Re: PATCH: Cygwin: file mode support and ufs writecleanswap bugfix

Sure. I was aiming to get it in as #1, and didn't know if everyone'd see it that way. I'll do that..

----- Original Message -----
From: "Henrik Nordstrom" <hno@hem.passagen.se>
To: "Robert Collins" <robert.collins@itdomain.com.au>
Cc: <squid-dev@squid-cache.org>
Sent: Thursday, January 04, 2001 6:16 AM
Subject: Re: PATCH: Cygwin: file mode support and ufs writecleanswap bugfix

> May I propose that you add definitions of O_TEXT and O_BINARY to
> defines.h instead of all the #ifdef CYGWIN...
>
> #ifndef O_TEXT
> #define O_TEXT 0
> #endif
> #ifndef O_BINARY
> #define O_BINARY 0
> #endif
>
> As you say the information that a file is text/binary is interesting to
> other platforms as well.
>
> /Henrik
>
>
> Robert Collins wrote:
> >
> > Hi,
> > If one of the core team could look over this patch and if acceptable drop into HEAD & squid 2.4 sometime soon I'd really
appreciate
> > it.
> >
> > It's all cleanly #if defined, except for one little bit in store_dir_ufs.c, where I've solved a somewhat general case (that
would
> > only affect platforms which don't let you rename open files - OS/2/Cygwin/NT).
> >
> > This should all go into squid 2.4 & head. Without it users on cygwin with text mode mounts will die badly, and all cygwin users
will
> > not ever see a (CLEAN) status on their cache_dir's. Why I was getting clean status when I first looked into this may be due to
some
> > of the 'emulation' work done on cygwin to make UN*X assumptions (ie unlinking open files) work somewhat.
> >
> > As a side note, I've reactivated the cygwin branch, (where this work currently resides), and I will be doing some experimental
stuff
> > there in the near future: so use the attached patch, not a CVS diff.
> >
> > Thanks,
> > Rob
> >
> > ---------------------------------------------------------------
> >
> > Name: cygwin.patch
> > Part 1.2 Type: unspecified type (application/octet-stream)
> > Encoding: quoted-printable
>
>
Received on Wed Jan 03 2001 - 16:27:26 MST