This is the mail archive of the cygwin-developers mailing list for the Cygwin project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: 1.7.1 release date?


Christopher Faylor schrieb:
/etc/termcap ...
I don't care if people casually peruse it. That's not a justification
for keeping it around.
That was only a side aspect. If you prefer to remove it, that's fine. I was mainly opposing to leaving a buggy file around.

On Thu, Nov 19, 2009 at 05:15:48PM +0100, Thomas Wolff wrote:
I reported trouble on one machine recently where I could not install cygwin to a network drive.
And, is this a regression?
Yes, after it failed, I reinstalled 1.5 to the same network drive.

Assume just 1% of cygwin users have this trouble and image the number of complaints...
Imagine of 10% of cygwin users started using Windows 7.
I understand the trade-off. An exotic problem reported by one person vs. the desire to support Windows 7 soon. Especially since it may be subtle to debug, don't wait for it.

I'm not overly concerned about people installing to network drives.
To explain the use case: it's a lab network, with user data on the network, users may have different PCs each time, and local drives are all temporary and often deleted (and even if they weren't I wouldn't wish to install N times).
We can always deal with that later. No one is claiming that 1.7.1 will be
perfection. I believe that it fixes many problems that are in 1.5.25
(witness the pthread_join deadlock report) and it's about time that it
went live.
Agreed.

I think this (charset/lang) stuff is working quite well now, with one small but significant issue: I had suggested to add set LC_CTYPE=C.UTF-8 to cygwin.bat and I'd like to strongly repeat this suggestion as it might avoid much trouble.
The point about releasing something is that, at some point, you have to
stop making "small but significant" changes and "just do it".
In this case, the change is really small, i.e. trivial, doesn't need any further debugging, might have much greater impact if not fixed, and I had suggested it some weeks ago already...

Thomas


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]