This is the mail archive of the cygwin 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: Problems with cygwin cvs over ssh.


>> Right.  I missed the "." in the original message.  The change that
>> prompted this behavior seems to be
>> <http://cygwin.com/ml/cygwin-cvs/2005-q3/msg00224.html>.  I'm assuming the
>> motivation for this patch was to duplicate Linux's behavior (which doesn't
>> allow trailing "." in a path passed to mkdir).
> 
> Indeed.  Eric mentioned that the coreutils testsuite tests thsi behaviour
> explicitely.  Since there's not much impact speedwise, we just added
> appropriate checks to be POIX compatible here.

I want to state, that cygwin might return the wrong error-code! Instead
of "file already exists" it returns "no such file or directory".


Cygwin's bahaviour:
$ mkdir /tmp/.
mkdir: cannot create directory `/tmp/.': No such file or directory

Linux' behaviour:
# mkdir /tmp/.
mkdir: cannot create directory `/tmp/.': File exists


Indeed, strace shows me, that on Linux mkdir() returns EEXIST in the
case of a mkdir("/tmp/.")-call. Cygwin doesn't seem to do it this way,
it seems to return ENOENT which would not be Linux-like ;-)

Unfortunatly i don't understand the output of cygwin's strace and cannot
check, what the mkdir("/tmp/.") returns there. Does anybody have a clue?


  Sven

--
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/


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