Re: [ANNOUNCEMENT] Updated: libreadline7-7.0.3-3

[Steven Penny <svnpenn@gmail.com>]

On Fri, 28 Jul 2017 19:54:27, Doug Henderson wrote:
> One of these may be the commit that I recall.
> 
> https://github.com/cygwin/cygwin/commit/ef007184874ead6f288e432eb23bfc76bf65929d
> 
> https://github.com/cygwin/cygwin/commit/5a3496c3e3c159e6cfb4879f5adae1092927483f
> 
> https://github.com/cygwin/cygwin/commit/7630e384623f778b88be8926e2c2a2e6fcbd9008

I looked at those 3, and I am not seeing something that would cause this. Hell:

http://github.com/cygwin/cygwin/commit/ef00718

is basically changing the variable names. Would be nice to have Corinna weigh in
though.

> I gave up trying to trying read C++ about 15 years ago, so all I
> understood was the English parts.

Right, so you are posting links to code that by your own admission you dont
understand.

> Also the following speak for themselves.
> 
> $ ldd omega_cygwin.exe
> $ ldd omega_mingw32.exe
> $ ldd $(which bash)
> $ ldd $(which dash)

The core of the issue here is proper UTF-8 input with Cygwin. This used to be
possible before the current version of readline. Before that, if you had the
proper font, and the proper codepage, Cygwin would accept UTF-8 input and could
handle UTF-8 output. Now at least the input part of it is broken.

The point of my example omega.c was to answer Corinnas question, which was
essentially, "why use 65001 codepage". The answer is because you have to, if you
want sane UTF-8 input/output across cygwin/mingw32 builds.

Your DLL dumps here serve about as little purpose as the GitHub links before,
Doug, unless you are willing to repro the issue as Eric has, you are not really
adding anything to the conversation. You are distracting from it.


--
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