On 5/08/12 7:13 AM, gmn wrote:
I'm sending this here rather than to vim_use because it's a Mac-specific problem, even though it's about the official console vim from Bram's hg repo (aka "BramVim").A short while ago, I discovered I had been calling --disable-darwin for the configuring of BramVim. I'd been doing this for years, disguised 'neath a bash alias. I couldn't recall why I was doing something so apparently dumb. Well, I discovered why, I think. If I don't --disable-darwin, then building vim chokes on enabling the python interpreter: objects/if_python.o objects/os_macosx.o objects/os_mac_conv.o objects/main.o objects/memfile.o -lm -lncurses -liconv -lintl -framework Cocoa -framework Python Undefined symbols: "__PyObject_NextNotImplemented", referenced from: __PyObject_NextNotImplemented$non_lazy_ptr in if_python.o "_PyCapsule_GetPointer", referenced from: _convert_dl in if_python.o "_PyCapsule_New", referenced from: _convert_dl in if_python.o ld: symbol(s) not found collect2: ld returned 1 exit status make[1]: *** [vim] Error 1 make: *** [first] Error 2
Hmm. Strange. I've never had any problem building (Mac)Vim against Python, and I certainly don't use --disable-darwin. I suspect I've always been building against a Python from MacPorts, though.
If I do --disable-darwin, I get python support just fine.
Interesting.
It should be known that: 0. This is on both Tiger and Leopard. There's no problem on 10.7 or even 10.8
I guess I'll have to fire up an older machine to see if I can investigate this further at all. I should be able to do that if I need to.
1. My $PYTHONPATH is set to /usr/local/gmnpyth, which is where python 2.7.3 lives.
How did it get there?
2. Even on Tiger and Leopard, building *MacVim* with the same $PYTHONPATH and --enable-pythoninterp works just fine.
Also interesting. I wonder if there's a patch in MacVim that should be 'pushed upstream' or moved outside a preprocessor conditional or something. When you attempt to build BramVim, are you doing it using the MacVim sources, but disabling the GUI, or with sources directly from Bram. Can you try it both ways and see if there is any difference?
3. The culprit is not any of the recent patches to vim dealing with python compilation problems (surprisingly).
Good to know.
4. It took a while to show up after I stopped using --disable-darwin because I hadn't obliterated the build caches inside my clone of the hg repo in a long time. Once I did so, pow!
Two of Murphy's Laws: "Everything that can go wrong will go wrong, and at the worst possible time." "If everything seems to be going well, you have obviously overlooked something." Ben. -- You received this message from the "vim_mac" maillist. Do not top-post! Type your reply below the text you are replying to. For more information, visit http://www.vim.org/maillist.php
