wrote:
> On Wednesday, July 06, 2011 7:35:21 PM, Vlad Irnov wrote:
> > On Jun 30, 10:35 am, Alexandre Martani<alexandre.mart...@gdx.com.br>
> > wrote:
> >> Hello all,
>
> >> I am trying to use Vim R plugin [1], which is a Python plugin that uses
> >> PyWin32 on Windows. I have installed Python 2.7.1 and PyWin32. It is
> >> installed correctly, as I am able to do a "import win32api" from python
> >> shell without errors. But, on GVim, if I do:
>
> >> :python import win32api
>
> >> I receive the following error:
>
> >> Traceback (most recent call last):
> >> File "<string>", line 1, in<module>
> >> ImportError: DLL load failed: The specified module could not be found.
>
> >> Notice that this is not the standard error for module not found ("No
> >> module named x"). It is probably related to the path python uses for its
> >> DLL modules. I have played with sys.path, but without success.
>
> > I cannot reproduce your problem on my system:
> > Win2k; Vim 7.3.145; Python 2.6.5; pywin32-214.win32-py2.6.exe
> > inst
> alled
> > My PATH has Python's top level directory "C:\Python" where it's
> > installed and no subdirs or other Python-related stuff.
I should mention that I installed Vim from sources. I compiled
gvim.exe and vim.exe with MinGW (GCC), with Python 2.6 support only,
no Python 3. In contrast, your version output shows your Vim was
probably built with Microsoft VC toolkit (or whatever it is called).
It is possible that it does not play nice with PyWin32. It also has
support for both Python 2.7 and 3.1. May be these two clash somehow.
I suggest you compile your own gvim.exe and/or vim.exe as per my
instructions:
http://vim.wikia.com/wiki/Build_Python-enabled_Vim_on_Windows_with_MinGW?useskin=monobook
If that helps, then it's a Vim bug.
You can also try to find a differently built version of Vim but you
have to make sure it supports Python 2.7. The Cream version needs
Python 2.6.
> >> I also noticed the first path in sys.pat
>
> h is "C:\\must>not&exist", which
>
> >> is probably a bug. Don't know if this could be related, but I tried do a
> >> sys.path.pop(0), and it didn't work.
>
> > It used to be an empty string. It was changed to this bogus path for
> > some security reasons:
> >http://ftp.nluug.nl/vim/patches/7.2/7.2.045
>
> It was changed, but I suppose from the path that it should have been
> removed later:
>
> + /* Remove the element from sys.path that was added because of our
> + * argv[0] value in PythonMod_Init(). Previously we used an empty
> + * string, but dependinding on the OS we then get an empty entry or
> + * the current directory in sys.path. */
> + PyRun_SimpleString("import sys; sys.path = filter(lambda x: x !=
> '/must>not&exist', sys.path)");
>
> I think it is not being removed because Python on Windows changes it to
> 'C:\\must>not&exist'.
You are probably right that it should be removed. It's not going to
cause any problems though. Such nonexistent directories are ignored.
--
You received this message from the "vim_use" 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
No comments:
Post a Comment