[PyQt] Bug in QScintilla-2.8.3 configure.py? (with fix)

[Patrick Stinson]

Uh oh, once I get configure.py to complete successfully the make file (and Qsci.pro) don’t have any sources, and the dylib is empty.

I am running latest yosemite with pyqt5/5.4.1 installed via homebrew. Here is the output with the unmodified configure.py:


turin:Python patrick$ python3 configure.py --pyqt=PyQt5
Configuring QScintilla 2.8.3...
QScintilla 2.8.3 is being used.
The QScintilla .sip files will be installed in
/usr/local/Cellar/python3/3.4.3/Frameworks/Python.framework/Versions/3.4/share/sip/PyQt5.
The QScintilla module will be installed in
/usr/local/Cellar/python3/3.4.3/Frameworks/Python.framework/Versions/3.4/lib/python3.4/site-packages/PyQt5.
PyQt 5.4 is being used.
Qt 5.4.1 is being used.
sip 4.16.5 is being used.
The sip executable is /usr/local/bin/sip.
The QScintilla module is being built with 'protected' redefined as 'public'.
The QScintilla API file will be installed in
/usr/local/Cellar/qt5/5.4.1/api/python.
Generating the C++ source for the Qsci module...
Error: Unable to create the C++ code.
turin:Python patrick$ 


Thoughts?

> On Mar 11, 2015, at 1:57 AM, Patrick Stinson <patrickkidd at gmail.com> wrote:
> 
> Hello, it’s been a long time.
> 
> I noticed that configure.py won’t run unless I disabled the file existence check at line 1265:
> 
> #    if not os.access(check_file, os.F_OK):                                                                                                               
> #        error("Unable to create the C++ code.") 
> 
> it’s checking for the *existence* of ./Python/sipAPIQsci.h, which of course does not exist. Perhaps it should check    for write access instead?
> 
> Thanks for keeping this excellent project going for all of these years.
> -P

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.riverbankcomputing.com/pipermail/pyqt/attachments/20150311/bfef6df0/attachment.html>