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: path conversion regression?


On Fri, Jan 09, 2009 at 02:42:09PM +0100, Corinna Vinschen wrote:
>On Jan  9 10:09, Corinna Vinschen wrote:
>> On Jan  9 00:43, Christopher Faylor wrote:
>> > Shouldn't this:
>> > 
>> > cygpath -u 'c:\cygwin\bin\cygwin1.dll'
>> >             ^
>> > 
>> > be equivalent to:
>> > 
>> > cygpath -u 'C:\cygwin\bin\cygwin1.dll'
>> >             ^
>> > [...]
>> I thought the code I added lately (*) had taken care of DOS paths so that
>> they are always treated caseinsensitive so your patch shouldn't have been
>> necessary.  Looks like that doesn't work as advertised.  I take a look.
>
>Thinko, sorry.  That's not the problem here.
>
>> Hmm, the mount table entries are another point.  I think what we really
>> should do is to convert all incoming DOS paths, no matter where they
>> come from, to an uppercased drive letter.  Later code then doesn't have
>> to care for the case anymore.
>
>I've created a patch which always creates uppercased drive letters
>so it doesn't matter if the path is casesensitive or not.  It also
>allows to revert your patch to path_prefix_p.  See below.  Is that
>ok with you or would you rather stick to the test in path_prefix_p?

No, go ahead and revert my patch.


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