Mailing-List: contact cygwin-help AT sourceware DOT cygnus DOT com; run by ezmlm List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-owner AT sources DOT redhat DOT com Delivered-To: mailing list cygwin AT sources DOT redhat DOT com Message-ID: <3B3967A3.1090406@gruntose.com> Date: Wed, 27 Jun 2001 00:57:07 -0400 From: "Fred T. Hamster" User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:0.9.1) Gecko/20010607 X-Accept-Language: en-us MIME-Version: 1.0 To: cygwin AT cygwin DOT com Subject: Re: Cygwin: Interoperability Is Important (was Cygwin: Open or Closed System, etc) References: Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Robert Collins wrote: >I'm not agreeing or disagreeing about the design goal. I simply want to >make a very small but important point... "cygwin" that is the >cygwin1.dll has full support for ms dos or more accurately win32 >pathnames. Call fstat(), or fopen() or ... and give it a ms dos >pathname. If there is a real file with that msdos pathname the call will >behave as you would expect it to. >Finito. Done. No problems. Possibly minor tweaks needed for performance or bugfixing. > i agree, on specific file names without wildcards, there are usually no problems unless one is foolish enough to provide the wrong case for characters in the name. however, it still seems to me that wildcards are an important topic in parsing pathnames. that is where i see some possibilities for improvement that would make the usage of cygwin a bit more seamless with ms-windows apps. >$ echo "c" > "foo" >$ echo "a" > "foo:1" >$ echo "b" > "foo:2" > now cat foo - it will still be "c". cat foo:1 and get "a". > that is really freaky. thanks for the explanation. so, if one were to interpret a situation where the "second character is a colon" as meaning "that the colon is always a drive letter", then that would be incorrect for the set of filenames that are exactly one letter long, providing that one wanted to set up streams on those filenames. i'm certainly not inclined towards asking for that kind of support to be added to cygwin, although it does appear to be part of win32. even some win32 programs have issues with it though; when i tried "type foo:2", i got a complaint that "The filename, directory name, or volume label syntax is incorrect." so microsoft's api is incompletely applied within their distributed toolset. for some reason, i am not surprised at that. thanks, fred. -- _____ chosen by the Nechung Oracle Program [ http://www.gruntose.com/ ] _____ WANTED: Meaningful overnight relationship. _____________ not necessarily my opinions, not necessarily not. _____________ -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/