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: CWD and long paths


On Oct 12 09:22, Lev Bishop wrote:
> On 10/12/07, Corinna Vinschen wrote:
> 
> > In Cygwin we can do as we like, so we can handle CWDs with long path
> > names without problems.  Unfortunately we have to call non-Cygwin
> > applications once in a while and then it gets weird.
> >
> > So, here's the question: How should we handle the CWD in future?
> >
> > - We could stick to only supporting CWDs of MAX_PATH length so we don't
> >   have a problem when calling native apps.
> >
> > - We could support long CWDs and change the CWD to the longest path
> >   prefix <= MAX_PATH as soon as we call a native app.
> >
> > - Something else?
> 
> Something like:
> 
> subst t: prefix\of\long\path
> t:
> cd rest\of\long\path
> win32_command
> subst /d t:
> 
> This allows to get up to 2*MAX_PATH and can be iterated to get deeper.
> 
> Ugly, I know, but maybe better than the alternatives?

I don't think so.  First, it only helps for 2 * MAX_PATH, as you say.
Second, it's a method only working for users.  How should that work for
the DLL without user intervention?  Searching the next free drive letter
and use it?  What if this is by chance the drive letter used for, say,
the user's USB drive and scripts or apps are relying on the drive
letter?  What if all drive letters are taken?  That doesn't sound very
reliable.


Corinna

-- 
Corinna Vinschen                  Please, send mails regarding Cygwin to
Cygwin Project Co-Leader          cygwin AT cygwin DOT com
Red Hat


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