file times still not quite right with 1.5.14 on Windows 98

[Josef Drexler]

Corinna Vinschen wrote:
> On Apr  3 15:54, Josef Drexler wrote:
> 
>>Eric Blake wrote:
>>
>>>The bug is not that ctime was touched, but that mtime was not
>>>touched.  Normally, Windows updates mtime automatically if you edit a
>>>file, only the ctime needed special treatment from cygwin.  I have no
>>>idea why Win98 is not touching the mtime on appending or truncation.
>>
>>Oh, I have no problem with ctime changing, but mtime not changing is
>>definitely a bug in cygwin.  It must somehow intentionally copy the old
>>mtime when a file is modified.  If I repeat the sequence I posted using 
> 
> 
> It's not exactly Cygwin, it's the incredible braindead Windows 95/98/Me
> and I'm more and more wondering why anybody is still using it voluntarily.

In my case the reason is that the programs I have don't and can't be 
made to run on NT/2K/XP.

> No offence meant, I'm just venting.
> 
> Cygwin is touching ctime right before closing the file, when a write or
> one of chmod/chown/acl has been called successfully.
> 
> Win98 is apparently "confused" by the fact that ctime is changed to a value
> bigger than the modification time and then simply refuses to change the
> modification time on file close.  I first thought this might be a FAT
> problem, but NT changes the modification time just fine.  I can't stop
> shaking my head about 9x.

So, is that fixable or do I need to compile a cygwin1.dll with the ctime 
change removed?

-- 
    Josef Drexler                 |       http://jdrexler.com/home/
---------------------------------+----------------------------------------
  Please help Conserve Gravity    |  Email address is *valid*.
  Play Chess, not Basketball.     |  Don't remove the "nospam" part.

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