You are viewing a plain text version of this content. The canonical link for it is here.
Posted to cvs@httpd.apache.org by dr...@locus.apache.org on 2000/06/12 15:51:23 UTC
cvs commit: apache-2.0/src/lib/apr/file_io/unix fileio.h readwrite.c
dreid 00/06/12 06:51:22
Modified: src/lib/apr/file_io/unix fileio.h readwrite.c
Log:
Another change to allow the newest version of BeOS to be used to it's
full potential.
Revision Changes Path
1.20 +3 -0 apache-2.0/src/lib/apr/file_io/unix/fileio.h
Index: fileio.h
===================================================================
RCS file: /home/cvs/apache-2.0/src/lib/apr/file_io/unix/fileio.h,v
retrieving revision 1.19
retrieving revision 1.20
diff -u -r1.19 -r1.20
--- fileio.h 2000/04/22 06:16:20 1.19
+++ fileio.h 2000/06/12 13:51:22 1.20
@@ -98,6 +98,9 @@
#ifdef BEOS
#include <kernel/OS.h>
#endif
+#if BEOS && HAVE_ARPA_INET_H
+#include <sys/socket.h> /* for fd_set definitions */
+#endif
/* End System headers */
#define APR_FILE_BUFSIZE 4096
1.52 +7 -3 apache-2.0/src/lib/apr/file_io/unix/readwrite.c
Index: readwrite.c
===================================================================
RCS file: /home/cvs/apache-2.0/src/lib/apr/file_io/unix/readwrite.c,v
retrieving revision 1.51
retrieving revision 1.52
diff -u -r1.51 -r1.52
--- readwrite.c 2000/06/01 22:23:16 1.51
+++ readwrite.c 2000/06/12 13:51:22 1.52
@@ -55,7 +55,11 @@
#include "fileio.h"
#include "apr_lock.h"
-#ifndef BEOS
+#if !BEOS || (BEOS && HAVE_ARPA_INET_H)
+#define USE_WAIT_FOR_IO
+#endif
+
+#ifdef USE_WAIT_FOR_IO
static ap_status_t wait_for_io_or_timeout(ap_file_t *file, int for_read)
{
struct timeval tv, *tvptr;
@@ -159,7 +163,7 @@
do {
rv = read(thefile->filedes, buf, *nbytes);
} while (rv == -1 && errno == EINTR);
-#ifndef BEOS
+#ifdef USE_WAIT_FOR_IO
if (rv == -1 &&
(errno == EAGAIN || errno == EWOULDBLOCK) &&
thefile->timeout != 0) {
@@ -232,7 +236,7 @@
do {
rv = write(thefile->filedes, buf, *nbytes);
} while (rv == (ap_size_t)-1 && errno == EINTR);
-#ifndef BEOS
+#ifdef USE_WAIT_FOR_IO
if (rv == (ap_size_t)-1 &&
(errno == EAGAIN || errno == EWOULDBLOCK) &&
thefile->timeout != 0) {
Re: cvs commit: apache-2.0/src/lib/apr/file_io/unix fileio.h readwrite.c
Posted by David Reid <dr...@jetnet.co.uk>.
I'll be changing this tomorrow when I return from Edinburgh in the morning.
david
----- Original Message -----
From: "David Reid" <dr...@jetnet.co.uk>
To: <ne...@apache.org>; <tr...@bellsouth.net>
Sent: Monday, June 12, 2000 5:03 PM
Subject: Re: cvs commit: apache-2.0/src/lib/apr/file_io/unix fileio.h
readwrite.c
> > Why do we need BEOS as part of the check? just include it if it is
> > available...
>
> Fair comment. I didn't want to include files unless they're needed, and
> AFAIK most unix's don't need socket.h to get fd_set defined.
>
> >
> >
> > Why do we check HAVE_ARPA_INET_H for including sys/socket.h?
> >
>
> I'm not checking for the presence of arpa_inet.h but using it to
> differentiate between versions of BeOS.
>
> This code is for a developmental version of BeOS with a new networking
> stack, and so far it doesn't have a version number I can verify. The
> arpa_inet.h header is a difference and it's just what I'm using. It'll go
> away once the development is further along and code better established.
>
> I'm already sailing near the wind with a piece of paper I signed by
> including the code and saying as much as I have done! Nuff said?
>
> david
>
>
>
Re: cvs commit: apache-2.0/src/lib/apr/file_io/unix fileio.h readwrite.c
Posted by Greg Stein <gs...@lyra.org>.
On Mon, Jun 12, 2000 at 05:03:48PM +0100, David Reid wrote:
>...
> > Why do we check HAVE_ARPA_INET_H for including sys/socket.h?
>
> I'm not checking for the presence of arpa_inet.h but using it to
> differentiate between versions of BeOS.
This brings up a good point. Maybe there should be a check somewhere to
require a specific version of BeOS? From some of these changes, I'd guess
that Apache will no longer work on older versions of BeOS.
> This code is for a developmental version of BeOS with a new networking
> stack, and so far it doesn't have a version number I can verify. The
> arpa_inet.h header is a difference and it's just what I'm using. It'll go
> away once the development is further along and code better established.
Maybe create a symbol that you can use, so this is clearer? For example:
#ifdef BEOS
#if HAVE_ARPA_INET_H
#define BEOS_R46 /* or whatever it is you're working with :-) */
#else
#undef BEOS_R46
#endif
oh... your other note says you'll be changing this tomorrow. Whatever :-),
just figured to throw in a couple ideas.
Cheers,
-g
--
Greg Stein, http://www.lyra.org/
Re: cvs commit: apache-2.0/src/lib/apr/file_io/unix fileio.h readwrite.c
Posted by David Reid <dr...@jetnet.co.uk>.
> Why do we need BEOS as part of the check? just include it if it is
> available...
Fair comment. I didn't want to include files unless they're needed, and
AFAIK most unix's don't need socket.h to get fd_set defined.
>
>
> Why do we check HAVE_ARPA_INET_H for including sys/socket.h?
>
I'm not checking for the presence of arpa_inet.h but using it to
differentiate between versions of BeOS.
This code is for a developmental version of BeOS with a new networking
stack, and so far it doesn't have a version number I can verify. The
arpa_inet.h header is a difference and it's just what I'm using. It'll go
away once the development is further along and code better established.
I'm already sailing near the wind with a piece of paper I signed by
including the code and saying as much as I have done! Nuff said?
david
Re: cvs commit: apache-2.0/src/lib/apr/file_io/unix fileio.h readwrite.c
Posted by Jeff Trawick <tr...@bellsouth.net>.
> From: dreid@locus.apache.org
> Date: 12 Jun 2000 13:51:23 -0000
>
> dreid 00/06/12 06:51:22
>
> Modified: src/lib/apr/file_io/unix fileio.h readwrite.c
> Log:
> Another change to allow the newest version of BeOS to be used to it's
> full potential.
>
> Revision Changes Path
> 1.20 +3 -0 apache-2.0/src/lib/apr/file_io/unix/fileio.h
>
> Index: fileio.h
> ===================================================================
> RCS file: /home/cvs/apache-2.0/src/lib/apr/file_io/unix/fileio.h,v
> retrieving revision 1.19
> retrieving revision 1.20
> diff -u -r1.19 -r1.20
> --- fileio.h 2000/04/22 06:16:20 1.19
> +++ fileio.h 2000/06/12 13:51:22 1.20
> @@ -98,6 +98,9 @@
> #ifdef BEOS
> #include <kernel/OS.h>
> #endif
> +#if BEOS && HAVE_ARPA_INET_H
> +#include <sys/socket.h> /* for fd_set definitions */
> +#endif
Why do we need BEOS as part of the check? just include it if it is
available...
Why do we check HAVE_ARPA_INET_H for including sys/socket.h?
--
Jeff Trawick | trawick@ibm.net | PGP public key at web site:
http://www.geocities.com/SiliconValley/Park/9289/
Born in Roswell... married an alien...
Re: cvs commit: apache-2.0/src/lib/apr/file_io/unix fileio.h readwrite.c
Posted by Jeff Trawick <tr...@bellsouth.net>.
> From: dreid@locus.apache.org
> Date: 12 Jun 2000 13:51:23 -0000
>
> dreid 00/06/12 06:51:22
>
> Modified: src/lib/apr/file_io/unix fileio.h readwrite.c
> Log:
> Another change to allow the newest version of BeOS to be used to it's
> full potential.
>
> Revision Changes Path
> 1.20 +3 -0 apache-2.0/src/lib/apr/file_io/unix/fileio.h
>
> Index: fileio.h
> ===================================================================
> RCS file: /home/cvs/apache-2.0/src/lib/apr/file_io/unix/fileio.h,v
> retrieving revision 1.19
> retrieving revision 1.20
> diff -u -r1.19 -r1.20
> --- fileio.h 2000/04/22 06:16:20 1.19
> +++ fileio.h 2000/06/12 13:51:22 1.20
> @@ -98,6 +98,9 @@
> #ifdef BEOS
> #include <kernel/OS.h>
> #endif
> +#if BEOS && HAVE_ARPA_INET_H
> +#include <sys/socket.h> /* for fd_set definitions */
> +#endif
Why do we need BEOS as part of the check? just include it if it is
available...
Why do we check HAVE_ARPA_INET_H for including sys/socket.h?
--
Jeff Trawick | trawick@ibm.net | PGP public key at web site:
http://www.geocities.com/SiliconValley/Park/9289/
Born in Roswell... married an alien...