the version. yup. Thanks for working in this! -Chris -- Christopher Barker, Ph.D. Oceanographer Emergency Response Division NOAA/NOS/OR&R (206) 526-6959 voice 7600 Sand Point Way NE (206) 526-6329 fax Seattle, WA 98115 (206) 526-6317 main reception [email protected] _______________________________________________ NumPy-Discussion mailing list [email protected] http://mail.scipy.org/mailman/listinfo/numpy-discussion
