Re: fish PATH problem

[Andrew Schulman <schulman dot andrew at epa dot gov>]

> On 08/10/2014 18:36, Alive4Ever wrote:
> > On Wednesday, October 08, 2014 11:08:43 AM you wrote:
> >> Does that work. It seemed to have the same issue as noted in the
> >> following thread https://cygwin.com/ml/cygwin/2014-04/msg00111.html
> >>
> > It should work, although I don't recommend running cygwin apps directly
> > via wincmd, unless the underlying shell is invoked as interactive login
> > shell.
> So I just re-tested starting a fish login shell after updating cygwin64. 
> As I suspected, fish has not been updated since the thread I quoted (Apr 
> 2014), and the issue remains (with my install). Google doesn't show the 
> maintainer on list since the announcement (Oct 2013), and his e-mail 
> bounces. Cc'ing anyway, just in case.

If fish needs a new maintainer, it looks like not a hard package to
maintain.  It has a simple and up-to-date cygport script, and builds OOTB
with just one small source code patch.

> Andrew: appreciated if you can confirm whether simply invoking as a 
> login shell works for you.

It sort of works.  When I change my shell to /usr/bin/fish in /etc/passwd
and run 

C:\cygwin64\bin\mintty.exe -

I don't get any error messages, as I first reported.  But the environment
isn't set up correctly.  The main problem seems to be that although I'm
running 64-bit cygwin, the directory for 32-bit cygwin is mounted as the
root filesystem:

 ~> cygpath -w /
C:\cygwin

By comparison when I start fish from bash, I get:

aschulma@LC77E1AASCHULMA ~> cygpath -w /
C:\cygwin64

This is a pretty serious problem.  Any thoughts on why it happens?

Andrew


--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple