Message6212
 
            
            
            
 
   
   
 
  | As I understand it, this issue is related to this same problem seen in JRuby of the same setup (cygwin + alt terminal):
http://jira.codehaus.org/browse/JRUBY-5068
The reason it worked in the past was that the logic was backwards - isatty should return False, given the workaround used by such terminals by using pipes. This meant that just for this setup, our previous buggy logic actually worked!
-i does seem to be the obvious thing to do, but looking at org.python.util.jython#run shows that opts.fixInteractive is treated differently than opts.interactive. Need to figure out why. But maybe we need a separate flag, or more likely property specific to Jython, that means "terminal interactive", preserving the other case where it's only has a part of the interactive quality.
This was also recommend in the JRuby case. |  |
 
| Date | User | Action | Args |  | 2010-10-26 16:54:51 | zyasoft | set | messageid: <1288112091.67.0.486911155735.issue1669@psf.upfronthosting.co.za> |  | 2010-10-26 16:54:51 | zyasoft | set | recipients:
  + zyasoft, neuroaudio |  | 2010-10-26 16:54:51 | zyasoft | link | issue1669 messages |  | 2010-10-26 16:54:50 | zyasoft | create |  | 
 |