Does it work if you simply copy ext/standard/filestat.c from 4.2.0 over current CVS ? There has been only a little modification which at a brief glance shouldn't do anything bad .. but you never know.

For CVS with it's original filestat.c, do you get any compiler warnings about pointer mismatch ?

First the compiler warnings:

http://www.phpcrawler.de/news/newsdetail.php?id=33

(im going to put that in a "real" file this afternoon, but from here, I can't use FTP to put it on server that is accessable for everyone)

When copying filestat.c like you described, everything works fine. PHP compiles as expected (well there are some warnings, but it runs at least). The file-functions work too.

strip_slashes from my bug-testing-script still failes ( http://nohn.net/lalafarm/php4-test.zip )

When starting PHP it still gives some strange errors:

> ~/php4-200207292100_install/bin/php 
Unaligned access pid=20430 <php> va=0x14006387c pc=0x1201b2870 ra=0x1201b2864 inst=0xb4010000

Like described in Bugs #18558 #16360 #17449

reproduced on AIX 4.3.3.

Relevant compiler warnings:
/data/mdev/_src/php-HEAD/ext/standard/filestat.c: In function `php_stat':
/data/mdev/_src/php-HEAD/ext/standard/filestat.c:560: warning: passing arg 1 of `php_check_open_basedir' discar
ds qualifiers from pointer target type
/home/thies/devel/php4/ext/standard/url_scanner_ex.re: In function `php_url_scanner_output_handler':
/home/thies/devel/php4/ext/standard/url_scanner_ex.re:392: warning: passing arg 3 of `url_adapt_ext' from incom
patible pointer type
/data/mdev/_src/php-HEAD/sapi/cgi/cgi_main.c: In function `main':
/data/mdev/_src/php-HEAD/sapi/cgi/cgi_main.c:603: warning: passing arg 2 of `cfg_get_long' from incompatible po
inter type

Since 'man' means installing X and a whole lot of other stuff on AIX I don't have them available, but will get back with some links from the IBM docserver, later today.

changed this to be verfied by msopacua@idg.nl 

Now the Error-Log is at http://www.nohn.net/lalafarm/php4-test/error.log

can you test the patch that stef@php.net just sent to PHP-Dev and see if this solves the touch issues?  

I think, you meant this one (it's from sfox@php.net, not stef@php.net):

--- old_filestat.c      2002-07-28 23:35:50.000000000 +0000
+++ filestat.c  2002-07-30 17:16:48.000000000 +0000
@@ -481,9 +481,7 @@
        newtime = &newtimebuf;

        if (ac == 1 && zend_get_parameters_ex(1, &filename) != FAILURE) {
-#ifndef HAVE_UTIME_NULL
                newtime->modtime = newtime->actime = time(NULL);
-#endif
        } else if (ac == 2 && zend_get_parameters_ex(2, &filename, &filetime) != FAILURE) {
                convert_to_long_ex(filetime);
                newtime->actime = time(NULL);



----

and yes... It works fine. Tested with php4-200207301200 tested on Compaq Testdrive machine (a little bit slow, but it does what it should do). Full report here: http://www.nohn.net/lalafarm/php4-test-reports/4.3.0-dev-1028061991.xml

woo, that was unexpected :)  I thought I'd come across a separate win32 issue.

Does that mean I can close this one now?

msopacua@idg.nl reports Stef patch fixes the bug on AIX.

This bug has been fixed in CVS. You can grab a snapshot of the
CVS version at http://snaps.php.net/. In case this was a documentation 
problem, the fix will show up soon at http://www.php.net/manual/.
In case this was a PHP.net website problem, the change will show
up on the PHP.net site and on the mirror sites.
Thank you for the report, and for helping us make PHP better.

Sebastian agrees.