#26309 [NEW]: imageftbbox() requires the optional parameter

[kputnam at putnamcabinets dot com]

From: kputnam at putnamcabinets dot com
Operating system: Debian GNU/Linux (Sid)
PHP version:  5.0.0b2 (beta2)
PHP Bug Type: GD related
Bug description:  imageftbbox() requires the optional parameter

Description:

The documentation shows that the last parameter is optional (extrainfo)
but an error is generated when it is left out. Calling with an empty array
as the last parameter resolves the issue.

Reproduce code:
---
$box = imageftbbox($size,$angle,$this->fonts[$font],$text);

Actual result:
--
PHP Warning:  Wrong parameter count for imageftbbox()

-- 
Edit bug report at http://bugs.php.net/?id=26309&edit=1
-- 
Try a CVS snapshot (php4):  http://bugs.php.net/fix.php?id=26309&r=trysnapshot4
Try a CVS snapshot (php5):  http://bugs.php.net/fix.php?id=26309&r=trysnapshot5
Fixed in CVS:   http://bugs.php.net/fix.php?id=26309&r=fixedcvs
Fixed in release:   http://bugs.php.net/fix.php?id=26309&r=alreadyfixed
Need backtrace: http://bugs.php.net/fix.php?id=26309&r=needtrace
Try newer version:  http://bugs.php.net/fix.php?id=26309&r=oldversion
Not developer issue:http://bugs.php.net/fix.php?id=26309&r=support
Expected behavior:  http://bugs.php.net/fix.php?id=26309&r=notwrong
Not enough info:http://bugs.php.net/fix.php?id=26309&r=notenoughinfo
Submitted twice:http://bugs.php.net/fix.php?id=26309&r=submittedtwice
register_globals:   http://bugs.php.net/fix.php?id=26309&r=globals
PHP 3 support discontinued: http://bugs.php.net/fix.php?id=26309&r=php3
Daylight Savings:   http://bugs.php.net/fix.php?id=26309&r=dst
IIS Stability:  http://bugs.php.net/fix.php?id=26309&r=isapi
Install GNU Sed:http://bugs.php.net/fix.php?id=26309&r=gnused
Floating point limitations: http://bugs.php.net/fix.php?id=26309&r=float

#26304 [Com]: Unexpected data loss when opening dba file

[vorlon at debian dot org]

 ID:   26304
 Comment by:   vorlon at debian dot org
 Reported By:  vesely at tana dot it
 Status:   Assigned
 Bug Type: DBM/DBA related
 Operating System: Solaris
 PHP Version:  4.3.4
 Assigned To:  helly
 New Comment:

This bug has also been reported at .

The source of this behavior is quite clear -- 
ext/dba/dba.c:

case 'c':
modenr = DBA_CREAT;
lock_mode = (lock_flag & DBA_LOCK_CREAT) ? LOCK_EX : 0;
file_mode = "a+b";
if (!lock_mode || !lock_dbf) {
break;
}
/* When we lock the db file it will be created before the handler
 * even tries to open it, hence we must change to truncate mode.
 */
case 'n':
modenr = DBA_TRUNC;
lock_mode = (lock_flag & DBA_LOCK_TRUNC) ? LOCK_EX : 0;
file_mode = "w+b";
break;

So unless locking is explicitly disabled (or explicitly configured to
use an external lockfile), 'CREAT' mode results in automatic truncation
of the database?  What kind of sense does that make?

The behavior on the HEAD branch looks correct, but doesn't seem to
interact well with the 4.3 version of the code ("Driver initialization
failed for handler: db4: Bad file descriptor").  The current behavior
certainly is NOT correct, for db4.


Previous Comments:


[2003-11-18 12:59:58] vesely at tana dot it

Also, that correction around line 67 in dba_db4.c
(DBA_OPEN_FUNC) is bogus: if stat returns 0 you want
to say DB_UNKNOWN, since you cannot say it is DB_BTREE
when it was created by some other program.

To reproduce the bug you should create a DB with a different
type, e.g. in C if you just include db.h and then call
the dbm_open compatibility layer. The Sleepycat message
will then say "call implies an access method which is
inconsistent with previous calls."



[2003-11-18 12:09:46] vesely at tana dot it

Description:

Opening a file in 'c' mode (see example below)
truncates the file!! The docs say '"c" for read/write
access and database creation if it doesn't currently exist.'

I cannot understand that comment at line 658 in
ext/dba/dba.c, as it seems to imply that truncating
the database is necessary for locking it (??).

Reproduce code:
---
dba_open("important_data.db", "c", "db4");

Expected result:

open db (create if doesn't exist)

Actual result:
--
truncated db





-- 
Edit this bug report at http://bugs.php.net/?id=26304&edit=1

#26267 [Ver->Csd]: gmp_random() leaks memory and does not produce random numbers

[sniper]

 ID:   26267
 Updated by:   [EMAIL PROTECTED]
-Summary:  Memory Leaks
 Reported By:  cunha17 at uol dot com dot br
-Status:   Verified
+Status:   Closed
 Bug Type: Math related
-Operating System: *
+Operating System: RedHat Linux 9.0
-PHP Version:  5CVS, 4CVS
+PHP Version:  4CVS, 5CVS
 New Comment:

This bug has been fixed in CVS.

In case this was a PHP problem, snapshots of the sources are packaged
every three hours; this change will be in the next snapshot. You can
grab the snapshot 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 in short time.
 
Thank you for the report, and for helping us make PHP better.

gmp_random() didn't really produce random numbers either.
That is also fixed now.



Previous Comments:


[2003-11-15 16:06:07] cunha17 at uol dot com dot br

Description:

The piece of code below leaks memory.

Reproduce code:
---


Expected result:

Nothing.

Actual result:
--
/usr/src/zend2/php-src/ext/gmp/gmp.c(119) :  Freeing 0x405018B8 (8
bytes), script=-
Last leak repeated 1 time
/usr/src/zend2/php-src/ext/gmp/gmp.c(127) :  Freeing 0x40501878 (8
bytes), script=-
=== Total 3 memory leaks detected ===






-- 
Edit this bug report at http://bugs.php.net/?id=26267&edit=1

#21197 [Com]: socket_read() outputs error with PHP_NORMAL_READ

[faenr at yahoo dot it]

 ID:   21197
 Comment by:   faenr at yahoo dot it
 Reported By:  bool at boolsite dot net
 Status:   Verified
 Bug Type: Sockets related
 Operating System: win32 only
 PHP Version:  4.3.4-dev
 New Comment:

example from--
http://ca2.php.net/manual/en/ref.sockets.php

#!/usr/local/bin/php -q


my system-
windows xp, php 5.0.0b2

error-
C:\PHP>php server.php
PHP Warning:  socket_read() [function.socket-read
]: unable to read from socket [0]: Operazione completata.
 in C:\PHP\server.php on line 39
Content-type: text/html
X-Powered-By: PHP/5.0.0b2


Warning:  socket_read() [function.socket-r
ead]: unable to read from socket [0]: Operazione completata.
 in C:\PHP\server.php on line 39
socket_read() failed: reason: Funzione non corretta.

Translection
Operazione completata = operation completed
Funzione non corretta = uncorrect function

Guesssing---
The problem isn't fixed in php 5

thank you


Previous Comments:


[2003-08-26 02:00:58] bool at boolsite dot net

Ok, this is a short example : (a little echo server)

 Debut de la connexion...',"\r\n";
$EndTime=time()+15;
do{
$buffer=socket_read($MsgSock,1024,PHP_NORMAL_READ);
if($buffer===false) {
echo 'socket_read() a échoué :
',socket_strerror(socket_last_error($MsgSock)),"\r\n";
break;
}
elseif(!$buffer){
continue;
}
$buffer=trim($buffer);
echo '< ',$buffer,"\r\n";
if($buffer=='quit') {
break;
}

$back='You sent : ['.$buffer.']';

echo '> ',$back,"\r\n";
socket_write($MsgSock,$back."\r\n");
} while(time()<$EndTime);

@socket_close($MsgSock);
echo '=> End...',"\r\n";
}
}
socket_close($Sock);
?>



[2003-08-25 20:17:06] [EMAIL PROTECTED]

Please provide a complete but short example script that can be used to
reproduce this bug.




[2003-08-25 18:19:27] bool at boolsite dot net

It's not the same bugs than http://bugs.php.net/bug.php?id=21760 . Here
it's a Win32 PHP version bug only, since the PHP 4.3.0 version, and
there is always the bug in the last version 4.3.3.

socket_read() always return FALSE, and display this error : "erreur [0]
avec socket_read : OpÚration rÚussie."
(in english it should be "error [0] with socket_read : operation
completed.")

maybe it's because of the socket_select() before the socket_read() ?


PS : it affects all my codes, but only with PHP 4.3.x Win32 version. On
unix or with PHP 4.2.3 all works fine.



[2003-06-29 21:42:30] [EMAIL PROTECTED]

Please do not submit the same bug more than once. An existing
bug report already describes this very problem. Even if you feel
that your issue is somewhat different, the resolution is likely
to be the same. Because of this, we hope you add your comments
to the existing bug instead.

Thank you for your interest in PHP.

The active discussion regarding this bug appears to be here
http://bugs.php.net/bug.php?id=21760

let's keep it all in one place.



[2002-12-26 20:39:22] [EMAIL PROTECTED]

If you omit the third parameter to socket_read() it seems to work fine.
However, adding PHP_NORMAL_READ causes error as described in the bug
report.



The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
http://bugs.php.net/21197

-- 
Edit this bug report at http://bugs.php.net/?id=21197&edit=1

#26285 [Opn->Asn]: escapeshellarg uses wrong quotes on windows

[iliaa]

 ID:   26285
 Updated by:   [EMAIL PROTECTED]
 Reported By:  jb at ez dot no
-Status:   Open
+Status:   Assigned
 Bug Type: *General Issues
 Operating System: Windows
 PHP Version:  4.3.3
-Assigned To:  
+Assigned To:  iliaa


Previous Comments:


[2003-11-17 06:43:20] jb at ez dot no

Description:

The escapeshellarg function in PHP creates wrongly escaped text on 
windows. It seems it always uses the single-quote (') for escaping the

strings, however on windows the correct way to escape is using the 
double-quote (") character. 
 
The same problems seems to be in the escapeshellcmd function as 
well. 

Reproduce code:
---
escapeshellarg( "-arg value" );

Expected result:

"-arg value" 

Actual result:
--
'-arg value' 





-- 
Edit this bug report at http://bugs.php.net/?id=26285&edit=1

#13961 [Ver->Bgs]: some characters in server variable names are silently changed

[iliaa]

 ID:   13961
 Updated by:   [EMAIL PROTECTED]
 Reported By:  lampa at fee dot vutbr dot cz
-Status:   Verified
+Status:   Bogus
 Bug Type: Apache related
 Operating System: any
 PHP Version:  4CVS, 5CVS
 Assigned To:  derick
 New Comment:

This is infact fixed php_register_variable() is actually a wrapper
around php_register_variable_safe() which always makes a copy of the
original before passing it to php_register_variable_ex().


Previous Comments:


[2003-11-14 07:06:13] lampa at fee dot vutbr dot cz

Still not fixed in 4.3.4



[2003-11-11 18:15:11] daniele at orlandi dot com

This bug is still present in php 4.3.4 and may be harmful since all the
BrowserMatch functionality to workaround browser bugs in Apache is
essentially disabled.

As a proof of concept i patched sapi/apache2handler/sapi_apache2.c
(apache2filter is probably affected too) and the problem went away.

Note that the patch may not be perfect as I don't know how Apache and
PHP work internally very well.

--- php-4.3.4/sapi/apache2handler/sapi_apache2.c2003-10-02
05:24:43.0 +0200
+++ php-4.3.4-patched/sapi/apache2handler/sapi_apache2.c   
2003-11-11 23:52:06.0 +0100
@@ -227,9 +227,14 @@
char *key, *val;
zval **path_translated_zv;
  
+   char *t;
+
APR_ARRAY_FOREACH_OPEN(arr, key, val)
if (!val) val = empty_string;
-   php_register_variable(key, val, track_vars_array
TSRMLS_CC);
+
+   t = estrndup(key, strlen(key));
+   php_register_variable(t, val, track_vars_array
TSRMLS_CC);
+   efree(t);
APR_ARRAY_FOREACH_CLOSE()



[2001-12-11 09:57:36] lampa at fee dot vutbr dot cz

Not fixed in 4.1.0. Why? To be clear, one call is neccessary:

for (i = 0; i < arr->nelts; i++) {
char *val,*key;

if (elts[i].val) {
val = elts[i].val;
} else {
val = empty_string;
}
key = estrdup(elts[i].key); /* HERE */
php_register_variable(key, val, track_vars_array 
ELS_CC PLS_CC)
;
}




[2001-11-07 04:33:03] [EMAIL PROTECTED]

This is not okay, PHP should not change the original key here.
Checking it out.



[2001-11-07 01:56:30] lampa at fee dot vutbr dot cz

I don't think that FAQ solves that problem.
Look at the source code of Apache server. There
are several tests of the variable "force-response-1.0" 
there. The problem is not that php code variable
is $force-response-1_0, that's OK, but the real
problem is that apache variable name in r->subprocess_env
is changed too. That's side effect and not pleasent.



The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
http://bugs.php.net/13961

-- 
Edit this bug report at http://bugs.php.net/?id=13961&edit=1

#26304 [Opn->Asn]: Unexpected data loss when opening dba file

[iliaa]

 ID:   26304
 Updated by:   [EMAIL PROTECTED]
 Reported By:  vesely at tana dot it
-Status:   Open
+Status:   Assigned
 Bug Type: DBM/DBA related
 Operating System: Solaris
 PHP Version:  4.3.4
-Assigned To:  
+Assigned To:  helly


Previous Comments:


[2003-11-18 12:59:58] vesely at tana dot it

Also, that correction around line 67 in dba_db4.c
(DBA_OPEN_FUNC) is bogus: if stat returns 0 you want
to say DB_UNKNOWN, since you cannot say it is DB_BTREE
when it was created by some other program.

To reproduce the bug you should create a DB with a different
type, e.g. in C if you just include db.h and then call
the dbm_open compatibility layer. The Sleepycat message
will then say "call implies an access method which is
inconsistent with previous calls."



[2003-11-18 12:09:46] vesely at tana dot it

Description:

Opening a file in 'c' mode (see example below)
truncates the file!! The docs say '"c" for read/write
access and database creation if it doesn't currently exist.'

I cannot understand that comment at line 658 in
ext/dba/dba.c, as it seems to imply that truncating
the database is necessary for locking it (??).

Reproduce code:
---
dba_open("important_data.db", "c", "db4");

Expected result:

open db (create if doesn't exist)

Actual result:
--
truncated db





-- 
Edit this bug report at http://bugs.php.net/?id=26304&edit=1

#26281 [Ver->Csd]: switch on string offset crashes w/bus error

[iliaa]

 ID:   26281
 Updated by:   [EMAIL PROTECTED]
 Reported By:  tater at potatoe dot com
-Status:   Verified
+Status:   Closed
 Bug Type: Strings related
 Operating System: OS X 10.2
 PHP Version:  5CVS-2003-11-16 (dev)
 New Comment:

This bug has been fixed in CVS.

In case this was a PHP problem, snapshots of the sources are packaged
every three hours; this change will be in the next snapshot. You can
grab the snapshot 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 in short time.
 
Thank you for the report, and for helping us make PHP better.




Previous Comments:


[2003-11-16 23:06:18] tater at potatoe dot com

Description:

a switch() statement testing a string offset, like 
'switch($string{0})', crashes after executing the
matching case or default block. this only happens
if there is at least one case. workarounds: assign
offset to another variable first, or put it in 
quotes like "{$string{0}}".

This breaks a lot of PEAR code, btw.

Reproduce code:
---
$x = 'abc';
switch ($x{0}) { default: print "default - works\n"; }
switch ($x{0}) { case 'a': print "case 'a' - bus error\n"; }
print "done\n";

Expected result:

default - works
case 'a' - bus error
done

Actual result:
--
default - works
case 'a' - bus error
Bus error

backtrace:
(gdb) run bug.php
Starting program: /usr/local/book/php/bin/php bug.php
[Switching to process 5410 thread 0xb03]
Reading symbols for shared libraries . done
Reading symbols for shared libraries  done
default - works
case 'a' - bus error

Program received signal EXC_BAD_ACCESS, Could not access memory.
0x00229f8c in _zval_dtor (zvalue=0xbb4c10, __zend_filename=0x2c2a24
"/Users/tater/book/php-src/Zend/zend_execute.c", __zend_lineno=203) at
/Users/tater/book/php-src/Zend/zend_variables.c:44
44  CHECK_ZVAL_STRING_REL(zvalue);
(gdb)





-- 
Edit this bug report at http://bugs.php.net/?id=26281&edit=1

#26267 [Opn->Ver]: Memory Leaks

[sniper]

 ID:   26267
 Updated by:   [EMAIL PROTECTED]
 Reported By:  cunha17 at uol dot com dot br
-Status:   Open
+Status:   Verified
 Bug Type: Math related
-Operating System: RedHat Linux 9.0
+Operating System: *
-PHP Version:  5CVS-2003-11-15 (dev)
+PHP Version:  5CVS, 4CVS


Previous Comments:


[2003-11-15 16:06:07] cunha17 at uol dot com dot br

Description:

The piece of code below leaks memory.

Reproduce code:
---


Expected result:

Nothing.

Actual result:
--
/usr/src/zend2/php-src/ext/gmp/gmp.c(119) :  Freeing 0x405018B8 (8
bytes), script=-
Last leak repeated 1 time
/usr/src/zend2/php-src/ext/gmp/gmp.c(127) :  Freeing 0x40501878 (8
bytes), script=-
=== Total 3 memory leaks detected ===






-- 
Edit this bug report at http://bugs.php.net/?id=26267&edit=1

#26275 [Opn->Ver]: Server Crash when throw exception

[sniper]

 ID:   26275
 Updated by:   [EMAIL PROTECTED]
 Reported By:  php dot bug at stre dot it
-Status:   Open
+Status:   Verified
 Bug Type: Zend Engine 2 problem
 Operating System: win32 nt5.1
-PHP Version:  5CVS-2003-11-16 (dev)
+PHP Version:  5CVS-2003-11-19 (dev)
 New Comment:

#0  0x08318170 in _zval_ptr_dtor (zval_ptr=0x40e33bf8, 
__zend_filename=0x856f580
"/usr/src/web/php/php5/Zend/zend_execute.h", __zend_lineno=122)
at /usr/src/web/php/php5/Zend/zend_execute_API.c:352
#1  0x0834e4cb in zend_ptr_stack_clear_multiple () at
zend_execute.h:122
#2  0x08349a4e in zend_do_fcall_common_helper (execute_data=0xbfffd7b0,
op_array=0x40e428b4)
at /usr/src/web/php/php5/Zend/zend_execute.c:2644
#3  0x08349c57 in zend_do_fcall_handler (execute_data=0xbfffd7b0,
op_array=0x40e428b4)
at /usr/src/web/php/php5/Zend/zend_execute.c:2696
#4  0x0834571f in execute (op_array=0x40e428b4) at
/usr/src/web/php/php5/Zend/zend_execute.c:1271
#5  0x08324148 in zend_execute_scripts (type=8, retval=0x0,
file_count=3) at /usr/src/web/php/php5/Zend/zend.c:1016
#6  0x082e303a in php_execute_script (primary_file=0xbbb0) at
/usr/src/web/php/php5/main/main.c:1622
#7  0x0835d87b in main (argc=2, argv=0xbc44) at
/usr/src/web/php/php5/sapi/cli/php_cli.c:910



Previous Comments:


[2003-11-16 13:20:35] php dot bug at stre dot it

sorry short version work, because there is first a standalone badfunc
call, this version dont work



i think is a failure by building stack trace



[2003-11-16 12:20:32] php dot bug at stre dot it

Description:

this code below bring my server to crash

Reproduce code:
---
getMessage();
}

?>

Expected result:

no response from server, must restart service






-- 
Edit this bug report at http://bugs.php.net/?id=26275&edit=1

#26263 [Fbk->Opn]: Apache doesn't start up wiht LoadModule phpX_module .. in config

[peoyli at bredband dot net]

 ID:   26263
 User updated by:  peoyli at bredband dot net
 Reported By:  peoyli at bredband dot net
-Status:   Feedback
+Status:   Open
 Bug Type: Apache2 related
 Operating System: OpenBSD 3.4
 PHP Version:  Irrelevant
 New Comment:

I'm using just one of the sample php.ini's (php.ini-recommended). No
enabled extensions there, what I can see.


Previous Comments:


[2003-11-18 17:13:06] [EMAIL PROTECTED]

Are you loading any modules and/or extensions via php.ini?



[2003-11-18 02:00:56] peoyli at bredband dot net

PHP5 + MySQL 4.1 core dumps when calling mysql related functions
(mysql_connect ...). I will try the internal support for mysql 4.0.16
or older.



[2003-11-16 15:52:47] peoyli at bredband dot net

One other note:

I DID get it running using the internal mysql support (no path given to
--with-mysql) after some hassle. I also now have PHP 5 running together
with mysql 4.1.0 alpha.

The problem seems to have something to do with mysql 4.0.16 (haven't
tried earlier versions), although, as I mentioned before, there is not
problem with mysql 4.0.16 by itself.



[2003-11-16 15:48:40] peoyli at bredband dot net

no output using 'apachectl configtest'. Seems like httpd crasches as
soon as the LoadModule line is appended to the configuration.



[2003-11-16 13:15:23] [EMAIL PROTECTED]

Have you tried running:
apachectl configtest ?



The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
http://bugs.php.net/26263

-- 
Edit this bug report at http://bugs.php.net/?id=26263&edit=1

#26308 [Opn->Bgs]: php extension not loaded

[iliaa]

 ID:   26308
 Updated by:   [EMAIL PROTECTED]
 Reported By:  muhlig at us dot edu dot pl
-Status:   Open
+Status:   Bogus
 Bug Type: CGI related
 Operating System: Solaris 8
 PHP Version:  4.3.4
 New Comment:

Thank you for taking the time to write to us, but this is not
a bug. Please double-check the documentation available at
http://www.php.net/manual/ and the instructions on how to report
a bug at http://bugs.php.net/how-to-report.php

BTW mmcache won't work with CGI or CLI.

You should be using:
extension="mmcache.so"


Previous Comments:


[2003-11-18 17:19:50] muhlig at us dot edu dot pl

Description:

php extension is not loaded while running as cli binary. 

from php.ini:

  extension="/usr/local/lib/php/extensions/mmcache.so"

it can be seen "./" is prepended to "extension" path, ie. web module
wants absolute path while cli binary wants relative path, which is
inconsistent.

it is also possible extension should NOT be loaded at all while running
as cli binary.

Reproduce code:
---
N/A

Expected result:

# php -v
PHP 4.3.4 (cli) (built: Nov  5 2003 09:08:10)
Copyright (c) 1997-2003 The PHP Group
Zend Engine v1.3.0, Copyright (c) 1998-2003 Zend Technologies
# 


Actual result:
--
# php -v
PHP Warning:  Unknown(): Unable to load dynamic library
'.//usr/local/lib/php/extensions/mmcache.so' - ld.so.1: php: fatal:
.//usr/local/lib/php/extensions/mmcache.so: open failed: No such file
or directory in Unknown on line 0
PHP 4.3.4 (cli) (built: Nov  5 2003 09:08:10)
Copyright (c) 1997-2003 The PHP Group
Zend Engine v1.3.0, Copyright (c) 1998-2003 Zend Technologies
# 






-- 
Edit this bug report at http://bugs.php.net/?id=26308&edit=1

#26308 [NEW]: php extension not loaded

[muhlig at us dot edu dot pl]

From: muhlig at us dot edu dot pl
Operating system: Solaris 8
PHP version:  4.3.4
PHP Bug Type: CGI related
Bug description:  php extension not loaded

Description:

php extension is not loaded while running as cli binary. 

from php.ini:

  extension="/usr/local/lib/php/extensions/mmcache.so"

it can be seen "./" is prepended to "extension" path, ie. web module wants
absolute path while cli binary wants relative path, which is
inconsistent.

it is also possible extension should NOT be loaded at all while running as
cli binary.

Reproduce code:
---
N/A

Expected result:

# php -v
PHP 4.3.4 (cli) (built: Nov  5 2003 09:08:10)
Copyright (c) 1997-2003 The PHP Group
Zend Engine v1.3.0, Copyright (c) 1998-2003 Zend Technologies
# 


Actual result:
--
# php -v
PHP Warning:  Unknown(): Unable to load dynamic library
'.//usr/local/lib/php/extensions/mmcache.so' - ld.so.1: php: fatal:
.//usr/local/lib/php/extensions/mmcache.so: open failed: No such file or
directory in Unknown on line 0
PHP 4.3.4 (cli) (built: Nov  5 2003 09:08:10)
Copyright (c) 1997-2003 The PHP Group
Zend Engine v1.3.0, Copyright (c) 1998-2003 Zend Technologies
# 


-- 
Edit bug report at http://bugs.php.net/?id=26308&edit=1
-- 
Try a CVS snapshot (php4):  http://bugs.php.net/fix.php?id=26308&r=trysnapshot4
Try a CVS snapshot (php5):  http://bugs.php.net/fix.php?id=26308&r=trysnapshot5
Fixed in CVS:   http://bugs.php.net/fix.php?id=26308&r=fixedcvs
Fixed in release:   http://bugs.php.net/fix.php?id=26308&r=alreadyfixed
Need backtrace: http://bugs.php.net/fix.php?id=26308&r=needtrace
Try newer version:  http://bugs.php.net/fix.php?id=26308&r=oldversion
Not developer issue:http://bugs.php.net/fix.php?id=26308&r=support
Expected behavior:  http://bugs.php.net/fix.php?id=26308&r=notwrong
Not enough info:http://bugs.php.net/fix.php?id=26308&r=notenoughinfo
Submitted twice:http://bugs.php.net/fix.php?id=26308&r=submittedtwice
register_globals:   http://bugs.php.net/fix.php?id=26308&r=globals
PHP 3 support discontinued: http://bugs.php.net/fix.php?id=26308&r=php3
Daylight Savings:   http://bugs.php.net/fix.php?id=26308&r=dst
IIS Stability:  http://bugs.php.net/fix.php?id=26308&r=isapi
Install GNU Sed:http://bugs.php.net/fix.php?id=26308&r=gnused
Floating point limitations: http://bugs.php.net/fix.php?id=26308&r=float

#25597 [NoF->Opn]: fastcgi + pfsockopen = crash

[dietrich dot ayala at foundstone dot com]

 ID:   25597
 User updated by:  dietrich dot ayala at foundstone dot com
 Reported By:  dietrich dot ayala at foundstone dot com
-Status:   No Feedback
+Status:   Open
 Bug Type: CGI related
 Operating System: windows 2000
 PHP Version:  4.3.3
 New Comment:

Ilia,

was there an attempt to fix it, or is this an automated request to try
lates cvs? we're you able to repro using the repro script provided? wez
furlong said he's going to look into this.

thanks!

dietrich


Previous Comments:


[2003-11-17 18:15:50] [EMAIL PROTECTED]

No feedback was provided. The bug is being suspended because
we assume that you are no longer experiencing the problem.
If this is not the case and you are able to provide the
information that was requested earlier, please do so and
change the status of the bug back to "Open". Thank you.





[2003-11-12 14:31:59] [EMAIL PROTECTED]

Please try using this CVS snapshot:

  http://snaps.php.net/php4-STABLE-latest.tar.gz
 
For Windows:
 
  http://snaps.php.net/win32/php4-win32-STABLE-latest.zip





[2003-09-24 13:42:08] dietrich dot ayala at foundstone dot com

Thanks Edin.

I've updated to the latest stable Windows build, and applied Edin's
patched php4ts.dll.

The example script I gave still reproduces the crash.

The text of the Application Error alert is:
The instruction at "0x100108b4" referenced memory at "0x'. The
memory could not be "read".

Then, as before, I get the FastCGI error:
Server Error, unable to connect to fastcgi server.

Thanks for any help in fixing this!



[2003-09-24 05:15:51] [EMAIL PROTECTED]

This is the latest ssl enabled build:

ftp://ftp.proventum.net/pub/php/win32/misc/openssl/php-4.3.4-dev-ssl.zip




[2003-09-19 13:58:54] dietrich dot ayala at foundstone dot com

My apologies for leaving out an important detail:

This only occurrs when using SSL. I cannot reproduce this issue using
unencrypted communication.

Since there still exists a problem with dynamically linking OpenSSL in
the 4.x tree, I have asked EdinK to compile a new php4ts.dll with
OpenSSL. Once that's available, I'll test using the latest.

Thanks!



The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
http://bugs.php.net/25597

-- 
Edit this bug report at http://bugs.php.net/?id=25597&edit=1

#26263 [Opn->Fbk]: Apache doesn't start up wiht LoadModule phpX_module .. in config

[iliaa]

 ID:   26263
 Updated by:   [EMAIL PROTECTED]
 Reported By:  peoyli at bredband dot net
-Status:   Open
+Status:   Feedback
 Bug Type: Apache2 related
 Operating System: OpenBSD 3.4
 PHP Version:  Irrelevant
 New Comment:

Are you loading any modules and/or extensions via php.ini?


Previous Comments:


[2003-11-18 02:00:56] peoyli at bredband dot net

PHP5 + MySQL 4.1 core dumps when calling mysql related functions
(mysql_connect ...). I will try the internal support for mysql 4.0.16
or older.



[2003-11-16 15:52:47] peoyli at bredband dot net

One other note:

I DID get it running using the internal mysql support (no path given to
--with-mysql) after some hassle. I also now have PHP 5 running together
with mysql 4.1.0 alpha.

The problem seems to have something to do with mysql 4.0.16 (haven't
tried earlier versions), although, as I mentioned before, there is not
problem with mysql 4.0.16 by itself.



[2003-11-16 15:48:40] peoyli at bredband dot net

no output using 'apachectl configtest'. Seems like httpd crasches as
soon as the LoadModule line is appended to the configuration.



[2003-11-16 13:15:23] [EMAIL PROTECTED]

Have you tried running:
apachectl configtest ?



[2003-11-15 13:50:14] peoyli at bredband dot net

the apachectl command gives me no error at all, it just doesn't start
apache, and if I happened to have a previous started httpd (without the
LoadModule line for PHP), I can't use apachectl to stop it either.



The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
http://bugs.php.net/26263

-- 
Edit this bug report at http://bugs.php.net/?id=26263&edit=1

#25441 [Opn->Csd]: public properties of parent class not accessible

[sniper]

 ID:   25441
 Updated by:   [EMAIL PROTECTED]
 Reported By:  tater at potatoe dot com
-Status:   Open
+Status:   Closed
 Bug Type: Zend Engine 2 problem
 Operating System: OS X 10.2
 PHP Version:  5CVS-2003-09-08 (dev)
 New Comment:

Works as expected with latest CVS. Seems to be fixed..



Previous Comments:


[2003-09-08 18:21:54] tater at potatoe dot com

Description:

If a child class directly references a property defined in
the parent class, it no longer inherits it.

Reproduce code:
---
x);
}
}
$base = new base;
print_r($base);
$foo = new foo;
print_r($foo);
$bar = new bar;
print_r($bar);
$bar->bartest();
?>

Expected result:

base Object
(
[x] => 1
)
foo Object
(
[x] => 1
)
bar Object
(
[x] => 1
)
int(1)

Actual result:
--
base Object
(
[x] => 1
)
foo Object
(
[x] => 1
)
bar Object
(
)

Notice: Undefined property:  bar::$x in /path/bug.php on line 12
NULL





-- 
Edit this bug report at http://bugs.php.net/?id=25441&edit=1

#25541 [Opn->Ver]: destructor won't be called

[sniper]

 ID:   25541
 Updated by:   [EMAIL PROTECTED]
 Reported By:  ich at christoph-emonds dot de
-Status:   Open
+Status:   Verified
 Bug Type: Zend Engine 2 problem
 Operating System: Gentoo Linux 2.4.20
-PHP Version:  5CVS-2003-09-15 (dev)
+PHP Version:  5CVS-2003-11-18 (dev)
 New Comment:

Construct A!
Construct B!
/usr/src/web/php/php5/Zend/zend_hash.c(236) :  Freeing 0x40E45A98 (41
bytes), script=t.php
Last leak repeated 1 time
/usr/src/web/php/php5/Zend/zend_execute.c(3098) :  Freeing 0x40E45A34
(44 bytes), script=t.php
/usr/src/web/php/php5/Zend/zend_API.c(713) : Actual location (location
was relayed)
Last leak repeated 1 time
/usr/src/web/php/php5/Zend/zend_objects.c(106) :  Freeing 0x40E459F0
(12 bytes), script=t.php
Last leak repeated 1 time
/usr/src/web/php/php5/Zend/zend_execute.c(3097) :  Freeing 0x40E459AC
(16 bytes), script=t.php
Last leak repeated 1 time
/usr/src/web/php/php5/Zend/zend_API.c(714) :  Freeing 0x40E455E8 (32
bytes), script=t.php
/usr/src/web/php/php5/Zend/zend_hash.c(157) : Actual location (location
was relayed)
Last leak repeated 1 time
=== Total 10 memory leaks detected ===



Previous Comments:


[2003-09-15 06:35:27] ich at christoph-emonds dot de

Description:

If you create two objects which contain a reference to the  
other object, none of both __destruct() functions will be 
called.  
  
Configure: './configure' '--with-mysql' '--with-xsl'  
'--with-sqlite' '--enable-sockets' '--with-mysql-sock'  
'--with-gd' '--with-dom' '--with-apxs2=/usr/sbin/apxs2'  
'--with-zlib'   

Reproduce code:
---
class A {
function __construct()  {
print "Construct A!";
$this->other = new B( $this );
}
function __destruct() {
print "Destruct A!";
}
}

class B {
function __construct( A $obj ) {
print "Construct B!";
$this->other = $obj;
}
function __destruct() {
print "Destruct B!";
}
}
new A();

Expected result:

Construct A!Construct B!Destruct A!Destruct B! 

Actual result:
--
Construct A!Construct B! 





-- 
Edit this bug report at http://bugs.php.net/?id=25541&edit=1

#25652 [Opn->Ver]: Calling Global functions dynamically fails from Class scope

[sniper]

 ID:   25652
 Updated by:   [EMAIL PROTECTED]
 Reported By:  [EMAIL PROTECTED]
-Status:   Open
+Status:   Verified
 Bug Type: Zend Engine 2 problem
 Operating System: Redhat 8
 PHP Version:  5CVS-2003-09-25 (dev)
 New Comment:

Works with PHP 4 fine, not with PHP 5..



Previous Comments:


[2003-09-25 06:44:37] [EMAIL PROTECTED]

Opps. Bad code example -- this is the code:

arr[0]('testvalue');

}

}

$a = new foo();
$a->bar();

?>



[2003-09-25 06:29:07] [EMAIL PROTECTED]

Description:

ZE2 is bailing on with a fatal error when calling a method dyanmically
using a properity of the class (an array) as the dynamic function name.
I.e.

$this->myarr[0]("function paramer");

returns:

Fatal error: Method name must be a string in

It worked in PHP4, so I don't know if this is intended as invalid in
PHP5, or if it is a bug...


Reproduce code:
---
"; var_dump($a); echo "";
$a->arr[0]("testvalue");

?>

Expected result:

Test testvalue

Actual result:
--
Fatal error: Method name must be a string in...





-- 
Edit this bug report at http://bugs.php.net/?id=25652&edit=1

#26307 [Opn->Bgs]: Variable reference assignments in functions

[sniper]

 ID:   26307
 Updated by:   [EMAIL PROTECTED]
 Reported By:  soywiz at hotmail dot com
-Status:   Open
+Status:   Bogus
 Bug Type: Variables related
 Operating System: Windows
 PHP Version:  4.3.4
 New Comment:

RTFM:

  http://www.php.net/manual/en/language.references.arent.php



Previous Comments:


[2003-11-18 15:58:01] soywiz at hotmail dot com

Description:

Within a function a reference to another variable cannot be assigned to
its to a global variable.

I also proved this example in php 4.3.3

Sorry by my bad english

In spanish: Dentro de una función no se le puede asignar a una variable
global una referencia a otra variable (también global).

He probado este ejemplo en php 4.3.3 y php 4.3.4

Reproduce code:
---


Expected result:

aa

Actual result:
--
ba





-- 
Edit this bug report at http://bugs.php.net/?id=26307&edit=1

#26307 [NEW]: Variable reference assignments in functions

[soywiz at hotmail dot com]

From: soywiz at hotmail dot com
Operating system: Windows
PHP version:  4.3.4
PHP Bug Type: Variables related
Bug description:  Variable reference assignments in functions

Description:

Within a function a reference to another variable cannot be assigned to
its to a global variable.

I also proved this example in php 4.3.3

Sorry by my bad english

In spanish: Dentro de una función no se le puede asignar a una variable
global una referencia a otra variable (también global).

He probado este ejemplo en php 4.3.3 y php 4.3.4

Reproduce code:
---


Expected result:

aa

Actual result:
--
ba

-- 
Edit bug report at http://bugs.php.net/?id=26307&edit=1
-- 
Try a CVS snapshot (php4):  http://bugs.php.net/fix.php?id=26307&r=trysnapshot4
Try a CVS snapshot (php5):  http://bugs.php.net/fix.php?id=26307&r=trysnapshot5
Fixed in CVS:   http://bugs.php.net/fix.php?id=26307&r=fixedcvs
Fixed in release:   http://bugs.php.net/fix.php?id=26307&r=alreadyfixed
Need backtrace: http://bugs.php.net/fix.php?id=26307&r=needtrace
Try newer version:  http://bugs.php.net/fix.php?id=26307&r=oldversion
Not developer issue:http://bugs.php.net/fix.php?id=26307&r=support
Expected behavior:  http://bugs.php.net/fix.php?id=26307&r=notwrong
Not enough info:http://bugs.php.net/fix.php?id=26307&r=notenoughinfo
Submitted twice:http://bugs.php.net/fix.php?id=26307&r=submittedtwice
register_globals:   http://bugs.php.net/fix.php?id=26307&r=globals
PHP 3 support discontinued: http://bugs.php.net/fix.php?id=26307&r=php3
Daylight Savings:   http://bugs.php.net/fix.php?id=26307&r=dst
IIS Stability:  http://bugs.php.net/fix.php?id=26307&r=isapi
Install GNU Sed:http://bugs.php.net/fix.php?id=26307&r=gnused
Floating point limitations: http://bugs.php.net/fix.php?id=26307&r=float

#25377 [Opn->Bgs]: Class variables can be added out of class definition

[sniper]

 ID:   25377
 Updated by:   [EMAIL PROTECTED]
 Reported By:  forseti at oak dot rpg dot pl
-Status:   Open
+Status:   Bogus
 Bug Type: Class/Object related
 Operating System: Windows 98 SE
 PHP Version:  5CVS-2003-09-03 (dev)
 New Comment:

This is actually a feature, not bug.



Previous Comments:


[2003-09-03 04:16:27] forseti at oak dot rpg dot pl

Description:

Class variables can be added freely out of class declaration context.
This can be done by simply assigning a value to existing object's
non-existing variable. Resulting modified object remains of his old
type.

Reproduce code:
---
b = 'bar';
$test3 = new Test;
echo 'test1: ';print_r($test1);echo '';
echo 'test2: ';print_r($test2);echo '';
echo 'test3: ';print_r($test3);echo '';
$hint = new HintTest($test2);
?>

Expected result:

Adding new class variables this way shouldn't be possible because
modified object is no longer of the same type. 
And as last line shows it is treated by engine as such.

Actual result:
--
Modified object is nevertheless treated as if it was of Test type.





-- 
Edit this bug report at http://bugs.php.net/?id=25377&edit=1

#25957 [Opn->Fbk]: bad one? mixingup of array / $this-> in memory

[sniper]

 ID:   25957
 Updated by:   [EMAIL PROTECTED]
 Reported By:  sven at leon dot uia dot mx
-Status:   Open
+Status:   Feedback
 Bug Type: Zend Engine 2 problem
 Operating System: All (redhat 90)
 PHP Version:  5.0.0b1 (beta1)
 New Comment:

Provide a complete (preferrably short as possible) example script.



Previous Comments:


[2003-10-22 17:32:16] sven at leon dot uia dot mx

Description:

In a class I send an array $test as a parameter to a protected
function. Then, if I try to access that array in any way in that
function, suddenly the $this object contains the data of that array
$test (!)

Reproduce code:
---
Sorry, cant make a short-and-clear reproducing code, but here I have a
small code sample that actually caused the problem: (note that
$this->params["source"] _IS_ an array)

protected function get_body($css){

echo "B0 (";
print_r($this);
echo ")";
flush();

  $retval="";
  while(list($key, $cell_data)=each($this->params["source"])){

  echo "B1 (";
  print_r($this);
  echo ")";
  flush();



Expected result:

I will try to get a .zip together with the classes that seem to cause
this problem.

Actual result:
--
B0 (table_key_value_a Object ( [params:protected] => Array ( [source]
=> Array ( [crc] => mBjsS50UkhAnxjWp [sid] => 75 [test_site] =>
75,mBjsS50UkhAnxjWp ) [col headers] => Array ( [0] => Key [1] => Value
) [table header] => HTTP VARIABLES [css prefix] => debug [table header
extra] => align='center' ) ) )

B1 (table_key_value_a Object ( [cell_data] => mBjsS50UkhAnxjWp [key] =>
crc ) )





-- 
Edit this bug report at http://bugs.php.net/?id=25957&edit=1

#26121 [Opn->WFx]: Object defined before object - error

[sniper]

 ID:   26121
 Updated by:   [EMAIL PROTECTED]
 Reported By:  closer at netnitco dot net
-Status:   Open
+Status:   Wont fix
 Bug Type: Zend Engine 2 problem
 Operating System: Redhat Linux 9.0 / Apache 2.0.46
 PHP Version:  5.0.0b2 (beta2)
 New Comment:

This is expected and correct behaviour.



Previous Comments:


[2003-11-04 14:16:22] closer at netnitco dot net

Description:

With PHP 4 you could create the object variable before the actual
object was defined. This doesnt seem to be true in 5. If you move lines
2 and 3 of the code to the bottom the script oes work, but this does
break backward compatibility.

Reproduce code:
---
hello ();

class new_class
{
function hello ()
{
echo "Hello";
}
}
?>

Expected result:

Hello

Actual result:
--
Fatal error: Class 'new_class' not found in
/usr/local/apache2/htdocs/scripts/class_test.php on line 2





-- 
Edit this bug report at http://bugs.php.net/?id=26121&edit=1

#26248 [Opn->Bgs]: Can't compile PHP (both 4 and 5) with IMAP

[sniper]

 ID:   26248
 Updated by:   [EMAIL PROTECTED]
 Reported By:  lists at nutshell dot nu
-Status:   Open
+Status:   Bogus
 Bug Type: Compile Failure
 Operating System: Linux (Mandrake 9.2)
 PHP Version:  5.0.0b2 (beta2)
 New Comment:

Works fine for me.



Previous Comments:


[2003-11-18 08:14:12] lists at nutshell dot nu

A few notes to my last comment.

To be able to compile php (4/5) you either have to compile imap from a
tar ball or change the imap.spec so it compiles the libc-client-PHP4.a
with kerberos (it only compiles with ssl default).

My experience here is that it is impossible to compile imap support for
php unless you have kerberos. In my original post I show that I tried
with both --without-kerberos and --without-imap-ssl, but the
compilation failed due to that libc-client had no kerberos support.



[2003-11-17 20:33:49] lists at nutshell dot nu

I think I have found the problem.

Mandrake builds three c-clients.

libc-client.a
libc-client-PHP4.a
libc-client-nossl.a

Note than the -PHP4 is built without kerberos support.

I renamed libc-client.a to libc-client-org.a and libc-client-PHP4.a to
libc-client.a.

Then PHP5 compiles just fine with a shared imap.

Is there any way of building without having to rename the libs?



[2003-11-16 14:25:38] lists at nutshell dot nu

Yes, as stated in my original post it fails both with and without
shared. I have tried every possible combination I can think of.

It compiles just fine without imap, but I need imap.

Can also add that Mandrake 9.2 comes with a php imap rpm and the only
things I compile from tar-balls are apache and php4/5.



[2003-11-16 13:16:33] [EMAIL PROTECTED]

Does it fail if you do not compile imap as a shared extension?



[2003-11-14 04:52:38] lists at nutshell dot nu

Description:

Please note that Mandrake 9.2 comes with a mod-imap for php4, but I
wanted to compile it to suite my needs.

I have tryed compiling php5 (b2 and later snapshots, as well as latest
official php4) with imap. I have tried with Mandrake 9.2's own
imap-2002d, recompiling it with "rpm -bb --target i686", as well as
downgrading to imap-2001a.

I have tried to configure with:
--with-imap=shared
--with-imap-ssl
or
--with-imap=shared
--without-imap-ssl
or
--with-imap=shared

Plus also adding "/usr" as dir and without shared.

Note that if I try to configure "--without-kerberos" the configure
stops at imap.

I have all normal and devel packages installed for imap, openssl and
keberos, as well as any util packages for them.

./configure works just fine, but when I run make I get errors.

I get the following error:
/usr/local/src/php5-200311122030/ext/imap/php_imap.c: In function
`zm_startup_imap':
/usr/local/src/php5-200311122030/ext/imap/php_imap.c:445: error:
`auth_gss' undeclared (first use in this function)
/usr/local/src/php5-200311122030/ext/imap/php_imap.c:445: error: (Each
undeclared identifier is reported only once
/usr/local/src/php5-200311122030/ext/imap/php_imap.c:445: error: for
each function it appears in.)
make: *** [ext/imap/php_imap.lo] Error 1






-- 
Edit this bug report at http://bugs.php.net/?id=26248&edit=1

#26300 [Opn->Csd]: PB serialization ascii code greater than 127

[iliaa]

 ID:   26300
 Updated by:   [EMAIL PROTECTED]
 Reported By:  pierre-marie dot mouliere at arc-intl dot com
-Status:   Open
+Status:   Closed
 Bug Type: WDDX related
 Operating System: HPUX
 PHP Version:  4.3.3
 New Comment:

Thank you for your bug report. This issue has already been fixed
in the latest released version of PHP, which you can download at 
http://www.php.net/downloads.php




Previous Comments:


[2003-11-18 10:12:41] pierre-marie dot mouliere at arc-intl dot com

Description:

Hi,
There's a little bug with strings serialization, for strings containing
characters with ascii code greater than 127 (with HP-UX v11).

We found that the problem is located in function  :
php_wddx_serialize_string.
There is a line of code like :
 if (iscntrl((int)*p) {
When casting character (*p) to int for values greater than 127 (witch
are considered negative) there is a sign extension so character
'ñ'(ascii code 0xf0) become integer 0xfff0.
But the function iscntrl expects an integrer between 0 and 255, or the
result is unpredictable.
The problem can be fixed by modifying the line above like :
 if(iscntrl((int)(unsigned char)*p) {   
Best Regards
Pierre-Marie MOULIERE

Reproduce code:
---
Without locale :";
print "àâäçéèêëîïôù ÀÂÄÉÈÊËÎÏÔÙ ñ Ñ" . "\n";
print wddx_deserialize(wddx_serialize_value("àâäçéèêëîïôù ÀÂÄÉÈÊËÎÏÔÙ ñ
Ñ")); 
print "";
print "With locale (fr_FR.iso88591) :";
setlocale(LC_CTYPE, "fr_FR.iso88591");
print "àâäçéèêëîïôù ÀÂÄÉÈÊËÎÏÔÙ ñ Ñ" . "\n";
print wddx_deserialize(wddx_serialize_value("àâäçéèêëîïôù ÀÂÄÉÈÊËÎÏÔÙ ñ
Ñ")); 
?>  


Expected result:

Result : 
Without locale :

àâäçéèêëîïôù ÀÂÄÉÈÊËÎÏÔÙ ñ Ñ
àâäçéèêëîïôù ÀÂÄÉÈÊËÎÏÔÙ ñ Ñ

---

With locale (fr_FR.iso88591) :

àâäçéèêëîïôù ÀÂÄÉÈÊËÎÏÔÙ ñ Ñ
àâäçéèêëîïôù ÀÂÄÉÈÊËÎÏÔÙ ñ Ñ



Actual result:
--
Result : 
Without locale :

àâäçéèêëîïôù ÀÂÄÉÈÊËÎÏÔÙ ñ Ñ
àâäçéèêëîïôù ÀÂÄÉÈÊËÎÏÔÙ ? Ñ

---

With locale (fr_FR.iso88591) :

àâäçéèêëîïôù ÀÂÄÉÈÊËÎÏÔÙ ñ Ñ
àâäçéèêëîïôù ÀÂÄ?ÈÊËÎÏÔÙ ñ Ñ





-- 
Edit this bug report at http://bugs.php.net/?id=26300&edit=1

#26221 [Opn->Bgs]: Configure with imap-support broken

[sniper]

 ID:   26221
 Updated by:   [EMAIL PROTECTED]
 Reported By:  admin at sportsandbytes dot de
-Status:   Open
+Status:   Bogus
 Bug Type: IMAP related
 Operating System: Linux, RedHat 7.3
 PHP Version:  4.3.4
 New Comment:

Check the config.log for the reason why this fails.
This works just fine when your libraries and headers are correctly
installed and you use the correct configure options.

Ask further support questions on the mailing lists.



Previous Comments:


[2003-11-18 06:26:19] admin at sportsandbytes dot de

from configure --help 
--with-imap-ssl= IMAP: Include SSL support. DIR is the 
OpenSSL install dir. 
 
so either the configure option is wrong or 
--with-imap-ssl=/usr/local/openssl/current  
should work 
 
We specifically DONT want it to link against the systems openssl in 
/usr/lib, which is what probably would happen when not giving a 
openssl dir.



[2003-11-12 11:00:52] [EMAIL PROTECTED]

Sorry, but your problem does not imply a bug in PHP itself.  For a
list of more appropriate places to ask for help using PHP, please
visit http://www.php.net/support.php as this bug system is not the
appropriate forum for asking support questions. 

Thank you for your interest in PHP.

Use --with-imap-ssl not 
--with-imap-ssl=/usr/local/openssl/current/



[2003-11-12 10:29:02] admin at sportsandbytes dot de

Description:

Hi,

the following Configure-Options work with php-4.3.3 but not with
php-4.3.4:
./configure \
--with-apxs=/usr/local/apache/current/bin/apxs \
--with-zlib \
--with-jpeg-dir=/usr/lib \
--with-png-dir=/usr/lib \
--enable-ftp \
--with-gd \
--enable-versioning \
--enable-track-vars=yes \
--enable-url-includes \
--enable-sysvshm=yes \
--enable-sysvsem=yes \
--with-config-file-path=/etc \
--with-freetype-dir=/usr/lib \
--enable-fastcgi \
--enable-debug \
--disable-ipv6 \
--with-openssl=/usr/local/openssl/current/ \
--with-imap \
--with-imap-ssl=/usr/local/openssl/current/ \
--with-kerberos

The Configure-Script ends with this:

...
...
...
checking whether to enable dbx support... no
checking whether to enable direct I/O support... no
checking for DOM support... no
checking for DOM XSLT support... no
checking for DOM EXSLT support... no
checking whether to enable EXIF (metadata from images) support... no
checking for FrontBase SQL92 (fbsql) support... no
checking for FDF support... no
checking whether to enable the bundled filePro support... no
checking for FriBidi support... no
checking whether to enable FTP support... yes
checking for GD support... yes
checking for the location of libjpeg... /usr/lib
checking for the location of libpng... /usr/lib
checking for the location of libXpm... no
checking for FreeType 1.x support... no
checking for FreeType 2... /usr/lib
checking for T1lib support... no
checking whether to enable truetype string function in GD... no
checking whether to enable JIS-mapped Japanese font support in GD...
no
checking for fabsf... yes
checking for floorf... yes
checking for jpeg_read_header in -ljpeg... yes
checking for png_write_image in -lpng... yes
If configure fails try --with-xpm-dir=
checking for GNU gettext support... no
checking for GNU MP support... no
checking for hwapi support... no
checking for Hyperwave support... no
checking for iconv support... no
checking for IMAP support... yes
checking for pam_start in -lpam... yes
checking for crypt in -lcrypt... yes
configure: error: Cannot find imap library (libc-client.a). Please
check your IMAP installation.


The file is located here:
[EMAIL PROTECTED] src]# locate libc-client.a
/usr/lib/libc-client.a
[EMAIL PROTECTED] src]# cd /usr/lib
[EMAIL PROTECTED] lib]# ls -al | grep c-client
-rw-r--r--1 root root  2404414 Apr 18  2002 c-client.a
lrwxrwxrwx1 root root   10 Jun  3 14:29 libc-client.a
-> c-client.a
[EMAIL PROTECTED] lib]# 

Without the imap-configure-options, configure and make work without a
failure!

The other Software:

apache_1.3.29 at /usr/local/apache/current
openssl-0.9.7c at /usr/local/openssl/current
courier-imap-2.1.2 at /usr/local/courier-imap/current









-- 
Edit this bug report at http://bugs.php.net/?id=26221&edit=1

#26254 [Opn->Fbk]: Inaccuracy in imagecopyresized

[sniper]

 ID:   26254
 Updated by:   [EMAIL PROTECTED]
 Reported By:  rc at opelgt dot org
-Status:   Open
+Status:   Feedback
 Bug Type: GD related
 Operating System: Mac OS X 10.2.8
 PHP Version:  4.3.4
 New Comment:

Provide a complete (and short as possible) example script which we can
use to try and reproduce this ourselves.
Also, if this happens with specific images, provide URLs to those too.



Previous Comments:


[2003-11-18 03:27:59] rc at opelgt dot org

I used the downloadable version of www.entropy.ch with 
an installer. So I didnt change something of that. Here 
is what phpinfo tells about the configuration command:

'./configure' '--prefix=/usr/local/php' '--with-config-
file-path=/usr/local/php/lib' '--with-apxs' '--with-
iconv=/usr/local/php' '--with-openssl=/usr' '--with-
zlib=/usr' '--with-mysql' '--with-pgsql=/Users/marc/
cvs/entropy/php-module/build/postgresql-build' '--
enable-cgi' '--with-gd' '--with-png-dir=/usr/local/php' 
'--with-freetype-dir=/usr/local/php' '--with-t1lib=/
usr/local/php' '--with-jpeg-dir=/usr/local/php' '--
with-tiff-dir=/usr/local/php' '--with-curl=/usr/local/
php' '--with-mcal=/usr/local/php' '--with-mcrypt=/usr/
local/php' '--with-mhash=/usr/local/php' '--with-
pdflib=/usr/local/php' '--with-imap=../imap-2002d' '--
with-imap-ssl=/usr' '--with-expat-dir=/usr/local/php' 
'--with-gettext=/usr/local/php' '--with-dom=/usr/local/
php' '--with-dom-xslt=/usr/local/php' '--with-dom-
exslt=/usr/local/php' '--with-xslt-sablot=/usr/local/
php' '--with-mssql=/usr/local/php' '--enable-xslt' '--
with-ldap' '--with-xmlrpc' '--with-xml' '--with-iodbc' 
'--enable-sockets' '--enable-dbx' '--enable-dbase' '--
enable-trans-sid' '--enable-exif' '--enable-wddx' '--
enable-ftp' '--enable-mbstring'



[2003-11-17 22:48:20] [EMAIL PROTECTED]

What was the configure line you used? 




[2003-11-17 08:22:47] rc at opelgt dot org

When the destination image is created with 
imagecreatetruecolor the behaviour seems to be ok.
So the bug is obviously for a palette based image.



[2003-11-17 07:48:31] rc at opelgt dot org

The bug is still existing in 4.3.4 R1.
So in PHP5 it is solved?



[2003-11-16 13:13:10] [EMAIL PROTECTED]

Thank you for your bug report. This issue has already been fixed
in the latest released version of PHP, which you can download at 
http://www.php.net/downloads.php





The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
http://bugs.php.net/26254

-- 
Edit this bug report at http://bugs.php.net/?id=26254&edit=1

#26249 [Opn->Bgs]: snmpget doesn't free memory

[sniper]

 ID:   26249
 Updated by:   [EMAIL PROTECTED]
 Reported By:  dacol dot jeremy at caramail dot com
-Status:   Open
+Status:   Bogus
 Bug Type: SNMP related
 Operating System: linux
 PHP Version:  4.3.2
 New Comment:

PHP uses memory, there's no bug here. (0.7% is nothing)




Previous Comments:


[2003-11-18 09:58:16] dacol dot jeremy at caramail dot com

i test a script with:
while(true)
{
system("snmpget.");
}
and see with "top" command the memory used : 0.7% all time.
it doesn't depend on the number of snmpget so i think it's a php
problem.



[2003-11-18 06:58:02] dacol dot jeremy at caramail dot com

hi,

i don't know how upgrad my php files with the latest version...

i up my old ucd-snmp version to net-snmp 5.1 and my bug still
happened.

please tell me exactly what to do with php4-STABLE-latest.tar.gz

the command, all things tou can

i am really sorry but i search on web and i find nothing on it.



[2003-11-17 17:52:10] [EMAIL PROTECTED]

Update your über-old ucd-snmp to the latest net-snmp version.
Then reopen if you still experience leaks which you can trace to happen
in PHP code (not in the net-snmp)




[2003-11-17 05:40:55] dacol dot jeremy at caramail dot com

hi 
i have copied the CVS on my php files and replace it.
but it doesn't work each time i use snmpget function it takes memory
which it doesn't free after a response.
in fact each 100 snmpget use 0.1% of my memory.
it's really the snmpget function which doesn't work.



[2003-11-17 03:06:59] response dot sorry at notremember dot pass

hi,

thank you for your answer.

can i only copy this on my existing php files or do i have to recompile
all?
in fact what i have to do with this file?

and sorry for the mistake i work with:
apache 1.3.28
php 4.3.3
ucd-snmp 4.2.2
on linux RH 7.1



The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
http://bugs.php.net/26249

-- 
Edit this bug report at http://bugs.php.net/?id=26249&edit=1

#26141 [Opn->Fbk]: Libtool + threads = library error

[sniper]

 ID:   26141
 Updated by:   [EMAIL PROTECTED]
 Reported By:  alan at pair dot com
-Status:   Open
+Status:   Feedback
 Bug Type: Compile Failure
 Operating System: FreeBSD 4.8-STABLE
 PHP Version:  4.3.4
 New Comment:

First you need to stop touching the libtool stuff yourself, that
doesn't help us finding what really is the problem.
Secondly, you need to find out which option in your configure line
causes this. And try also using the correct option values, like for
--with-jpeg-dir=/usr/local/lib <-- this /lib does NOT belong there.
Same for ANY other option.



Previous Comments:


[2003-11-18 13:52:33] alan at pair dot com

We have tried this with several different versions of libtool, and it
failed with all of them.  libtool is bundled with PHP, isn't it?  

Does this mean I need to go get libtool to fix the bug, and then
convince you guys to use the new version of libtool?

Alan



[2003-11-17 18:05:53] [EMAIL PROTECTED]

Obviously yet another libtool bug.




[2003-11-14 10:38:29] alan at pair dot com

OK, with the old ltmain.sh and with those commands, PHP builds
normally. So one of the configure options is triggering the pthread
stuff.  configure definitely knows to try -pthread, look in the source
around line 9215.   
   
 
# Create a list of thread flags to try.  Items starting with a "-" are 
 
# C compiler flags, and other items are library names, except for
"none" 
# which indicates that we try without any flags at all.
 
   
 
acx_pthread_flags="pthreads none -Kthread -kthread lthread -pthread
-pthreads -mthreads pthread --thread-safe -mt"   
   
 
# The ordering *is* (sometimes) important.  Some notes on the  
 
# individual items follow: 
 
...
 
# -pthread: Linux/gcc (kernel threads), BSD/gcc (userland threads) 
 
   
 
So when it decides to use -pthread, libtool later sees that and thinks
it's a library reference (e.g. -lm is libm.so) and decides to try to
analyze it with ld or whatever.



[2003-11-06 14:38:46] [EMAIL PROTECTED]

Try this configure line:

# rm config.cache && ./configure --disable-all --disable-cgi
# make clean && make

Does it work? That -pthread option is not coming from PHP,
are you sure it's not set by some environment variable when you run
configure/make ??





[2003-11-06 10:31:44] alan at pair dot com

You are correct in the sense that phpinfo() reports "Thread Safety
disabled".  But using this patch, the compilation was definitely taking
place with -pthread as an option   
to gcc, for whatever reason.

Truthfully, even if threads aren't enabled, I'd rather have it compile
than not compile.



The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
http://bugs.php.net/26141

-- 
Edit this bug report at http://bugs.php.net/?id=26141&edit=1

#26117 [Opn->Bgs]: Persistent connection not reused

[sniper]

 ID:   26117
 Updated by:   [EMAIL PROTECTED]
 Reported By:  spam at vrana dot cz
-Status:   Open
+Status:   Bogus
 Bug Type: MySQL related
 Operating System: Linux
 PHP Version:  4.3.3
 New Comment:

Well, it does work fine for me. 



Previous Comments:


[2003-11-18 07:33:03] spam at vrana dot cz

Can you please read my bug report carefuly before stating bug as bogus?
Persistent connection in PHP should reuse previous persistent
connection with the same host, user and password. The bug is that it
doesn't.



[2003-11-17 18:21:48] [EMAIL PROTECTED]

And this is PHP bug in where..?




[2003-11-05 12:20:11] spam at vrana dot cz

Yes. There were 466 processes with the same user, host and db. All
processes were in state Sleep.



[2003-11-05 11:21:33] [EMAIL PROTECTED]

When the "too many connections" problem occurs, have you tried running
"show full processlist;" query to see if infact all connections use the
same database?



[2003-11-05 03:24:59] spam at vrana dot cz

This is not the case. Believe me that I have read all similar bugs and
this is different. I also have read carefuly the manual and I
understand everything written there about persistent connections.

There is written in the Persistent Database Connections chapter of
manual: An 'identical' connection is a connection that was opened to
the same host, with the same username and the same password (where
applicable). So it's not true that by mysql_select_db() with different
database name the connection can't be reused. Anyway I use the same
database in all scripts.

I know that connection is persistent only over the child. Thus I wrote
that Apache configuration directive MaxClients is set to 150.



The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
http://bugs.php.net/26117

-- 
Edit this bug report at http://bugs.php.net/?id=26117&edit=1

#26306 [Opn->Bgs]: Recursive function call crashes engine

[sniper]

 ID:   26306
 Updated by:   [EMAIL PROTECTED]
 Reported By:  toolshed51 at yahoo dot com
-Status:   Open
+Status:   Bogus
 Bug Type: Reproducible crash
 Operating System: FreeBSD 4.9
 PHP Version:  4.3.3
 New Comment:

Thank you for taking the time to write to us, but this is not
a bug. Please double-check the documentation available at
http://www.php.net/manual/ and the instructions on how to report
a bug at http://bugs.php.net/how-to-report.php




Previous Comments:


[2003-11-18 14:23:45] toolshed51 at yahoo dot com

Description:

Engine crashes when passed buggy recursive function calling code.

PHP 4.3.3 (cli) (built: Sep 17 2003 22:54:17)
Copyright (c) 1997-2003 The PHP Group
Zend Engine v1.3.0, Copyright (c) 1998-2003 Zend Technologies
with Zend Optimizer v2.1.0, Copyright (c) 1998-2003, by Zend
Technologies

Reproduce code:
---


Expected result:

I expect an out of memory error or something to be caught.  Instead it
seems the engine gets caught in an infinite loop like my code and dies
painfully.

Actual result:
--
0x28ade8cb in zend_assign_to_variable_reference () from
/usr/local/libexec/apache/ZendOptimizer.so

Or!  If I add a call to error_log to count function calls (in my case
it was 8976 before engine crashed)... then I get...

0x81a07c1 in zend_hash_quick_add_or_update () from
/usr/local/libexec/apache/ZendOptimizer.so

If I remove zend_optimzer.so from php.ini I get this error
0x819aa2e in zendi_smart_strcmp ()

Backtrace is thousands and thousands of any of the given above
messages.





-- 
Edit this bug report at http://bugs.php.net/?id=26306&edit=1

#26305 [Opn->Bgs]: Segmentation Fault when endlessly including the same file

[sniper]

 ID:   26305
 Updated by:   [EMAIL PROTECTED]
 Reported By:  vesely at tana dot it
-Status:   Open
+Status:   Bogus
 Bug Type: Reproducible crash
 Operating System: Solaris
 PHP Version:  4.3.4
 New Comment:

Thank you for taking the time to write to us, but this is not
a bug. Please double-check the documentation available at
http://www.php.net/manual/ and the instructions on how to report
a bug at http://bugs.php.net/how-to-report.php




Previous Comments:


[2003-11-18 13:41:39] vesely at tana dot it

Description:

Yes, I've been so stupid to do that.
(I thought ../ could be relative to the file
being included rather than the file being executed.)

However, triggering a segmentation fault isn't the
nicest way out of that. How does php deal with VERY
big sources?

Reproduce code:
---
somedir/subdir/example.php:

   require("../common.inc");

somedir/common.inc:

   require("../common.inc");


Expected result:

Fatal: out of memory

Actual result:
--
Segmentation Fault (core dumped)





-- 
Edit this bug report at http://bugs.php.net/?id=26305&edit=1

#26306 [NEW]: Recursive function call crashes engine

[toolshed51 at yahoo dot com]

From: toolshed51 at yahoo dot com
Operating system: FreeBSD 4.9
PHP version:  4.3.3
PHP Bug Type: Reproducible crash
Bug description:  Recursive function call crashes engine

Description:

Engine crashes when passed buggy recursive function calling code.

PHP 4.3.3 (cli) (built: Sep 17 2003 22:54:17)
Copyright (c) 1997-2003 The PHP Group
Zend Engine v1.3.0, Copyright (c) 1998-2003 Zend Technologies
with Zend Optimizer v2.1.0, Copyright (c) 1998-2003, by Zend
Technologies

Reproduce code:
---


Expected result:

I expect an out of memory error or something to be caught.  Instead it
seems the engine gets caught in an infinite loop like my code and dies
painfully.

Actual result:
--
0x28ade8cb in zend_assign_to_variable_reference () from
/usr/local/libexec/apache/ZendOptimizer.so

Or!  If I add a call to error_log to count function calls (in my case it
was 8976 before engine crashed)... then I get...

0x81a07c1 in zend_hash_quick_add_or_update () from
/usr/local/libexec/apache/ZendOptimizer.so

If I remove zend_optimzer.so from php.ini I get this error
0x819aa2e in zendi_smart_strcmp ()

Backtrace is thousands and thousands of any of the given above messages.

-- 
Edit bug report at http://bugs.php.net/?id=26306&edit=1
-- 
Try a CVS snapshot (php4):  http://bugs.php.net/fix.php?id=26306&r=trysnapshot4
Try a CVS snapshot (php5):  http://bugs.php.net/fix.php?id=26306&r=trysnapshot5
Fixed in CVS:   http://bugs.php.net/fix.php?id=26306&r=fixedcvs
Fixed in release:   http://bugs.php.net/fix.php?id=26306&r=alreadyfixed
Need backtrace: http://bugs.php.net/fix.php?id=26306&r=needtrace
Try newer version:  http://bugs.php.net/fix.php?id=26306&r=oldversion
Not developer issue:http://bugs.php.net/fix.php?id=26306&r=support
Expected behavior:  http://bugs.php.net/fix.php?id=26306&r=notwrong
Not enough info:http://bugs.php.net/fix.php?id=26306&r=notenoughinfo
Submitted twice:http://bugs.php.net/fix.php?id=26306&r=submittedtwice
register_globals:   http://bugs.php.net/fix.php?id=26306&r=globals
PHP 3 support discontinued: http://bugs.php.net/fix.php?id=26306&r=php3
Daylight Savings:   http://bugs.php.net/fix.php?id=26306&r=dst
IIS Stability:  http://bugs.php.net/fix.php?id=26306&r=isapi
Install GNU Sed:http://bugs.php.net/fix.php?id=26306&r=gnused
Floating point limitations: http://bugs.php.net/fix.php?id=26306&r=float

#26141 [Bgs->Opn]: Libtool + threads = library error

[alan at pair dot com]

 ID:   26141
 User updated by:  alan at pair dot com
 Reported By:  alan at pair dot com
-Status:   Bogus
+Status:   Open
 Bug Type: Compile Failure
 Operating System: FreeBSD 4.8-STABLE
 PHP Version:  4.3.4
 New Comment:

We have tried this with several different versions of libtool, and it
failed with all of them.  libtool is bundled with PHP, isn't it?  

Does this mean I need to go get libtool to fix the bug, and then
convince you guys to use the new version of libtool?

Alan


Previous Comments:


[2003-11-17 18:05:53] [EMAIL PROTECTED]

Obviously yet another libtool bug.




[2003-11-14 10:38:29] alan at pair dot com

OK, with the old ltmain.sh and with those commands, PHP builds
normally. So one of the configure options is triggering the pthread
stuff.  configure definitely knows to try -pthread, look in the source
around line 9215.   
   
 
# Create a list of thread flags to try.  Items starting with a "-" are 
 
# C compiler flags, and other items are library names, except for
"none" 
# which indicates that we try without any flags at all.
 
   
 
acx_pthread_flags="pthreads none -Kthread -kthread lthread -pthread
-pthreads -mthreads pthread --thread-safe -mt"   
   
 
# The ordering *is* (sometimes) important.  Some notes on the  
 
# individual items follow: 
 
...
 
# -pthread: Linux/gcc (kernel threads), BSD/gcc (userland threads) 
 
   
 
So when it decides to use -pthread, libtool later sees that and thinks
it's a library reference (e.g. -lm is libm.so) and decides to try to
analyze it with ld or whatever.



[2003-11-06 14:38:46] [EMAIL PROTECTED]

Try this configure line:

# rm config.cache && ./configure --disable-all --disable-cgi
# make clean && make

Does it work? That -pthread option is not coming from PHP,
are you sure it's not set by some environment variable when you run
configure/make ??





[2003-11-06 10:31:44] alan at pair dot com

You are correct in the sense that phpinfo() reports "Thread Safety
disabled".  But using this patch, the compilation was definitely taking
place with -pthread as an option   
to gcc, for whatever reason.

Truthfully, even if threads aren't enabled, I'd rather have it compile
than not compile.



[2003-11-05 11:12:57] [EMAIL PROTECTED]

With that configure line you won't be enabling any thread stuff.




The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
http://bugs.php.net/26141

-- 
Edit this bug report at http://bugs.php.net/?id=26141&edit=1

#26305 [NEW]: Segmentation Fault when endlessly including the same file

[vesely at tana dot it]

From: vesely at tana dot it
Operating system: Solaris
PHP version:  4.3.4
PHP Bug Type: Reproducible crash
Bug description:  Segmentation Fault when endlessly including the same file 

Description:

Yes, I've been so stupid to do that.
(I thought ../ could be relative to the file
being included rather than the file being executed.)

However, triggering a segmentation fault isn't the
nicest way out of that. How does php deal with VERY
big sources?

Reproduce code:
---
somedir/subdir/example.php:

   require("../common.inc");

somedir/common.inc:

   require("../common.inc");


Expected result:

Fatal: out of memory

Actual result:
--
Segmentation Fault (core dumped)

-- 
Edit bug report at http://bugs.php.net/?id=26305&edit=1
-- 
Try a CVS snapshot (php4):  http://bugs.php.net/fix.php?id=26305&r=trysnapshot4
Try a CVS snapshot (php5):  http://bugs.php.net/fix.php?id=26305&r=trysnapshot5
Fixed in CVS:   http://bugs.php.net/fix.php?id=26305&r=fixedcvs
Fixed in release:   http://bugs.php.net/fix.php?id=26305&r=alreadyfixed
Need backtrace: http://bugs.php.net/fix.php?id=26305&r=needtrace
Try newer version:  http://bugs.php.net/fix.php?id=26305&r=oldversion
Not developer issue:http://bugs.php.net/fix.php?id=26305&r=support
Expected behavior:  http://bugs.php.net/fix.php?id=26305&r=notwrong
Not enough info:http://bugs.php.net/fix.php?id=26305&r=notenoughinfo
Submitted twice:http://bugs.php.net/fix.php?id=26305&r=submittedtwice
register_globals:   http://bugs.php.net/fix.php?id=26305&r=globals
PHP 3 support discontinued: http://bugs.php.net/fix.php?id=26305&r=php3
Daylight Savings:   http://bugs.php.net/fix.php?id=26305&r=dst
IIS Stability:  http://bugs.php.net/fix.php?id=26305&r=isapi
Install GNU Sed:http://bugs.php.net/fix.php?id=26305&r=gnused
Floating point limitations: http://bugs.php.net/fix.php?id=26305&r=float

#26304 [Opn]: Unexpected data loss when opening dba file

[vesely at tana dot it]

 ID:   26304
 User updated by:  vesely at tana dot it
 Reported By:  vesely at tana dot it
 Status:   Open
 Bug Type: DBM/DBA related
 Operating System: Solaris
 PHP Version:  4.3.4
 New Comment:

Also, that correction around line 67 in dba_db4.c
(DBA_OPEN_FUNC) is bogus: if stat returns 0 you want
to say DB_UNKNOWN, since you cannot say it is DB_BTREE
when it was created by some other program.

To reproduce the bug you should create a DB with a different
type, e.g. in C if you just include db.h and then call
the dbm_open compatibility layer. The Sleepycat message
will then say "call implies an access method which is
inconsistent with previous calls."


Previous Comments:


[2003-11-18 12:09:46] vesely at tana dot it

Description:

Opening a file in 'c' mode (see example below)
truncates the file!! The docs say '"c" for read/write
access and database creation if it doesn't currently exist.'

I cannot understand that comment at line 658 in
ext/dba/dba.c, as it seems to imply that truncating
the database is necessary for locking it (??).

Reproduce code:
---
dba_open("important_data.db", "c", "db4");

Expected result:

open db (create if doesn't exist)

Actual result:
--
truncated db





-- 
Edit this bug report at http://bugs.php.net/?id=26304&edit=1

#26289 [Fbk->Bgs]: Random problem in mail() function

[iliaa]

 ID:   26289
 Updated by:   [EMAIL PROTECTED]
 Reported By:  julien at touslesdrivers dot com
-Status:   Feedback
+Status:   Bogus
 Bug Type: *Mail Related
 Operating System: Windows 2000 Server SP4
 PHP Version:  4.3.4
 New Comment:

This is not a PHP bug, somewhere along the line $code is not populated
correctly, hence the error you are seeing.


Previous Comments:


[2003-11-18 12:34:36] [EMAIL PROTECTED]

Sorry, but your problem does not imply a bug in PHP itself.  For a
list of more appropriate places to ask for help using PHP, please
visit http://www.php.net/support.php as this bug system is not the
appropriate forum for asking support questions. 

Thank you for your interest in PHP.





[2003-11-18 06:59:42] julien at touslesdrivers dot com

up



[2003-11-17 14:45:46] julien at touslesdrivers dot com

Thanks you fo your reply.

I don't think the problem is the $code beacause this variable is set
and is set correctly. Otherwise the value of $code would not appear in
the URL...

I just want to know where this "VOID" come from?! And why it replaces
the string "script" in the file name?

Thanks



[2003-11-17 14:16:12] [EMAIL PROTECTED]

Sorry, but your problem does not imply a bug in PHP itself.  For a
list of more appropriate places to ask for help using PHP, please
visit http://www.php.net/support.php as this bug system is not the
appropriate forum for asking support questions. 

Thank you for your interest in PHP.

I suspect the $code is not set or not set correctly.

First of all, change http://www.domain.com/inscription.php?code=$code
to
http://www.domain.com/inscription.php?code={$code}
and always validate value of $code before using it.



[2003-11-17 11:10:45] julien at touslesdrivers dot com

Description:

Hello,

I have a problem with the mail() function. The OS is Windows 2000
Server SP4 + IIS 5.0 + MySQL + PHP 4.3.4 (ISAPI).

On my website, there is a form to subscribe to the site as a member. To
confirm the subscription, I send a mail to the person who has to click
on a link to confirm the subscription.

I use so the mail() function to send the mail as text format but there
is a random problem.

Sometimes (about 2-3 times per week) persons who receive the mail can't
confirm their subscription and forward the mail to me and I can then
constat the problem:

A little part of the body (the string "script") had been replaced by
another string ("VOID").

This problem is not dependant with the Mail client (Outlook, Netscape
etc.).

PHP 4.3.4 doesn't resolves this problem.

I am not arrived to reproduce this problem and it it seems to be
totally random. It is a very strange bug!! An idea?

Reproduce code:
---
The code source is:

$message = "
Bonjour,

Pour valider définitivement votre inscription comme membre, merci de
clicker sur le lien ci-dessous:
http://www.domain.com/inscription.php?code=$code
";

mail($email,"Inscription",$message,"From:
[EMAIL PROTECTED]: text/plain; charset=iso-8859-1");

Expected result:

Normally the person would see:

Bonjour,

Pour valider définitivement votre inscription comme membre, merci de
clicker sur le lien ci-dessous:
http://www.domain.com/inscription.php?code=1587

Actual result:
--
But some persons obtain:

Bonjour,

Pour valider définitivement votre inscription comme membre, merci de
clicker sur le lien ci-dessous:
http://www.domain.com/inVOIDion.php?code=1587





-- 
Edit this bug report at http://bugs.php.net/?id=26289&edit=1

#26289 [Opn->Fbk]: Random problem in mail() function

[iliaa]

 ID:   26289
 Updated by:   [EMAIL PROTECTED]
 Reported By:  julien at touslesdrivers dot com
-Status:   Open
+Status:   Feedback
 Bug Type: *Mail Related
 Operating System: Windows 2000 Server SP4
 PHP Version:  4.3.4
 New Comment:

Sorry, but your problem does not imply a bug in PHP itself.  For a
list of more appropriate places to ask for help using PHP, please
visit http://www.php.net/support.php as this bug system is not the
appropriate forum for asking support questions. 

Thank you for your interest in PHP.




Previous Comments:


[2003-11-18 06:59:42] julien at touslesdrivers dot com

up



[2003-11-17 14:45:46] julien at touslesdrivers dot com

Thanks you fo your reply.

I don't think the problem is the $code beacause this variable is set
and is set correctly. Otherwise the value of $code would not appear in
the URL...

I just want to know where this "VOID" come from?! And why it replaces
the string "script" in the file name?

Thanks



[2003-11-17 14:16:12] [EMAIL PROTECTED]

Sorry, but your problem does not imply a bug in PHP itself.  For a
list of more appropriate places to ask for help using PHP, please
visit http://www.php.net/support.php as this bug system is not the
appropriate forum for asking support questions. 

Thank you for your interest in PHP.

I suspect the $code is not set or not set correctly.

First of all, change http://www.domain.com/inscription.php?code=$code
to
http://www.domain.com/inscription.php?code={$code}
and always validate value of $code before using it.



[2003-11-17 11:10:45] julien at touslesdrivers dot com

Description:

Hello,

I have a problem with the mail() function. The OS is Windows 2000
Server SP4 + IIS 5.0 + MySQL + PHP 4.3.4 (ISAPI).

On my website, there is a form to subscribe to the site as a member. To
confirm the subscription, I send a mail to the person who has to click
on a link to confirm the subscription.

I use so the mail() function to send the mail as text format but there
is a random problem.

Sometimes (about 2-3 times per week) persons who receive the mail can't
confirm their subscription and forward the mail to me and I can then
constat the problem:

A little part of the body (the string "script") had been replaced by
another string ("VOID").

This problem is not dependant with the Mail client (Outlook, Netscape
etc.).

PHP 4.3.4 doesn't resolves this problem.

I am not arrived to reproduce this problem and it it seems to be
totally random. It is a very strange bug!! An idea?

Reproduce code:
---
The code source is:

$message = "
Bonjour,

Pour valider définitivement votre inscription comme membre, merci de
clicker sur le lien ci-dessous:
http://www.domain.com/inscription.php?code=$code
";

mail($email,"Inscription",$message,"From:
[EMAIL PROTECTED]: text/plain; charset=iso-8859-1");

Expected result:

Normally the person would see:

Bonjour,

Pour valider définitivement votre inscription comme membre, merci de
clicker sur le lien ci-dessous:
http://www.domain.com/inscription.php?code=1587

Actual result:
--
But some persons obtain:

Bonjour,

Pour valider définitivement votre inscription comme membre, merci de
clicker sur le lien ci-dessous:
http://www.domain.com/inVOIDion.php?code=1587





-- 
Edit this bug report at http://bugs.php.net/?id=26289&edit=1

#26304 [NEW]: Unexpected data loss when opening dba file

[vesely at tana dot it]

From: vesely at tana dot it
Operating system: Solaris
PHP version:  4.3.4
PHP Bug Type: DBM/DBA related
Bug description:  Unexpected data loss when opening dba file

Description:

Opening a file in 'c' mode (see example below)
truncates the file!! The docs say '"c" for read/write
access and database creation if it doesn't currently exist.'

I cannot understand that comment at line 658 in
ext/dba/dba.c, as it seems to imply that truncating
the database is necessary for locking it (??).

Reproduce code:
---
dba_open("important_data.db", "c", "db4");

Expected result:

open db (create if doesn't exist)

Actual result:
--
truncated db

-- 
Edit bug report at http://bugs.php.net/?id=26304&edit=1
-- 
Try a CVS snapshot (php4):  http://bugs.php.net/fix.php?id=26304&r=trysnapshot4
Try a CVS snapshot (php5):  http://bugs.php.net/fix.php?id=26304&r=trysnapshot5
Fixed in CVS:   http://bugs.php.net/fix.php?id=26304&r=fixedcvs
Fixed in release:   http://bugs.php.net/fix.php?id=26304&r=alreadyfixed
Need backtrace: http://bugs.php.net/fix.php?id=26304&r=needtrace
Try newer version:  http://bugs.php.net/fix.php?id=26304&r=oldversion
Not developer issue:http://bugs.php.net/fix.php?id=26304&r=support
Expected behavior:  http://bugs.php.net/fix.php?id=26304&r=notwrong
Not enough info:http://bugs.php.net/fix.php?id=26304&r=notenoughinfo
Submitted twice:http://bugs.php.net/fix.php?id=26304&r=submittedtwice
register_globals:   http://bugs.php.net/fix.php?id=26304&r=globals
PHP 3 support discontinued: http://bugs.php.net/fix.php?id=26304&r=php3
Daylight Savings:   http://bugs.php.net/fix.php?id=26304&r=dst
IIS Stability:  http://bugs.php.net/fix.php?id=26304&r=isapi
Install GNU Sed:http://bugs.php.net/fix.php?id=26304&r=gnused
Floating point limitations: http://bugs.php.net/fix.php?id=26304&r=float

#26302 [NEW]: open_basedir /home outside the virtual host

[sanry at now dot net dot cn]

From: sanry at now dot net dot cn
Operating system: redhat9
PHP version:  4.3.4
PHP Bug Type: Feature/Change Request
Bug description:  open_basedir /home  outside the virtual host

Description:

when use mod vhost in apache1.3.x  
like this 


  php_admin_value open_basedir /home

..



..


open_basedir will affect sometimes in all the virualhost after one day or
more runing ( this bug appear since php4.3.0 and php4.2.2 is ok)



-- 
Edit bug report at http://bugs.php.net/?id=26302&edit=1
-- 
Try a CVS snapshot (php4):  http://bugs.php.net/fix.php?id=26302&r=trysnapshot4
Try a CVS snapshot (php5):  http://bugs.php.net/fix.php?id=26302&r=trysnapshot5
Fixed in CVS:   http://bugs.php.net/fix.php?id=26302&r=fixedcvs
Fixed in release:   http://bugs.php.net/fix.php?id=26302&r=alreadyfixed
Need backtrace: http://bugs.php.net/fix.php?id=26302&r=needtrace
Try newer version:  http://bugs.php.net/fix.php?id=26302&r=oldversion
Not developer issue:http://bugs.php.net/fix.php?id=26302&r=support
Expected behavior:  http://bugs.php.net/fix.php?id=26302&r=notwrong
Not enough info:http://bugs.php.net/fix.php?id=26302&r=notenoughinfo
Submitted twice:http://bugs.php.net/fix.php?id=26302&r=submittedtwice
register_globals:   http://bugs.php.net/fix.php?id=26302&r=globals
PHP 3 support discontinued: http://bugs.php.net/fix.php?id=26302&r=php3
Daylight Savings:   http://bugs.php.net/fix.php?id=26302&r=dst
IIS Stability:  http://bugs.php.net/fix.php?id=26302&r=isapi
Install GNU Sed:http://bugs.php.net/fix.php?id=26302&r=gnused
Floating point limitations: http://bugs.php.net/fix.php?id=26302&r=float

#26301 [NEW]: open_basedir /home outside the virtual host

[sanry at now dot net dot cn]

From: sanry at now dot net dot cn
Operating system: redhat9
PHP version:  4.3.4
PHP Bug Type: Feature/Change Request
Bug description:  open_basedir /home  outside the virtual host

Description:

when use mod vhost in apache1.3.x  
like this 


  php_admin_value open_basedir /home

..



..


open_basedir will affect sometimes in all the virualhost after one day or
more runing ( this appear since php4.3.0)
a day or more runing 


-- 
Edit bug report at http://bugs.php.net/?id=26301&edit=1
-- 
Try a CVS snapshot (php4):  http://bugs.php.net/fix.php?id=26301&r=trysnapshot4
Try a CVS snapshot (php5):  http://bugs.php.net/fix.php?id=26301&r=trysnapshot5
Fixed in CVS:   http://bugs.php.net/fix.php?id=26301&r=fixedcvs
Fixed in release:   http://bugs.php.net/fix.php?id=26301&r=alreadyfixed
Need backtrace: http://bugs.php.net/fix.php?id=26301&r=needtrace
Try newer version:  http://bugs.php.net/fix.php?id=26301&r=oldversion
Not developer issue:http://bugs.php.net/fix.php?id=26301&r=support
Expected behavior:  http://bugs.php.net/fix.php?id=26301&r=notwrong
Not enough info:http://bugs.php.net/fix.php?id=26301&r=notenoughinfo
Submitted twice:http://bugs.php.net/fix.php?id=26301&r=submittedtwice
register_globals:   http://bugs.php.net/fix.php?id=26301&r=globals
PHP 3 support discontinued: http://bugs.php.net/fix.php?id=26301&r=php3
Daylight Savings:   http://bugs.php.net/fix.php?id=26301&r=dst
IIS Stability:  http://bugs.php.net/fix.php?id=26301&r=isapi
Install GNU Sed:http://bugs.php.net/fix.php?id=26301&r=gnused
Floating point limitations: http://bugs.php.net/fix.php?id=26301&r=float

#18347 [Csd->Bgs]: ibase_pconnect() spawn connections to InterBase

[abies]

 ID:   18347
 Updated by:   [EMAIL PROTECTED]
 Reported By:  michael at mediacom dot ru
-Status:   Closed
+Status:   Bogus
 Bug Type: InterBase related
 Operating System: nt4 sp6a
 PHP Version:  4.2.1


Previous Comments:


[2002-07-16 02:04:03] michael at mediacom dot ru

Yes! You almost right.
I have looked in httpd.conf and understood that reason in settings of
KeepAlive (KeepAliveTimeout & ThreadsPerChild) of my Apache server.

Thank for support.



[2002-07-15 20:23:10] [EMAIL PROTECTED]

The persistent connection does not mean that you have ONE single
connection to the database, but you have one connection from each httpd
child. Are you sure this is not what is happening? ie. if the child you
hit does not have the connection established yet, it will be created.




[2002-07-15 08:15:44] michael at mediacom dot ru

Settings for interbase in php.ini? Noting. I didn't found description
for those in PHP Manual.
Interbase is not closing connection - thay are not closing. PHP spawned
new connections in ibase_pconnect() function, when there are available
persistent connections.



[2002-07-15 06:27:22] [EMAIL PROTECTED]

What settings you have for interbase in php.ini?
Are you sure it's not interbase closing the connection?
Check it's settings too..




[2002-07-15 04:43:13] michael at mediacom dot ru

My script calls ibase_pconnect("127.0.0.1:/Database/test.gdb",
"SYSDBA", "masterkey"). I expected that php reuse persistent connection
to InterBase. But if time from last script executing is over 15 seconds
- php spawn new InterBase connection.



The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
http://bugs.php.net/18347

-- 
Edit this bug report at http://bugs.php.net/?id=18347&edit=1

#19915 [Csd->Bgs]: ibase_query() fails but returns true

[abies]

 ID:   19915
 Updated by:   [EMAIL PROTECTED]
 Reported By:  irie at gmx dot de
-Status:   Closed
+Status:   Bogus
 Bug Type: InterBase related
 Operating System: Linux 2.4.18
 PHP Version:  4.3.0-pre1


Previous Comments:


[2003-08-09 22:13:22] [EMAIL PROTECTED]

Because InterBase uses a transaction to process the metadata updates,
the dependency conflict is not reported
before the is transaction is committed.



[2002-10-15 05:43:02] irie at gmx dot de

Hi,
dropping a table fails in interbase if there are existing dependensies
for the table, i.e. if a view is using the table.

Executing such a query with the isql tool I get:

SQL> DROP TABLE SONGS;
Statement failed, SQLCODE = -607

unsuccessful metadata update
-cannot delete
-COLUMN ARTIST
-there are 1 dependencies


But executing the same query via php's ibase_query() returns TRUE :

var_dump(ibase_query($dbhandle, 'DROP TABLE SONGS'));

> bool(true)


Beside 4.3.0-pre1 I've just tested this with 4.0.6 and found the same
behavior.

Lutz




-- 
Edit this bug report at http://bugs.php.net/?id=19915&edit=1

#25282 [Csd->Bgs]: API missing two important parameters

[abies]

 ID:   25282
 Updated by:   [EMAIL PROTECTED]
 Reported By:  nigel at e-easy dot com dot au
-Status:   Closed
+Status:   Bogus
 Bug Type: InterBase related
 Operating System: Irrelevant
 PHP Version:  Irrelevant


Previous Comments:


[2003-08-28 18:06:43] [EMAIL PROTECTED]

The constants are useless without a way of specifying a table name.
Therefore, this will not be changed. PHP5 will support table
reservation by means of SET TRANSACTION statements.



[2003-08-28 17:38:07] nigel at e-easy dot com dot au

Handling Lock Conflicts is what I'm trying to do.
By stating that I want to reserve a table as shared write, I can then
update a single field much faster, as any transaction can write to the
table - even if it's been updated inside this transaction space.

...Or so I thought...

On the Firebird list, they're stating that the default mode is Shared
Write - not mentioned anywhere in the PHP source.

Ah, another undocumented behaviour.

Thanks for your time!

N.



[2003-08-28 04:47:26] [EMAIL PROTECTED]

Passing the table reservation flags in this way will result in
absolutely nothing, because they require a table name as an extra
argument. You cannot just lock all tables for [shared|protected]
[read|write].

As PHP5 will contain support in ibase_trans() for multiple databases,
adding tablenames as arguments as well would clutter up the prototype
too much.

PHP5 _will_ allow you to execute SET TRANSACTION statements. Then you
can pass table reservation flags (with tablenames) anyway you like.

I seriously doubt, though, that unconditionally locking tables instead
of only when necessary would actually improve performance. As far as I
know, these parameters were introduced to handle lock conflicts.



[2003-08-27 23:52:17] nigel at e-easy dot com dot au

The patch for interbase.c where we apply the mask for 
PHP_IBASE_PROTECTED should read: 
+   tpb[tpb_len++] = isc_tpb_protected 
 
and NOT: 
+   tpb[tpb_len++] = isc_tpb_shared 
 
sorry...



[2003-08-27 23:48:23] nigel at e-easy dot com dot au

Description:

PHP's support for Interbase(Firebird) is missing two important 
features for transaction handling - Shared, and Protected. 
These two additions enable performance to surpass Oracle on 
updates under heavy load. 
 
The patch is as follows: 
--- interbase.c.origThu Aug 28 11:39:22 2003 
+++ interbase.c Thu Aug 28 12:22:44 2003 
@@ -563,6 +563,8 @@ 

REGISTER_LONG_CONSTANT("IBASE_REC_NO_VERSION", 
PHP_IBASE_REC_NO_VERSION 
, CONST_PERSISTENT); 
REGISTER_LONG_CONSTANT("IBASE_NOWAIT", 
PHP_IBASE_NOWAIT, CONST_PERSISTE 
NT); 
REGISTER_LONG_CONSTANT("IBASE_WAIT", 
PHP_IBASE_WAIT, CONST_PERSISTENT); 
+   REGISTER_LONG_CONSTANT("IBASE_SHARED", 
PHP_IBASE_SHARED, CONST_PERSISTE 
NT); 
+   REGISTER_LONG_CONSTANT("IBASE_PROTECTED", 
PHP_IBASE_PROTECTED, CONST_PE 
RSISTENT); 
 
return SUCCESS; 
 } 
@@ -1574,6 +1576,13 @@ 
tpb[tpb_len++] = isc_tpb_nowait; 
} else { 
tpb[tpb_len++] = isc_tpb_wait;  /* default lock

resolut 
ion */ 
+   } 
+   /* Pre Allocation Modes */ 
+   if (trans_argl & PHP_IBASE_SHARED) { 
+   tpb[tpb_len++] = isc_tpb_shared; 
+   } 
+   if (trans_argl & PHP_IBASE_PROTECTED) { 
+   tpb[tpb_len++] = isc_tpb_shared; 
} 
} 
  
--- php_interbase.h.origThu Aug 28 11:59:34 2003 
+++ php_interbase.h Thu Aug 28 12:21:34 2003 
@@ -175,7 +175,9 @@ 
PHP_IBASE_REC_VERSION   = 64, 
PHP_IBASE_REC_NO_VERSION= 128, 
PHP_IBASE_NOWAIT= 256, 
-   PHP_IBASE_WAIT  = 512 
+   PHP_IBASE_WAIT  = 512, 
+   PHP_IBASE_SHARED= 1024, 
+   PHP_IBASE_PROTECTED = 2048 
 }; 
  
 #ifdef ZTS 
 






-- 
Edit this bug report at http://bugs.php.net/?id=25282&edit=1

#26299 [Opn->Bgs]: Inheritance problem: an object can have two same-named private variables

[jay]

 ID:   26299
 Updated by:   [EMAIL PROTECTED]
 Reported By:  forseti at oak dot rpg dot pl
-Status:   Open
+Status:   Bogus
 Bug Type: Zend Engine 2 problem
 Operating System: Windows 98 SE
 PHP Version:  5.0.0b2 (beta2)
 New Comment:

This is expected. When a member is private, it can only be 
seen by the class it is a member of. Class B has no idea 
class A also has a member called $a and goes ahead and 
creates one. It's just a matter of visibility. 
 
J 


Previous Comments:


[2003-11-18 10:05:05] forseti at oak dot rpg dot pl

Description:

When extending a class, private member variables from parent class are
retained even if child class defines same-named variable. 

Reproduce code:
---
';print_r(new B);echo '';
?>

Expected result:

b Object
(
[a:private] => From class B
)



Actual result:
--
b Object
(
[a:private] => From class B
[a:private] => From class A
)







-- 
Edit this bug report at http://bugs.php.net/?id=26299&edit=1

#26264 [Opn]: Compile failure with iconv enabled

[Joerg dot Dieter dot Friedrich at uni-konstanz dot de]

 ID:   26264
 User updated by:  Joerg dot Dieter dot Friedrich at uni-konstanz dot de
 Reported By:  Joerg dot Dieter dot Friedrich at uni-konstanz dot de
 Status:   Open
 Bug Type: ICONV related
 Operating System: Solaris 9
 PHP Version:  4.3.4
 New Comment:

Additional info:

I was not able to generate a core with --enable-debug, but here's the
backtrace with --disable-debug, maybe you can see someting.

Core was generated by `php login.php'.
Program terminated with signal 11, Segmentation fault.
Reading symbols from /usr/lib/libcrypt_i.so.1...done.
Loaded symbols for /usr/lib/libcrypt_i.so.1
Reading symbols from /usr/lib/libpam.so.1...done.
Loaded symbols for /usr/lib/libpam.so.1
Reading symbols from /usr/lib/libintl.so.1...
warning: Lowest section in /usr/lib/libintl.so.1 is .hash at 0074
done.
Loaded symbols for /usr/lib/libintl.so.1
Reading symbols from /usr/local/ssl/lib/libssl.so.0.9.7...done.
Loaded symbols for /usr/local/ssl/lib/libssl.so.0.9.7
Reading symbols from /usr/local/ssl/lib/libcrypto.so.0.9.7...done.
Loaded symbols for /usr/local/ssl/lib/libcrypto.so.0.9.7
Reading symbols from /usr/lib/libresolv.so.2...done.
Loaded symbols for /usr/lib/libresolv.so.2
Reading symbols from /usr/lib/libm.so.1...done.
Loaded symbols for /usr/lib/libm.so.1
Reading symbols from /usr/lib/libdl.so.1...done.
Loaded symbols for /usr/lib/libdl.so.1
Reading symbols from /usr/lib/libnsl.so.1...done.
Loaded symbols for /usr/lib/libnsl.so.1
Reading symbols from /usr/lib/libsocket.so.1...done.
Loaded symbols for /usr/lib/libsocket.so.1
Reading symbols from /usr/lib/libc.so.1...done.
Loaded symbols for /usr/lib/libc.so.1
Reading symbols from /usr/lib/libgen.so.1...done.
Loaded symbols for /usr/lib/libgen.so.1
Reading symbols from /usr/lib/libcmd.so.1...done.
Loaded symbols for /usr/lib/libcmd.so.1
Reading symbols from /usr/local/lib/libgcc_s.so.1...done.
Loaded symbols for /usr/local/lib/libgcc_s.so.1
Reading symbols from /usr/lib/libmp.so.2...done.
Loaded symbols for /usr/lib/libmp.so.2
Reading symbols from
/usr/platform/SUNW,Sun-Fire-V240/lib/libc_psr.so.1...done.
Loaded symbols for /usr/platform/SUNW,Sun-Fire-V240/lib/libc_psr.so.1
Reading symbols from
/opt/php4/lib/php/extensions/no-debug-non-zts-20020429/iconv.so...done.
Loaded symbols for
/opt/php4/lib/php/extensions/no-debug-non-zts-20020429/iconv.so
Reading symbols from
/opt/php4/lib/php/extensions/no-debug-non-zts-20020429/mbstring.so...done.
Loaded symbols for
/opt/php4/lib/php/extensions/no-debug-non-zts-20020429/mbstring.so
Reading symbols from
/opt/php4/lib/php/extensions/no-debug-non-zts-20020429/imap.so...done.
Loaded symbols for
/opt/php4/lib/php/extensions/no-debug-non-zts-20020429/imap.so
Reading symbols from
/opt/php4/lib/php/extensions/no-debug-non-zts-20020429/mysql.so...done.
Loaded symbols for
/opt/php4/lib/php/extensions/no-debug-non-zts-20020429/mysql.so
Reading symbols from /opt/mysql/lib/mysql/libmysqlclient.so.12...done.
Loaded symbols for /opt/mysql/lib/mysql/libmysqlclient.so.12
Reading symbols from /usr/lib/libz.so.1...done.
Loaded symbols for /usr/lib/libz.so.1
Reading symbols from /usr/lib/librt.so.1...done.
Loaded symbols for /usr/lib/librt.so.1
Reading symbols from /usr/lib/libaio.so.1...done.
Loaded symbols for /usr/lib/libaio.so.1
Reading symbols from /usr/lib/libmd5.so.1...done.
Loaded symbols for /usr/lib/libmd5.so.1
Reading symbols from
/opt/php4/lib/php/extensions/no-debug-non-zts-20020429/gd.so...done.
Loaded symbols for
/opt/php4/lib/php/extensions/no-debug-non-zts-20020429/gd.so
Reading symbols from /usr/sfw/lib/libpng.so.2...done.
Loaded symbols for /usr/sfw/lib/libpng.so.2
Reading symbols from /usr/sfw/lib/libjpeg.so.62...done.
Loaded symbols for /usr/sfw/lib/libjpeg.so.62
Reading symbols from
/opt/php4/lib/php/extensions/no-debug-non-zts-20020429/ldap.so...done.
Loaded symbols for
/opt/php4/lib/php/extensions/no-debug-non-zts-20020429/ldap.so
Reading symbols from /usr/local/lib/libldap.so.2...done.
Loaded symbols for /usr/local/lib/libldap.so.2
Reading symbols from /usr/local/lib/liblber.so.2...done.
Loaded symbols for /usr/local/lib/liblber.so.2
Reading symbols from
/opt/php4/lib/php/extensions/no-debug-non-zts-20020429/bcmath.so...done.
Loaded symbols for
/opt/php4/lib/php/extensions/no-debug-non-zts-20020429/bcmath.so
Reading symbols from
/opt/php4/lib/php/extensions/no-debug-non-zts-20020429/calendar.so...done.
Loaded symbols for
/opt/php4/lib/php/extensions/no-debug-non-zts-20020429/calendar.so
Reading symbols from
/opt/php4/lib/php/extensions/no-debug-non-zts-20020429/dba.so...done.
Loaded symbols for
/opt/php4/lib/php/extensions/no-debug-non-zts-20020429/dba.so
Reading symbols from /usr/local/lib/libdb-4.1.so...done.
Loaded symbols for /usr/local/lib/libdb-4.1.so
Reading symbols from
/opt/php4/lib/php/extensions/no-debug-non-zts-20020429/domxml.so...done.
Loaded symbols for
/opt/php4/lib/php/extensions/no-debug-non-zts-20020429/domxml.so
Reading symbols from /usr/local/lib/libe

#26300 [NEW]: PB serialization ascii code greater than 127

[pierre-marie dot mouliere at arc-intl dot com]

From: pierre-marie dot mouliere at arc-intl dot com
Operating system: HPUX
PHP version:  4.3.3
PHP Bug Type: WDDX related
Bug description:  PB serialization ascii code greater than 127

Description:

Hi,
There's a little bug with strings serialization, for strings containing
characters with ascii code greater than 127 (with HP-UX v11).

We found that the problem is located in function  :
php_wddx_serialize_string.
There is a line of code like :
 if (iscntrl((int)*p) {
When casting character (*p) to int for values greater than 127 (witch are
considered negative) there is a sign extension so character 'ñ'(ascii code
0xf0) become integer 0xfff0.
But the function iscntrl expects an integrer between 0 and 255, or the
result is unpredictable.
The problem can be fixed by modifying the line above like :
 if(iscntrl((int)(unsigned char)*p) {   
Best Regards
Pierre-Marie MOULIERE

Reproduce code:
---
Without locale :";
print "àâäçéèêëîïôù ÀÂÄÉÈÊËÎÏÔÙ ñ Ñ" . "\n";
print wddx_deserialize(wddx_serialize_value("àâäçéèêëîïôù ÀÂÄÉÈÊËÎÏÔÙ ñ
Ñ")); 
print "";
print "With locale (fr_FR.iso88591) :";
setlocale(LC_CTYPE, "fr_FR.iso88591");
print "àâäçéèêëîïôù ÀÂÄÉÈÊËÎÏÔÙ ñ Ñ" . "\n";
print wddx_deserialize(wddx_serialize_value("àâäçéèêëîïôù ÀÂÄÉÈÊËÎÏÔÙ ñ
Ñ")); 
?>  


Expected result:

Result : 
Without locale :

àâäçéèêëîïôù ÀÂÄÉÈÊËÎÏÔÙ ñ Ñ
àâäçéèêëîïôù ÀÂÄÉÈÊËÎÏÔÙ ñ Ñ

---

With locale (fr_FR.iso88591) :

àâäçéèêëîïôù ÀÂÄÉÈÊËÎÏÔÙ ñ Ñ
àâäçéèêëîïôù ÀÂÄÉÈÊËÎÏÔÙ ñ Ñ



Actual result:
--
Result : 
Without locale :

àâäçéèêëîïôù ÀÂÄÉÈÊËÎÏÔÙ ñ Ñ
àâäçéèêëîïôù ÀÂÄÉÈÊËÎÏÔÙ ? Ñ

---

With locale (fr_FR.iso88591) :

àâäçéèêëîïôù ÀÂÄÉÈÊËÎÏÔÙ ñ Ñ
àâäçéèêëîïôù ÀÂÄ?ÈÊËÎÏÔÙ ñ Ñ

-- 
Edit bug report at http://bugs.php.net/?id=26300&edit=1
-- 
Try a CVS snapshot (php4):  http://bugs.php.net/fix.php?id=26300&r=trysnapshot4
Try a CVS snapshot (php5):  http://bugs.php.net/fix.php?id=26300&r=trysnapshot5
Fixed in CVS:   http://bugs.php.net/fix.php?id=26300&r=fixedcvs
Fixed in release:   http://bugs.php.net/fix.php?id=26300&r=alreadyfixed
Need backtrace: http://bugs.php.net/fix.php?id=26300&r=needtrace
Try newer version:  http://bugs.php.net/fix.php?id=26300&r=oldversion
Not developer issue:http://bugs.php.net/fix.php?id=26300&r=support
Expected behavior:  http://bugs.php.net/fix.php?id=26300&r=notwrong
Not enough info:http://bugs.php.net/fix.php?id=26300&r=notenoughinfo
Submitted twice:http://bugs.php.net/fix.php?id=26300&r=submittedtwice
register_globals:   http://bugs.php.net/fix.php?id=26300&r=globals
PHP 3 support discontinued: http://bugs.php.net/fix.php?id=26300&r=php3
Daylight Savings:   http://bugs.php.net/fix.php?id=26300&r=dst
IIS Stability:  http://bugs.php.net/fix.php?id=26300&r=isapi
Install GNU Sed:http://bugs.php.net/fix.php?id=26300&r=gnused
Floating point limitations: http://bugs.php.net/fix.php?id=26300&r=float

#26299 [NEW]: Inheritance problem: an object can have two same-named private variables

[forseti at oak dot rpg dot pl]

From: forseti at oak dot rpg dot pl
Operating system: Windows 98 SE
PHP version:  5.0.0b2 (beta2)
PHP Bug Type: Zend Engine 2 problem
Bug description:  Inheritance problem: an object can have two same-named private 
variables

Description:

When extending a class, private member variables from parent class are
retained even if child class defines same-named variable. 

Reproduce code:
---
';print_r(new B);echo '';
?>

Expected result:

b Object
(
[a:private] => From class B
)



Actual result:
--
b Object
(
[a:private] => From class B
[a:private] => From class A
)



-- 
Edit bug report at http://bugs.php.net/?id=26299&edit=1
-- 
Try a CVS snapshot (php4):  http://bugs.php.net/fix.php?id=26299&r=trysnapshot4
Try a CVS snapshot (php5):  http://bugs.php.net/fix.php?id=26299&r=trysnapshot5
Fixed in CVS:   http://bugs.php.net/fix.php?id=26299&r=fixedcvs
Fixed in release:   http://bugs.php.net/fix.php?id=26299&r=alreadyfixed
Need backtrace: http://bugs.php.net/fix.php?id=26299&r=needtrace
Try newer version:  http://bugs.php.net/fix.php?id=26299&r=oldversion
Not developer issue:http://bugs.php.net/fix.php?id=26299&r=support
Expected behavior:  http://bugs.php.net/fix.php?id=26299&r=notwrong
Not enough info:http://bugs.php.net/fix.php?id=26299&r=notenoughinfo
Submitted twice:http://bugs.php.net/fix.php?id=26299&r=submittedtwice
register_globals:   http://bugs.php.net/fix.php?id=26299&r=globals
PHP 3 support discontinued: http://bugs.php.net/fix.php?id=26299&r=php3
Daylight Savings:   http://bugs.php.net/fix.php?id=26299&r=dst
IIS Stability:  http://bugs.php.net/fix.php?id=26299&r=isapi
Install GNU Sed:http://bugs.php.net/fix.php?id=26299&r=gnused
Floating point limitations: http://bugs.php.net/fix.php?id=26299&r=float

#26249 [Opn]: snmpget doesn't free memory

[dacol dot jeremy at caramail dot com]

 ID:   26249
 User updated by:  dacol dot jeremy at caramail dot com
 Reported By:  dacol dot jeremy at caramail dot com
 Status:   Open
 Bug Type: SNMP related
 Operating System: linux
 PHP Version:  4.3.2
 New Comment:

i test a script with:
while(true)
{
system("snmpget.");
}
and see with "top" command the memory used : 0.7% all time.
it doesn't depend on the number of snmpget so i think it's a php
problem.


Previous Comments:


[2003-11-18 06:58:02] dacol dot jeremy at caramail dot com

hi,

i don't know how upgrad my php files with the latest version...

i up my old ucd-snmp version to net-snmp 5.1 and my bug still
happened.

please tell me exactly what to do with php4-STABLE-latest.tar.gz

the command, all things tou can

i am really sorry but i search on web and i find nothing on it.



[2003-11-17 17:52:10] [EMAIL PROTECTED]

Update your über-old ucd-snmp to the latest net-snmp version.
Then reopen if you still experience leaks which you can trace to happen
in PHP code (not in the net-snmp)




[2003-11-17 05:40:55] dacol dot jeremy at caramail dot com

hi 
i have copied the CVS on my php files and replace it.
but it doesn't work each time i use snmpget function it takes memory
which it doesn't free after a response.
in fact each 100 snmpget use 0.1% of my memory.
it's really the snmpget function which doesn't work.



[2003-11-17 03:06:59] response dot sorry at notremember dot pass

hi,

thank you for your answer.

can i only copy this on my existing php files or do i have to recompile
all?
in fact what i have to do with this file?

and sorry for the mistake i work with:
apache 1.3.28
php 4.3.3
ucd-snmp 4.2.2
on linux RH 7.1



[2003-11-14 10:36:16] [EMAIL PROTECTED]

Please try using this CVS snapshot:

  http://snaps.php.net/php4-STABLE-latest.tar.gz
 
For Windows:
 
  http://snaps.php.net/win32/php4-win32-STABLE-latest.zip





The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
http://bugs.php.net/26249

-- 
Edit this bug report at http://bugs.php.net/?id=26249&edit=1

#26264 [Opn]: Compile failure with iconv enabled

[Joerg dot Dieter dot Friedrich at uni-konstanz dot de]

 ID:   26264
 User updated by:  Joerg dot Dieter dot Friedrich at uni-konstanz dot de
 Reported By:  Joerg dot Dieter dot Friedrich at uni-konstanz dot de
 Status:   Open
 Bug Type: ICONV related
 Operating System: Solaris 9
 PHP Version:  4.3.4
 New Comment:

Hi again.
I tested a lot these weekend.
Summary:

OS: Solaris 9

I want to use horde cvs which supports utf-8 coding of each page
displayed.

I compiled php with --with-iconv=shared --enable-mbstring=shared,all.

No libiconv was installed. Solaris libc provides these
iconv-functions.

php always segfaults when charset is set to utf-8.

Then I tried to use GNU's libiconv, but php still segfaults.

then I tried to --enable-debug.
no segfaults anymore, there is a 
Warning: Unknown(): A session is active. You cannot change the session
module's ini settings at this time. in Unknown on line 0

btw no segfault no coredump.

I dont know what to do and how to debug this beast.


Previous Comments:


[2003-11-17 21:54:39] [EMAIL PROTECTED]

see also bug #25281




[2003-11-17 12:50:50] [EMAIL PROTECTED]

I should have been a bit more specific about my 
suggestion. I wrote it that way because you were 
implying you have libiconv installed also.

If you have both in your machine, it's possible that 
iconv extension wrongly detects the library to be used.





[2003-11-16 11:01:53] Joerg dot Dieter dot Friedrich at uni-konstanz
dot de

Why should this be needed. I haven't installed any libiconv in
/usr/local. Solaris provides the iconv in libc and the headers are in
/usr/include/iconv.h



[2003-11-16 06:11:34] [EMAIL PROTECTED]

Try giving the prefix of iconv library to the configure 
option, like --with-iconv=shared,/usr/local




[2003-11-16 04:31:13] Joerg dot Dieter dot Friedrich at uni-konstanz
dot de

Here you are:
./configure  --prefix=/opt/php4 --with-apxs=/opt/apache/bin/apxs
--sysconfdir=/opt/etc/php4 --with-config-file-path=/opt/etc/php4
--disable-debug --with-gd=shared --with-jpeg-dir=/usr/sfw
--with-png-dir=/usr/sfw --with-zlib=shared
--with-imap=shared,/usr/local/src/imap-2002e
--with-mysql=shared,/opt/mysql --with-mysql-sock=/var/run/mysql/sock
--enable-dba=shared --with-db4=shared,/usr/local --with-ndbm=shared
--with-ldap=shared,/usr/local --with-openssl=/usr/local/ssl
--with-gettext --with-pear --enable-safe-mode --enable-bcmath=shared
--enable-calendar=shared --enable-ftp=shared --enable-sysvmsg
--enable-sysvsem --enable-sysvshm --with-mcrypt=shared
--with-xml=shared --with-dom=shared --enable-mime-magic
--with-iconv=shared --with-mbstring=all



The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
http://bugs.php.net/26264

-- 
Edit this bug report at http://bugs.php.net/?id=26264&edit=1

#14066 [Com]: Can't suppress warnigns on accessing invalid string offset

[php at pleaseletusknow dot com]

 ID:   14066
 Comment by:   php at pleaseletusknow dot com
 Reported By:  mfischer at guru dot josefine dot at
 Status:   Wont fix
 Bug Type: Scripting Engine problem
 Operating System: Any
 PHP Version:  4.0CVS-2001-11-15
 New Comment:

This error, in latest PHP versions, only occurs when your trying to
treat a plain variable as an array.

In most occasions you can get round this by adding in an isArray check
before you try to use the var as an array.

For example;

Problem:
if (isset($parameter['align'])) { echo 'moo'; } // error: Uninitialized
string offset 

Solution:
if (!isarray($parameter)) { $parameter = array(); }
if (isset($parameter['align'])) { echo 'moo'; }

So your basically setting an empty array in place of an empty variable.


Previous Comments:


[2003-04-25 11:07:59] [EMAIL PROTECTED]

Since we now have this 'wont fix' status, let's use it. :)




[2002-03-03 17:47:39] [EMAIL PROTECTED]

It's an engine thing, and can really not be fixed. The accessing of the
{3} is done after the END_SILENCE op at the CONCAT op.
I'll mark it as suspended for now.



[2002-03-03 17:35:33] [EMAIL PROTECTED]

I'm not satisfied with this answer :-)

Why does it work with arrays but not with strings (and, bert, yeah I
know they aren't the same).

Just because "it's strings here" doesn't legitimate that a warnings is
thrown. Using your suggestion is rather cumbersome.

Is it an implemention issue? Can it be solve to be less long winded?

"it is"

a bug imho :)



[2002-03-03 17:35:31] [EMAIL PROTECTED]

I'm not satisfied with this answer :-)

Why does it work with arrays but not with strings (and, bert, yeah I
know they aren't the same).

Just because "it's strings here" doesn't legitimate that a warnings is
thrown. Using your suggestion is rather cumbersome.

Is it an implemention issue? Can it be solve to be less long winded?

"it is"

a bug imho :)



[2002-03-03 17:28:03] [EMAIL PROTECTED]

It's not a bug, $foo{3}. "\n" is done before the @, so that displays
the error.
so I'm closing this, this works BTW:
$foo = "foo"; echo @($foo{3}. "\n");

Derick



The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
http://bugs.php.net/14066

-- 
Edit this bug report at http://bugs.php.net/?id=14066&edit=1

#26288 [Opn]: Segmentation Fault while getting NULL values from a row of a table

[kiranhariharan at in dot ibm dot com]

 ID:   26288
 User updated by:  kiranhariharan at in dot ibm dot com
 Reported By:  kiranhariharan at in dot ibm dot com
 Status:   Open
 Bug Type: ODBC related
 Operating System: x86_64 SuSE Linux
-PHP Version:  4.3.4
+PHP Version:  4.3.5 (built: Nov 18 2003 19:22:58)
 New Comment:

Hi,
 I have tried using cvs snapshot it shows similar problem in 4.3.5
(built: Nov 18 2003 19:22:58)

Thank you


Previous Comments:


[2003-11-18 07:55:04] kiranhariharan at in dot ibm dot com

I have faced the same problem as mentioned with PHP v4.3.4 also. 

Could someone help me to tell what is going wrong?



[2003-11-17 14:22:27] [EMAIL PROTECTED]

Please try using this CVS snapshot:

  http://snaps.php.net/php4-STABLE-latest.tar.gz
 
For Windows:
 
  http://snaps.php.net/win32/php4-win32-STABLE-latest.zip





[2003-11-17 09:43:51] kiranhariharan at in dot ibm dot com

Description:

Hi,

I have installed PHP 4.2.2 on x86_64_AMD machine having SuSE Linux
installed on it. I have a DB2 database which has several tables in it.

I wrote a PHP code which fetches data from a row, that has null values
in it using odbc_fetch_row(), I get segmentation fault. The same PHP
script to access the table runs fine on Windows OS (to mention the PHP
version I have on Windows OS is 4.3.3). I beleive there is some problem
when it comes to handling the NULL values from a table at this
version.

1) Can someone please let me know whether there are any reported bugs
with 4.2.2 PHP version on Linux platforms. If that is so, also please
advice me to which version of PHP I would have to move (a stable PHP
version).

2) Also, is there any seperate PHP 4.3 rpm for x86_64AMD architecture
and where to download it from?


Thanks in advance,
Kiran H.






-- 
Edit this bug report at http://bugs.php.net/?id=26288&edit=1

#26248 [Opn]: Can't compile PHP (both 4 and 5) with IMAP

[lists at nutshell dot nu]

 ID:   26248
 User updated by:  lists at nutshell dot nu
 Reported By:  lists at nutshell dot nu
 Status:   Open
 Bug Type: Compile Failure
 Operating System: Linux (Mandrake 9.2)
 PHP Version:  5.0.0b2 (beta2)
 New Comment:

A few notes to my last comment.

To be able to compile php (4/5) you either have to compile imap from a
tar ball or change the imap.spec so it compiles the libc-client-PHP4.a
with kerberos (it only compiles with ssl default).

My experience here is that it is impossible to compile imap support for
php unless you have kerberos. In my original post I show that I tried
with both --without-kerberos and --without-imap-ssl, but the
compilation failed due to that libc-client had no kerberos support.


Previous Comments:


[2003-11-17 20:33:49] lists at nutshell dot nu

I think I have found the problem.

Mandrake builds three c-clients.

libc-client.a
libc-client-PHP4.a
libc-client-nossl.a

Note than the -PHP4 is built without kerberos support.

I renamed libc-client.a to libc-client-org.a and libc-client-PHP4.a to
libc-client.a.

Then PHP5 compiles just fine with a shared imap.

Is there any way of building without having to rename the libs?



[2003-11-16 14:25:38] lists at nutshell dot nu

Yes, as stated in my original post it fails both with and without
shared. I have tried every possible combination I can think of.

It compiles just fine without imap, but I need imap.

Can also add that Mandrake 9.2 comes with a php imap rpm and the only
things I compile from tar-balls are apache and php4/5.



[2003-11-16 13:16:33] [EMAIL PROTECTED]

Does it fail if you do not compile imap as a shared extension?



[2003-11-14 04:52:38] lists at nutshell dot nu

Description:

Please note that Mandrake 9.2 comes with a mod-imap for php4, but I
wanted to compile it to suite my needs.

I have tryed compiling php5 (b2 and later snapshots, as well as latest
official php4) with imap. I have tried with Mandrake 9.2's own
imap-2002d, recompiling it with "rpm -bb --target i686", as well as
downgrading to imap-2001a.

I have tried to configure with:
--with-imap=shared
--with-imap-ssl
or
--with-imap=shared
--without-imap-ssl
or
--with-imap=shared

Plus also adding "/usr" as dir and without shared.

Note that if I try to configure "--without-kerberos" the configure
stops at imap.

I have all normal and devel packages installed for imap, openssl and
keberos, as well as any util packages for them.

./configure works just fine, but when I run make I get errors.

I get the following error:
/usr/local/src/php5-200311122030/ext/imap/php_imap.c: In function
`zm_startup_imap':
/usr/local/src/php5-200311122030/ext/imap/php_imap.c:445: error:
`auth_gss' undeclared (first use in this function)
/usr/local/src/php5-200311122030/ext/imap/php_imap.c:445: error: (Each
undeclared identifier is reported only once
/usr/local/src/php5-200311122030/ext/imap/php_imap.c:445: error: for
each function it appears in.)
make: *** [ext/imap/php_imap.lo] Error 1






-- 
Edit this bug report at http://bugs.php.net/?id=26248&edit=1

#26298 [Bgs]: Intarbase fetch record of wide field freeze PHP execution

[abies]

 ID:   26298
 Updated by:   [EMAIL PROTECTED]
 Reported By:  tma at 2p dot cz
 Status:   Bogus
 Bug Type: Reproducible crash
 Operating System: Win98
 PHP Version:  4.3.4
 New Comment:

1000 records 
 


Previous Comments:


[2003-11-18 07:54:24] tma at 2p dot cz

Did you check it with data in table? How much?



[2003-11-18 07:45:08] [EMAIL PROTECTED]

This exact sample script works fine on Windows 2000 with 
Firebird 1.0.3 & CLI php4. [25+ runs with both Debug and 
Release builds] 
 
 
 
 
 



[2003-11-18 06:36:46] tma at 2p dot cz

Description:

It seems there is some problem when fetching interbase records
containing a wide field.

Execution of PHP code is frozen at random point - probably some memory
problem. Point of freezing depends on number of fetched records. 

Does not matter if ibase_fetch_row/assoc/object is used. There is one
presumtion - must be selected a field that is type of VARCHAR(200). I
did not test what is critical field length. There are in field national
chars but this occurs even pure english chars are used
(#32..#127)(first I've used UNICODE_FSS and thought that problem is in
unicode). If this wide column is not listed it works (SELECT ID,ID as
NAME). 

Does not matter if NAME's CHARACTER SET is ASCII or UNICODE_FSS

CREATE TABLE TBL (
   ID VARCHAR(10) CHARACTER SET ASCII NOT NULL,
   NAME VARCHAR(200) CHARACTER SET ASCII /* UNICODE_FSS*/ NOT
NULL,
   PARENT_ID VARCHAR(10) CHARACTER SET ASCII,
 CONSTRAINT GROUPS_PK PRIMARY KEY (ID)
);

Note:
Problem is neither in IBServer nor in IBClient. When using windows
based apps using gds32.dll it works.

I cannot fix it by myself because I have not MS-C to compile module for
Windows and I'm not expert for CYGWIN environment.

Reproduce code:
---
function CleanUp() {
global $IB_HANDLE;
  if ($IB_HANDLE) {
echo "close\n";
IBase_Close($IB_HANDLE);
echo "close done\n";
  }
}
Register_ShutDown_Function('CleanUp');

$maxn = 1000;  // limit max num. of listed records

Dl('php_interbase.dll');
$IB_HANDLE = IBase_Connect (IB_HOST, IB_USER, IB_PASSWORD);

$query = IBase_Query($IB_HANDLE, "SELECT * FROM TBL");
$n = 0;  while ($res = IBase_Fetch_Assoc($query)) {
  echo "$n.";
  printf("%s: %s\n", $res['ID'], $res['NAME']); 
  Flush();  // no effect
  Unset($res);
  $n++;
  if ($n > $maxn) {
break;
  }
}
echo "loop done\n";
IBase_Free_Result($query);
echo "free query done\n";


Expected result:

-open ibase
-open SQL
-list table records.
-close SQL
-close ibase

Actual result:
--
Execution of PHP code is frozen at random point - probably some memory
problem. Point of freeze depends on number of fetched records. For
example in my environment if $maxn < 6 it works, if $maxn < 100 it's
frozen in when ibase_free_result in action - if this line is commented
is frozen during ibase_close. If $maxn = 1000 is frozen when listing
record 355. But this depends on output and ASCII/UNICODE_FSS.

Notes:

Execution is frozen because time_limit has no effect.

If was script executed from Apache as CGI and execution was stopped
from browser there was still connection to IBServer. If was executed
from command line and pressed Ctrl+C when frozen, connection was
closed

It's not time_limit or output problem, this occurs only if
ibase_fetch_* called.





-- 
Edit this bug report at http://bugs.php.net/?id=26298&edit=1

#26288 [Fbk->Opn]: Segmentation Fault while getting NULL values from a row of a table

[kiranhariharan at in dot ibm dot com]

 ID:   26288
 User updated by:  kiranhariharan at in dot ibm dot com
-Summary:  Segmentation Fault while getting NULL values from a
   row of a table(PHP v 4.2.2)
 Reported By:  kiranhariharan at in dot ibm dot com
-Status:   Feedback
+Status:   Open
 Bug Type: ODBC related
 Operating System: x86_64 SuSE Linux
-PHP Version:  4.3.3
+PHP Version:  4.3.4
 New Comment:

I have faced the same problem as mentioned with PHP v4.3.4 also. 

Could someone help me to tell what is going wrong?


Previous Comments:


[2003-11-17 14:22:27] [EMAIL PROTECTED]

Please try using this CVS snapshot:

  http://snaps.php.net/php4-STABLE-latest.tar.gz
 
For Windows:
 
  http://snaps.php.net/win32/php4-win32-STABLE-latest.zip





[2003-11-17 09:43:51] kiranhariharan at in dot ibm dot com

Description:

Hi,

I have installed PHP 4.2.2 on x86_64_AMD machine having SuSE Linux
installed on it. I have a DB2 database which has several tables in it.

I wrote a PHP code which fetches data from a row, that has null values
in it using odbc_fetch_row(), I get segmentation fault. The same PHP
script to access the table runs fine on Windows OS (to mention the PHP
version I have on Windows OS is 4.3.3). I beleive there is some problem
when it comes to handling the NULL values from a table at this
version.

1) Can someone please let me know whether there are any reported bugs
with 4.2.2 PHP version on Linux platforms. If that is so, also please
advice me to which version of PHP I would have to move (a stable PHP
version).

2) Also, is there any seperate PHP 4.3 rpm for x86_64AMD architecture
and where to download it from?


Thanks in advance,
Kiran H.






-- 
Edit this bug report at http://bugs.php.net/?id=26288&edit=1

#26298 [Bgs]: Intarbase fetch record of wide field freeze PHP execution

[tma at 2p dot cz]

 ID:   26298
 User updated by:  tma at 2p dot cz
 Reported By:  tma at 2p dot cz
 Status:   Bogus
 Bug Type: Reproducible crash
 Operating System: Win98
 PHP Version:  4.3.4
 New Comment:

Did you check it with data in table? How much?


Previous Comments:


[2003-11-18 07:45:08] [EMAIL PROTECTED]

This exact sample script works fine on Windows 2000 with 
Firebird 1.0.3 & CLI php4. [25+ runs with both Debug and 
Release builds] 
 
 
 
 
 



[2003-11-18 06:36:46] tma at 2p dot cz

Description:

It seems there is some problem when fetching interbase records
containing a wide field.

Execution of PHP code is frozen at random point - probably some memory
problem. Point of freezing depends on number of fetched records. 

Does not matter if ibase_fetch_row/assoc/object is used. There is one
presumtion - must be selected a field that is type of VARCHAR(200). I
did not test what is critical field length. There are in field national
chars but this occurs even pure english chars are used
(#32..#127)(first I've used UNICODE_FSS and thought that problem is in
unicode). If this wide column is not listed it works (SELECT ID,ID as
NAME). 

Does not matter if NAME's CHARACTER SET is ASCII or UNICODE_FSS

CREATE TABLE TBL (
   ID VARCHAR(10) CHARACTER SET ASCII NOT NULL,
   NAME VARCHAR(200) CHARACTER SET ASCII /* UNICODE_FSS*/ NOT
NULL,
   PARENT_ID VARCHAR(10) CHARACTER SET ASCII,
 CONSTRAINT GROUPS_PK PRIMARY KEY (ID)
);

Note:
Problem is neither in IBServer nor in IBClient. When using windows
based apps using gds32.dll it works.

I cannot fix it by myself because I have not MS-C to compile module for
Windows and I'm not expert for CYGWIN environment.

Reproduce code:
---
function CleanUp() {
global $IB_HANDLE;
  if ($IB_HANDLE) {
echo "close\n";
IBase_Close($IB_HANDLE);
echo "close done\n";
  }
}
Register_ShutDown_Function('CleanUp');

$maxn = 1000;  // limit max num. of listed records

Dl('php_interbase.dll');
$IB_HANDLE = IBase_Connect (IB_HOST, IB_USER, IB_PASSWORD);

$query = IBase_Query($IB_HANDLE, "SELECT * FROM TBL");
$n = 0;  while ($res = IBase_Fetch_Assoc($query)) {
  echo "$n.";
  printf("%s: %s\n", $res['ID'], $res['NAME']); 
  Flush();  // no effect
  Unset($res);
  $n++;
  if ($n > $maxn) {
break;
  }
}
echo "loop done\n";
IBase_Free_Result($query);
echo "free query done\n";


Expected result:

-open ibase
-open SQL
-list table records.
-close SQL
-close ibase

Actual result:
--
Execution of PHP code is frozen at random point - probably some memory
problem. Point of freeze depends on number of fetched records. For
example in my environment if $maxn < 6 it works, if $maxn < 100 it's
frozen in when ibase_free_result in action - if this line is commented
is frozen during ibase_close. If $maxn = 1000 is frozen when listing
record 355. But this depends on output and ASCII/UNICODE_FSS.

Notes:

Execution is frozen because time_limit has no effect.

If was script executed from Apache as CGI and execution was stopped
from browser there was still connection to IBServer. If was executed
from command line and pressed Ctrl+C when frozen, connection was
closed

It's not time_limit or output problem, this occurs only if
ibase_fetch_* called.





-- 
Edit this bug report at http://bugs.php.net/?id=26298&edit=1

#25970 [Bgs]: Session management fails when I stop the browser while a PHP page is loading.

[bueno at catho dot com dot br]

 ID:   25970
 User updated by:  bueno at catho dot com dot br
-Reported By:  catia at catho dot com dot br
+Reported By:  bueno at catho dot com dot br
 Status:   Bogus
 Bug Type: Session related
 Operating System: Debian 3.0 Linux 2.4.20
 PHP Version:  4.3.4
 New Comment:

Please,
can you send me some of this sites in a private message ?


Previous Comments:


[2003-11-17 18:12:15] [EMAIL PROTECTED]

Can not reproduce with 10 sites heavily dependant on sessions.
Check your scripts.




[2003-11-12 11:00:46] catia at catho dot com dot br

I try to use this snapshot and the problem was not solved...
Then, I install php 4.3.4 and apache 1.3.29... and the problem still
happens...



[2003-10-25 20:14:35] catia at catho dot com dot br

ignore_user_abort(true)  didn´t help...

some how, system can´t see that php fails, or take too much time, to
delivery the page and the user press "Reload"

May we set some time limit to the session file be lock in a single
request?



[2003-10-23 22:22:07] [EMAIL PROTECTED]

ignore_user_abort(true); is always a good idea.

But try a newer version, the problem may have went away.



[2003-10-23 15:45:57] bueno at catho dot com dot br

Description:

Session management fails when I stop the browser while a PHP page is
loading.
As soon as the browser requests a file with a session_start or
session_register command the browser gets in a
kinda endless loop and the session file stay open. The browser keeps
waiting and waiting and nothing happens (nothing appears on the
screen).

I tryed to list the open files on system (lsof) when this happen.
And, see my session file opened by a httpd process for 20 minutes, even
when the browser has canceled the requisition. 

While the command : lsof | grep sess_SESSIONID return the file open, I
could not access any page with session_start() or session_register().

Pparently the apache process has locked the file, and any other process
can read it (the session file)

I'm using PHP4.3.2 / Debian 3.0 Linux 2.4.20 / Apache 1.3.28
My configure command:
'./configure'
'--with-mysql=../mysql/mysql-standard-4.0.12-pc-linux-i686'
'--with-apache=../../apache/build_deb/deb2/apache-1.3.28'
'--enable-track-vars'
'--disable-debug' '--with-gd' '--enable-ftp'
'--enable-inline-optimization'
'--enable-sockets' '--with-zlib' 






-- 
Edit this bug report at http://bugs.php.net/?id=25970&edit=1

#26298 [Opn->Bgs]: Intarbase fetch record of wide field freeze PHP execution

[abies]

 ID:   26298
 Updated by:   [EMAIL PROTECTED]
 Reported By:  tma at 2p dot cz
-Status:   Open
+Status:   Bogus
 Bug Type: Reproducible crash
 Operating System: Win98
 PHP Version:  4.3.4
 New Comment:

This exact sample script works fine on Windows 2000 with 
Firebird 1.0.3 & CLI php4. [25+ runs with both Debug and 
Release builds] 
 
 
 
 
 


Previous Comments:


[2003-11-18 06:36:46] tma at 2p dot cz

Description:

It seems there is some problem when fetching interbase records
containing a wide field.

Execution of PHP code is frozen at random point - probably some memory
problem. Point of freezing depends on number of fetched records. 

Does not matter if ibase_fetch_row/assoc/object is used. There is one
presumtion - must be selected a field that is type of VARCHAR(200). I
did not test what is critical field length. There are in field national
chars but this occurs even pure english chars are used
(#32..#127)(first I've used UNICODE_FSS and thought that problem is in
unicode). If this wide column is not listed it works (SELECT ID,ID as
NAME). 

Does not matter if NAME's CHARACTER SET is ASCII or UNICODE_FSS

CREATE TABLE TBL (
   ID VARCHAR(10) CHARACTER SET ASCII NOT NULL,
   NAME VARCHAR(200) CHARACTER SET ASCII /* UNICODE_FSS*/ NOT
NULL,
   PARENT_ID VARCHAR(10) CHARACTER SET ASCII,
 CONSTRAINT GROUPS_PK PRIMARY KEY (ID)
);

Note:
Problem is neither in IBServer nor in IBClient. When using windows
based apps using gds32.dll it works.

I cannot fix it by myself because I have not MS-C to compile module for
Windows and I'm not expert for CYGWIN environment.

Reproduce code:
---
function CleanUp() {
global $IB_HANDLE;
  if ($IB_HANDLE) {
echo "close\n";
IBase_Close($IB_HANDLE);
echo "close done\n";
  }
}
Register_ShutDown_Function('CleanUp');

$maxn = 1000;  // limit max num. of listed records

Dl('php_interbase.dll');
$IB_HANDLE = IBase_Connect (IB_HOST, IB_USER, IB_PASSWORD);

$query = IBase_Query($IB_HANDLE, "SELECT * FROM TBL");
$n = 0;  while ($res = IBase_Fetch_Assoc($query)) {
  echo "$n.";
  printf("%s: %s\n", $res['ID'], $res['NAME']); 
  Flush();  // no effect
  Unset($res);
  $n++;
  if ($n > $maxn) {
break;
  }
}
echo "loop done\n";
IBase_Free_Result($query);
echo "free query done\n";


Expected result:

-open ibase
-open SQL
-list table records.
-close SQL
-close ibase

Actual result:
--
Execution of PHP code is frozen at random point - probably some memory
problem. Point of freeze depends on number of fetched records. For
example in my environment if $maxn < 6 it works, if $maxn < 100 it's
frozen in when ibase_free_result in action - if this line is commented
is frozen during ibase_close. If $maxn = 1000 is frozen when listing
record 355. But this depends on output and ASCII/UNICODE_FSS.

Notes:

Execution is frozen because time_limit has no effect.

If was script executed from Apache as CGI and execution was stopped
from browser there was still connection to IBServer. If was executed
from command line and pressed Ctrl+C when frozen, connection was
closed

It's not time_limit or output problem, this occurs only if
ibase_fetch_* called.





-- 
Edit this bug report at http://bugs.php.net/?id=26298&edit=1

#26117 [Bgs->Opn]: Persistent connection not reused

[spam at vrana dot cz]

 ID:   26117
 User updated by:  spam at vrana dot cz
 Reported By:  spam at vrana dot cz
-Status:   Bogus
+Status:   Open
 Bug Type: MySQL related
 Operating System: Linux
 PHP Version:  4.3.3
 New Comment:

Can you please read my bug report carefuly before stating bug as bogus?
Persistent connection in PHP should reuse previous persistent
connection with the same host, user and password. The bug is that it
doesn't.


Previous Comments:


[2003-11-17 18:21:48] [EMAIL PROTECTED]

And this is PHP bug in where..?




[2003-11-05 12:20:11] spam at vrana dot cz

Yes. There were 466 processes with the same user, host and db. All
processes were in state Sleep.



[2003-11-05 11:21:33] [EMAIL PROTECTED]

When the "too many connections" problem occurs, have you tried running
"show full processlist;" query to see if infact all connections use the
same database?



[2003-11-05 03:24:59] spam at vrana dot cz

This is not the case. Believe me that I have read all similar bugs and
this is different. I also have read carefuly the manual and I
understand everything written there about persistent connections.

There is written in the Persistent Database Connections chapter of
manual: An 'identical' connection is a connection that was opened to
the same host, with the same username and the same password (where
applicable). So it's not true that by mysql_select_db() with different
database name the connection can't be reused. Anyway I use the same
database in all scripts.

I know that connection is persistent only over the child. Thus I wrote
that Apache configuration directive MaxClients is set to 150.



[2003-11-04 13:53:56] [EMAIL PROTECTED]

Thank you for taking the time to write to us, but this is not
a bug. Please double-check the documentation available at
http://www.php.net/manual/ and the instructions on how to report
a bug at http://bugs.php.net/how-to-report.php

If you go mysql_select_db() then the persistent connection you've
created becomes specific to that particular database. Which may explain
why those connections are not re-used.
Also, persistent connections are per-child, meaning that if 1 Apache
child opens connections only that child has a pre-existing connection
avaliable. If a subsequent request is handled by another child, which
does not yet have a MySQL connection, it'll create a new connection.



The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
http://bugs.php.net/26117

-- 
Edit this bug report at http://bugs.php.net/?id=26117&edit=1

#25777 [Com]: char and varchar fields are being rtrimmed (and also ltrimmed?) using freetds

[klaus dot albert at techconsult dot ie]

 ID:   25777
 Comment by:   klaus dot albert at techconsult dot ie
 Reported By:  duh at dowebwedo dot com
 Status:   Closed
 Bug Type: MSSQL related
 Operating System: Debian GNU/Linux 3.0
 PHP Version:  4.3.4RC1
 New Comment:

Description
---
mssql_fetch_row and mssql_fetch_array return a single space 
for columns that contain an empty string

Reproduce
-
$db=mssql_pconnect('server','user','pass');
$result=mssql_query("UPDATE table SET column='' WHERE 
id='1'",$db);
$result=mssql_query("SELECT column FROM table WHERE 
id='1'",$db);
echo strlen($result[0]);

Expected result
---
Expected length is 0, gotten result is 1 (a single space).
PHP 4.3.2 returns correct empty result.

Enviroment
--
Win2000 SP4, IIS5, MSSQL 2000
also reproducable on WinNT SP6, IIS5, MSSQL 7

Remarks
---
see Bug #9854


Previous Comments:


[2003-11-17 18:14:15] [EMAIL PROTECTED]

The sybase and mssql extensions do not handle returned data the exact
same way, and it might be that we need to readd the rtrim code to the
sybase extensions. It seams like the dbconvert() finctions are
implemented differently.

You could also try the native mssql extension. That will work with
FreeTDS as well, and give you more functions.



[2003-11-17 17:29:18] cquincy at whitewave dot com

I think I have run into a problem related to this bug fix.  I am using
sybase/freetds to interact with mssql on my company intranet like the
orginator of this bug.  Now with version 4.3.4 numbers, whether integer
of floating points are coming back with spaces on the right side.  i.e.
you put "123" in the database, then when you mssql_fetch* you get back
"123   ".  Note that the originator of the bug was only concerned with
char and varchar fields.  4.3.3 did not have this problem with numbers.
 So, it seems the "fix" has created an unforseen problem with numbers
which is worse in my opinion than "hi   " coming back as "hi".   


Obviously I expect to get exactly the same data that is stored in the
database instead of right padded data.

Expected result:

Put 123 in the database get back 123, not '123 '.  Put 123.456 in the
database get back 123.45, not '123.456 '

Thanks for your efforts.



[2003-10-14 23:32:13] [EMAIL PROTECTED]

This bug has been fixed in CVS.

In case this was a PHP problem, snapshots of the sources are packaged
every three hours; this change will be in the next snapshot. You can
grab the snapshot 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 in short time.
 
Thank you for the report, and for helping us make PHP better.





[2003-10-10 06:12:57] duh at dowebwedo dot com

the configure line is:

'./configure' '--with-mysql' '--with-apxs=/www/bin/apxs'
'--with-gd=/usr/local' '--with-png-dir' '--with-freetype-dir'
'--with-pear' '--with-zlib-dir' '--enable-track-vars'
'--enable-trans-sid' '--disable-posix-threads' '--enable-shared'
'--with-pgsql=/usr' '--with-unixODBC=/usr/local/easysoft/unixODBC'
'--with-mssql=/usr/local'



[2003-10-07 19:09:27] [EMAIL PROTECTED]

What was the configure line used to configure PHP?




The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
http://bugs.php.net/25777

-- 
Edit this bug report at http://bugs.php.net/?id=25777&edit=1

#26289 [Bgs->Opn]: Random problem in mail() function

[julien at touslesdrivers dot com]

 ID:   26289
 User updated by:  julien at touslesdrivers dot com
 Reported By:  julien at touslesdrivers dot com
-Status:   Bogus
+Status:   Open
 Bug Type: *Mail Related
 Operating System: Windows 2000 Server SP4
 PHP Version:  4.3.4
 New Comment:

up


Previous Comments:


[2003-11-17 14:45:46] julien at touslesdrivers dot com

Thanks you fo your reply.

I don't think the problem is the $code beacause this variable is set
and is set correctly. Otherwise the value of $code would not appear in
the URL...

I just want to know where this "VOID" come from?! And why it replaces
the string "script" in the file name?

Thanks



[2003-11-17 14:16:12] [EMAIL PROTECTED]

Sorry, but your problem does not imply a bug in PHP itself.  For a
list of more appropriate places to ask for help using PHP, please
visit http://www.php.net/support.php as this bug system is not the
appropriate forum for asking support questions. 

Thank you for your interest in PHP.

I suspect the $code is not set or not set correctly.

First of all, change http://www.domain.com/inscription.php?code=$code
to
http://www.domain.com/inscription.php?code={$code}
and always validate value of $code before using it.



[2003-11-17 11:10:45] julien at touslesdrivers dot com

Description:

Hello,

I have a problem with the mail() function. The OS is Windows 2000
Server SP4 + IIS 5.0 + MySQL + PHP 4.3.4 (ISAPI).

On my website, there is a form to subscribe to the site as a member. To
confirm the subscription, I send a mail to the person who has to click
on a link to confirm the subscription.

I use so the mail() function to send the mail as text format but there
is a random problem.

Sometimes (about 2-3 times per week) persons who receive the mail can't
confirm their subscription and forward the mail to me and I can then
constat the problem:

A little part of the body (the string "script") had been replaced by
another string ("VOID").

This problem is not dependant with the Mail client (Outlook, Netscape
etc.).

PHP 4.3.4 doesn't resolves this problem.

I am not arrived to reproduce this problem and it it seems to be
totally random. It is a very strange bug!! An idea?

Reproduce code:
---
The code source is:

$message = "
Bonjour,

Pour valider définitivement votre inscription comme membre, merci de
clicker sur le lien ci-dessous:
http://www.domain.com/inscription.php?code=$code
";

mail($email,"Inscription",$message,"From:
[EMAIL PROTECTED]: text/plain; charset=iso-8859-1");

Expected result:

Normally the person would see:

Bonjour,

Pour valider définitivement votre inscription comme membre, merci de
clicker sur le lien ci-dessous:
http://www.domain.com/inscription.php?code=1587

Actual result:
--
But some persons obtain:

Bonjour,

Pour valider définitivement votre inscription comme membre, merci de
clicker sur le lien ci-dessous:
http://www.domain.com/inVOIDion.php?code=1587





-- 
Edit this bug report at http://bugs.php.net/?id=26289&edit=1

#26249 [Bgs->Opn]: snmpget doesn't free memory

[dacol dot jeremy at caramail dot com]

 ID:   26249
 User updated by:  dacol dot jeremy at caramail dot com
 Reported By:  dacol dot jeremy at caramail dot com
-Status:   Bogus
+Status:   Open
 Bug Type: SNMP related
 Operating System: linux
 PHP Version:  4.3.2
 New Comment:

hi,

i don't know how upgrad my php files with the latest version...

i up my old ucd-snmp version to net-snmp 5.1 and my bug still
happened.

please tell me exactly what to do with php4-STABLE-latest.tar.gz

the command, all things tou can

i am really sorry but i search on web and i find nothing on it.


Previous Comments:


[2003-11-17 17:52:10] [EMAIL PROTECTED]

Update your über-old ucd-snmp to the latest net-snmp version.
Then reopen if you still experience leaks which you can trace to happen
in PHP code (not in the net-snmp)




[2003-11-17 05:40:55] dacol dot jeremy at caramail dot com

hi 
i have copied the CVS on my php files and replace it.
but it doesn't work each time i use snmpget function it takes memory
which it doesn't free after a response.
in fact each 100 snmpget use 0.1% of my memory.
it's really the snmpget function which doesn't work.



[2003-11-17 03:06:59] response dot sorry at notremember dot pass

hi,

thank you for your answer.

can i only copy this on my existing php files or do i have to recompile
all?
in fact what i have to do with this file?

and sorry for the mistake i work with:
apache 1.3.28
php 4.3.3
ucd-snmp 4.2.2
on linux RH 7.1



[2003-11-14 10:36:16] [EMAIL PROTECTED]

Please try using this CVS snapshot:

  http://snaps.php.net/php4-STABLE-latest.tar.gz
 
For Windows:
 
  http://snaps.php.net/win32/php4-win32-STABLE-latest.zip





[2003-11-14 04:58:57] dacol dot jeremy at caramail dot com

Description:

hello,

I use a lot of snmpget (more then 300) each 5 minutes.
snmpget works, it closes session. ( i read all bug reports on snmpget)
but it use memory which it doesn't free.

i look, with the "top" command, the ressources used by my script and
every 5 minutes it uses 0.3% more...

i don't understand why
maybe a bug in my snmp.c but.

i use ucd-snmp package with php 4.3.2 and i use my script with php
command.






-- 
Edit this bug report at http://bugs.php.net/?id=26249&edit=1

#26298 [NEW]: Intarbase fetch record of wide field freeze PHP execution

[tma at 2p dot cz]

From: tma at 2p dot cz
Operating system: Win98
PHP version:  4.3.4
PHP Bug Type: Reproducible crash
Bug description:  Intarbase fetch record of wide field freeze PHP execution 

Description:

It seems there is some problem when fetching interbase records containing
a wide field.

Execution of PHP code is frozen at random point - probably some memory
problem. Point of freezing depends on number of fetched records. 

Does not matter if ibase_fetch_row/assoc/object is used. There is one
presumtion - must be selected a field that is type of VARCHAR(200). I did
not test what is critical field length. There are in field national chars
but this occurs even pure english chars are used (#32..#127)(first I've
used UNICODE_FSS and thought that problem is in unicode). If this wide
column is not listed it works (SELECT ID,ID as NAME). 

Does not matter if NAME's CHARACTER SET is ASCII or UNICODE_FSS

CREATE TABLE TBL (
   ID VARCHAR(10) CHARACTER SET ASCII NOT NULL,
   NAME VARCHAR(200) CHARACTER SET ASCII /* UNICODE_FSS*/ NOT NULL,
   PARENT_ID VARCHAR(10) CHARACTER SET ASCII,
 CONSTRAINT GROUPS_PK PRIMARY KEY (ID)
);

Note:
Problem is neither in IBServer nor in IBClient. When using windows based
apps using gds32.dll it works.

I cannot fix it by myself because I have not MS-C to compile module for
Windows and I'm not expert for CYGWIN environment.

Reproduce code:
---
function CleanUp() {
global $IB_HANDLE;
  if ($IB_HANDLE) {
echo "close\n";
IBase_Close($IB_HANDLE);
echo "close done\n";
  }
}
Register_ShutDown_Function('CleanUp');

$maxn = 1000;  // limit max num. of listed records

Dl('php_interbase.dll');
$IB_HANDLE = IBase_Connect (IB_HOST, IB_USER, IB_PASSWORD);

$query = IBase_Query($IB_HANDLE, "SELECT * FROM TBL");
$n = 0;  while ($res = IBase_Fetch_Assoc($query)) {
  echo "$n.";
  printf("%s: %s\n", $res['ID'], $res['NAME']); 
  Flush();  // no effect
  Unset($res);
  $n++;
  if ($n > $maxn) {
break;
  }
}
echo "loop done\n";
IBase_Free_Result($query);
echo "free query done\n";


Expected result:

-open ibase
-open SQL
-list table records.
-close SQL
-close ibase

Actual result:
--
Execution of PHP code is frozen at random point - probably some memory
problem. Point of freeze depends on number of fetched records. For example
in my environment if $maxn < 6 it works, if $maxn < 100 it's frozen in
when ibase_free_result in action - if this line is commented is frozen
during ibase_close. If $maxn = 1000 is frozen when listing record 355. But
this depends on output and ASCII/UNICODE_FSS.

Notes:

Execution is frozen because time_limit has no effect.

If was script executed from Apache as CGI and execution was stopped from
browser there was still connection to IBServer. If was executed from
command line and pressed Ctrl+C when frozen, connection was closed

It's not time_limit or output problem, this occurs only if ibase_fetch_*
called.

-- 
Edit bug report at http://bugs.php.net/?id=26298&edit=1
-- 
Try a CVS snapshot (php4):  http://bugs.php.net/fix.php?id=26298&r=trysnapshot4
Try a CVS snapshot (php5):  http://bugs.php.net/fix.php?id=26298&r=trysnapshot5
Fixed in CVS:   http://bugs.php.net/fix.php?id=26298&r=fixedcvs
Fixed in release:   http://bugs.php.net/fix.php?id=26298&r=alreadyfixed
Need backtrace: http://bugs.php.net/fix.php?id=26298&r=needtrace
Try newer version:  http://bugs.php.net/fix.php?id=26298&r=oldversion
Not developer issue:http://bugs.php.net/fix.php?id=26298&r=support
Expected behavior:  http://bugs.php.net/fix.php?id=26298&r=notwrong
Not enough info:http://bugs.php.net/fix.php?id=26298&r=notenoughinfo
Submitted twice:http://bugs.php.net/fix.php?id=26298&r=submittedtwice
register_globals:   http://bugs.php.net/fix.php?id=26298&r=globals
PHP 3 support discontinued: http://bugs.php.net/fix.php?id=26298&r=php3
Daylight Savings:   http://bugs.php.net/fix.php?id=26298&r=dst
IIS Stability:  http://bugs.php.net/fix.php?id=26298&r=isapi
Install GNU Sed:http://bugs.php.net/fix.php?id=26298&r=gnused
Floating point limitations: http://bugs.php.net/fix.php?id=26298&r=float

#26221 [Bgs->Opn]: Configure with imap-support broken

[admin at sportsandbytes dot de]

 ID:   26221
 User updated by:  admin at sportsandbytes dot de
 Reported By:  admin at sportsandbytes dot de
-Status:   Bogus
+Status:   Open
 Bug Type: IMAP related
 Operating System: Linux, RedHat 7.3
 PHP Version:  4.3.4
 New Comment:

from configure --help 
--with-imap-ssl= IMAP: Include SSL support. DIR is the 
OpenSSL install dir. 
 
so either the configure option is wrong or 
--with-imap-ssl=/usr/local/openssl/current  
should work 
 
We specifically DONT want it to link against the systems openssl in 
/usr/lib, which is what probably would happen when not giving a 
openssl dir.


Previous Comments:


[2003-11-12 11:00:52] [EMAIL PROTECTED]

Sorry, but your problem does not imply a bug in PHP itself.  For a
list of more appropriate places to ask for help using PHP, please
visit http://www.php.net/support.php as this bug system is not the
appropriate forum for asking support questions. 

Thank you for your interest in PHP.

Use --with-imap-ssl not 
--with-imap-ssl=/usr/local/openssl/current/



[2003-11-12 10:29:02] admin at sportsandbytes dot de

Description:

Hi,

the following Configure-Options work with php-4.3.3 but not with
php-4.3.4:
./configure \
--with-apxs=/usr/local/apache/current/bin/apxs \
--with-zlib \
--with-jpeg-dir=/usr/lib \
--with-png-dir=/usr/lib \
--enable-ftp \
--with-gd \
--enable-versioning \
--enable-track-vars=yes \
--enable-url-includes \
--enable-sysvshm=yes \
--enable-sysvsem=yes \
--with-config-file-path=/etc \
--with-freetype-dir=/usr/lib \
--enable-fastcgi \
--enable-debug \
--disable-ipv6 \
--with-openssl=/usr/local/openssl/current/ \
--with-imap \
--with-imap-ssl=/usr/local/openssl/current/ \
--with-kerberos

The Configure-Script ends with this:

...
...
...
checking whether to enable dbx support... no
checking whether to enable direct I/O support... no
checking for DOM support... no
checking for DOM XSLT support... no
checking for DOM EXSLT support... no
checking whether to enable EXIF (metadata from images) support... no
checking for FrontBase SQL92 (fbsql) support... no
checking for FDF support... no
checking whether to enable the bundled filePro support... no
checking for FriBidi support... no
checking whether to enable FTP support... yes
checking for GD support... yes
checking for the location of libjpeg... /usr/lib
checking for the location of libpng... /usr/lib
checking for the location of libXpm... no
checking for FreeType 1.x support... no
checking for FreeType 2... /usr/lib
checking for T1lib support... no
checking whether to enable truetype string function in GD... no
checking whether to enable JIS-mapped Japanese font support in GD...
no
checking for fabsf... yes
checking for floorf... yes
checking for jpeg_read_header in -ljpeg... yes
checking for png_write_image in -lpng... yes
If configure fails try --with-xpm-dir=
checking for GNU gettext support... no
checking for GNU MP support... no
checking for hwapi support... no
checking for Hyperwave support... no
checking for iconv support... no
checking for IMAP support... yes
checking for pam_start in -lpam... yes
checking for crypt in -lcrypt... yes
configure: error: Cannot find imap library (libc-client.a). Please
check your IMAP installation.


The file is located here:
[EMAIL PROTECTED] src]# locate libc-client.a
/usr/lib/libc-client.a
[EMAIL PROTECTED] src]# cd /usr/lib
[EMAIL PROTECTED] lib]# ls -al | grep c-client
-rw-r--r--1 root root  2404414 Apr 18  2002 c-client.a
lrwxrwxrwx1 root root   10 Jun  3 14:29 libc-client.a
-> c-client.a
[EMAIL PROTECTED] lib]# 

Without the imap-configure-options, configure and make work without a
failure!

The other Software:

apache_1.3.29 at /usr/local/apache/current
openssl-0.9.7c at /usr/local/openssl/current
courier-imap-2.1.2 at /usr/local/courier-imap/current









-- 
Edit this bug report at http://bugs.php.net/?id=26221&edit=1

#26295 [Opn->Bgs]: Change from PCRE Library 3.9 to PCRE library 4.3 has resulted in broken code.

[sniper]

 ID:   26295
 Updated by:   [EMAIL PROTECTED]
 Reported By:  mnbob70 at earthlink dot net
-Status:   Open
+Status:   Bogus
 Bug Type: PCRE related
 Operating System: Linux lion.phpwebhosting.com 2.4
 PHP Version:  4.3.4
 New Comment:

You can always compile PHP using an external PCRE library.
Anyway, this is not PHP bug. Bundled pcre library was upgraded since it
fixed several other bugs. 



Previous Comments:


[2003-11-18 00:55:58] mnbob70 at earthlink dot net

Description:

A friend running PHP 4.3.4 is experiencing an error with code that I
wrote that works fine with PHP 4.3.2. The code parses stat information
from a Madden 2004 game log.

I'm guessing that something has changed in moving from PCRE Library 3.9
to 4.3.

Here is a sample of the text file being parsed.

Madden NFL 2004 PC v4.1.0.0 Game Log - Topdogs at Muleys
Game Time: Sun Oct 12 13:31:48 2003
Skill Level: Pro
Quarter Length: 9 minute(s)

Slider Human   CPU
QB Accuracy   4525
Pass Blocking 6540
Receiver Catching 4840
Running Ability   5250
Offensive line Run Blocking   5555
Defensive Awareness   6550
Defensive line Knockdowns 5050
Interceptions 2020
Defensive Break Blocks5550
Tackling  5550
Fieldgoal Length  6050
Fieldgoal Accuracy5040
Punt Length   7580
Punt Accuracy 7050
Kickoff Length5550

Weather
Forecast: Fair
Wind: 1 mph
Temp: 34 degrees

Team   Q1   Q2   Q3   Q4   OT  FINAL
Topdogs 0   14   2130 38
Muleys 14   14070 35

Game Statistics:  Topdogs Muleys
First Downs18 11
Third Down Conversions  5  6
Third Downs14 11
Third Down Percentage   35.7%  54.5%
Forth Down Conversions  0  0
Fourth Downs0  1
Fourth Down Percentage   0.0%   0.0%
Two Pt Conversions Made 0  0
Two Pt Conversions Attempted0  0
Two Pt Conversion Percentage 0.0%   0.0%
Offense Redzone Num 3  6
Offense Redzone TDs 3  4
Offense Redzone FGs 0  0
Offense Redzone Percentage 100.0%  66.7%
Rushing Attempts   35 15
Rushing Yards 147 40
Rushing Average   4.22.7
Rushing TDs 1  3
Passing Yards 219295
Passing Completions16 19
Passing Attempts   30 38
Completion Percentage   53.3%  50.0%
Passing Average   7.37.8
Passing TDs 3  2
Offensive Pass Interceptions3  4
TOTAL OFFENSE 366335
Defensive Pass Interceptions4  3
Punt Return Yards  15 75
Kick Return Yards 106131
Punts   5  4
Punt Avg 43.0   43.3
Fumbles 1  2
Fumbles Lost0  1
Penalties   6  2
Penalty Yds42 15
Turnovers   3  5
TIME OF POSSESSION  22:31  13:31

Individual Stats: Muleys
PASSINGCMP  ATT  YDS  PCT   YPA SACK   TD  INT
LONG RATING
Lyle Coen   19   38  310   50   8.1324 
 80   55.7

RUSHINGATT  YDS   AVG LONG   TD  FUM
Derrick Walton  10   26   2.5   1021
Paul Johnson 4   15   3.7810
Lyle Coen1   -1  -1.0001

RECEIVING  REC  YDS   AVG LONG   TD DROP  YAC
Wade Van Dyke3  117  39.0   8011   55
Derrick Walton   2   76  38.0   5700   45
Eugene Nixon 7   64   9.1   2210   11
Victor Fitzgerald4   29   7.2   16013
Jimmy Harper 1   14  14.0   14000
Willie Linde 16   6.06001
Seth Henry   14   4.04000

KICKINGFGM  FGA  PCT FGSBLOCKED  XPA  

#26296 [Opn->Bgs]: Warning: mssql_query(): message: Invalid object name

[sniper]

 ID:   26296
 Updated by:   [EMAIL PROTECTED]
 Reported By:  ameen at whitehouseit dot com
-Status:   Open
+Status:   Bogus
 Bug Type: MSSQL related
 Operating System: Windows 2000 Server
 PHP Version:  4.3.4
 New Comment:

Sorry, but your problem does not imply a bug in PHP itself.  For a
list of more appropriate places to ask for help using PHP, please
visit http://www.php.net/support.php as this bug system is not the
appropriate forum for asking support questions. 

Thank you for your interest in PHP.




Previous Comments:


[2003-11-18 01:48:03] ameen at whitehouseit dot com

Description:

Just i try to connect my database in SQL Server and fetch some records
from the database. But it shows warning message like this.

Warning: mssql_query(): message: Invalid object name 'mytable'.
(severity 16) in D:\abc\Conn.php on line 21

Warning: mssql_query(): Query failed in D:\abc\Conn.php on line 21

How to rectify? i'm sure tables exists in my db. and also connection
establishes perfectly.

Reproduce code:
---
$i=mssql_connect("localhost","uid","pwd")  or die("unable to
Connect");
if (mssql_select_db("mydb",$i)==true)
echo "DB Exists";
else
echo "DB Not Exists";

$sql="select * from mytable where login_id='myname'";

$rs=mssql_query($sql,$i);

Expected result:

Warning: mssql_query(): message: Invalid object name 'mytable'.
(severity 16) in D:\abc\Conn.php on line 21

Warning: mssql_query(): Query failed in D:\abc\Conn.php on line 21


Actual result:
--
Fetch Records from db





-- 
Edit this bug report at http://bugs.php.net/?id=26296&edit=1

#25863 [Opn->Fbk]: The specified CGI application misbehaved ...

[sniper]

 ID:   25863
 Updated by:   [EMAIL PROTECTED]
 Reported By:  salmanarshad2000 at yahoo dot com
-Status:   Open
+Status:   Feedback
 Bug Type: IIS related
 Operating System: Windows XP, 2000
 PHP Version:  4.3.3
 New Comment:

Check your cgi.* php.ini settings. Tuning those (set to opposite what
they are set right now) might help. 
And for the record: I can not reproduce this.



Previous Comments:


[2003-11-18 00:38:09] salmanarshad2000 at yahoo dot com

The problem disappears totally if we use Apache, but some people just
don't like running two webservers on a machine at a time.



[2003-11-17 18:39:03] [EMAIL PROTECTED]

Try with Apache 1.3.29.





[2003-10-14 08:44:21] salmanarshad2000 at yahoo dot com

Description:

This bug or issue has been around for quite a while and seems like
nobody cares. The bug list is filled with hundreds of complains about
the "The specified CGI application misbehaved ..." error each time
these people have BOGUSed or CLOSEd saying things like "The version you
are using is too old, please try the latest version ...", "This is not
a php bug, please go to ...", "Not enough evidence ..." or "Problem
with Windows, not PHP". Quite a few versions of php have been released
in the meanwhile, but this issue hasn't been fixed, people who upgrade
their php installation come back with the same complains. I see no good
reason for this ignorance.

Problem Statement
-
When browsing a php application, the IIS server randomly throws the
error message:

CGI Error
The specified CGI application misbehaved by not returning a complete
set of HTTP headers. The headers it did return are:


Observations

This happened only when:
- PHP.exe is used as a CGI on IIS 
- The php scripts contained 2 or more frames (e.g. phpMyAdmin)
- MySQL operation was executed (update, insert, delete etc.)
- header("Location: ...") is used to redirect user after a MySQL
operation to a page that also performs a MySQL operation
- The pages are viewed from local computer
- A very fast computer is used

This did not happened when:
- Apache server for windows with php support was used
- The php scripts contained 2 or more frames but all frames contained
php scripts with Hello World and a random number
- Frequency of errors was much lesser when same pages were accessed
from the network
- Pentium 2, 300 MHz was used (a slow computer)

History
---
Following bugs are all related to this problem. This is just a reminder
for the fact that this issue has been discussed quite a few times and
it is still present. People also found these interesting things that
might help to get the problem solved.

- BugID #25567 getting errors when doing a mysql_db_query() and then
header("location")
- BugID #24916 header("location")
- BugID #23208 using two or more frames
- BugID #19381 no details :(
- BugID #19676 works on one (slow) system but does not work on other
- BugID #18901 header("location") after delete or update, error occurs
under under load
- BugID #16313 header("location") and db operations
- BugID #23050 mysql_query() followed by header("location")
- BugID #17468 header("location")
- BugID #9852  thousands of lines describing the problem, including
frames, manually slowing down the script, pages work from outside the
machine nut not locally, two database connections in rapid succession
etc

Things that have been said earlier
--
"This is a problem with Microsoft OS"
No this is not, it works on exact same OS running on slower hardware or
when the application is accessed across a network. And even if it is,
the developers should try to find a work around instead of blaming M$
and telling it to fix it. After all, when you develop some app for an
environment, you don't change the environment to suit your app
(although sometimes it is easier to do so).

"This is not a php bug"
Well this could be right, since there is one other suspect, MySQL. But
somebody please figure out where the problem is? Also, MySQL is now an
integral part of php so problem could lie in the integration part.

My Opinion
---
May be php.exe is not fast enough to keep up with the pace at which IIS
can throw requests at it. Or ... may be it is a problem with the MySQL
connections ... attempting to create connections too quickly may be the
cause. Users having same problem please feel free to contribute with
their observations and suggestions.






-- 
Edit this bug report at http://bugs.php.net/?id=25863&edit=1

#26130 [Opn->Fbk]: IBM DB2 Unique Key Problem

[sniper]

 ID:   26130
 Updated by:   [EMAIL PROTECTED]
 Reported By:  jay at nicwr dot mah dot nic dot in
-Status:   Open
+Status:   Feedback
 Bug Type: ODBC related
 Operating System: Linux 8.0
 PHP Version:  4.3.2
 New Comment:

Please provide the requested feedback (see the code in my previous
comment..)



Previous Comments:


[2003-11-05 00:51:58] [EMAIL PROTECTED]

Please upgrade first to PHP 4.3.4.
And then try this script and paste the output here:






[2003-11-05 00:43:17] jay at nicwr dot mah dot nic dot in

Description:

I have IBM DB2 V 7.2 EE Fix Pack 7
My Php is enabled with ibm-db2 support

I have a table test (c1 int not null,c2 int,c3 int)
I have a unique key on table test as (c1) with include options for
(c2,C3)
If I have a sample data
1,null,null
2,1,null,
3,null,1
4,1,2
If I access the data thru my php script I do not get desirable result.


Reproduce code:
---
";
}
?>


Expected result:

Expected Result is as follows
-
Record:1---End Record
Record:2-1--End Record
Record:3--1-End Record
Record:4-1-2-End Record



Actual result:
--
Actual Results Appear as follows

Record : ---End Record
Record : ---End Record
Record : ---End Record
Record : 4-1-2-End Record





-- 
Edit this bug report at http://bugs.php.net/?id=26130&edit=1

#9876 [Com]: OCIPLogon flaky after restart Oracle

[aferaud at buyingpack dot com]

 ID:   9876
 Comment by:   aferaud at buyingpack dot com
 Reported By:  merrill at missioncriticallinux dot com
 Status:   Closed
 Bug Type: OCI8 related
 Operating System: Redhat 6.1 Linux
 PHP Version:  4.0.4pl1
 New Comment:

PHP 4.3.3 (Linux) - Oracle 8.1.7 (Win2000)
I've have the same problem, and it still persists with non-persistent
connection...

Is there a solution somewhere ??


Previous Comments:


[2002-04-08 22:53:07] bmironov at jonview dot com

Hello,

I have same problem under
RedHat 6.1 / 7.2
Oracle 7.3.4 / 8.1.5
PHP 4.1.2  (--with-oci8)
Apache 1.3.23

I have to restart whole webserver after Database restart. I would like
to use OCIPLogon because of performance issue for very small SELECT's.



[2001-05-04 10:35:58] [EMAIL PROTECTED]

plz try 4.0.5 - if the problem persists please reopen.



[2001-03-20 11:55:31] merrill at missioncriticallinux dot com

./configure' '--prefix=/var/php' '--with-config-file-path=/var/php'
'--enable-track-vars'
  '--enable-sigchild'
'--with-apache=../apache_1.3.14' '--with-ftp' '--with-xml'
 
'--with-mcrypt=/var/libmcrypt' '--with-pgsql=/var/postgresql'
 
'--with-oci8=/u01/app/oracle/product/8.1.6' '--with-mysql=/var/mysql'

When using OCIPLogon for persistent connections, I'm noticing that
after restarting Oracle, that my PHP application starts getting Oracle
errors(ORA-03113 and ORA-03114).  This can be fixed by restarting the
webserver after restarting Oracle.  I also notice that changing my
Oracle connections from OCIPLogon to OCILogon fixes the problem - after
making that change I get no errors after restarting Oracle, and I
*don't* have to restart the webserver.

The OCI8 interface must have a bug, right?  I shouldn't have to restart
the webserver after restarting Oracle just because I'm using persistent
connections, should I?




-- 
Edit this bug report at http://bugs.php.net/?id=9876&edit=1

#20473 [Com]: Cannot redeclare gc()

[frodelan at netcom dot no]

 ID:   20473
 Comment by:   frodelan at netcom dot no
 Reported By:  hofmann at isl dot org
 Status:   Closed
 Bug Type: Unknown/Other Function
 Operating System: Linux
 PHP Version:  4.3.0RC1
 New Comment:

I have the same problem still, because of this I have to run php 4.2.2
and this made me sad! I've tried to run php 4.3.4 and I get the "Cannot
redeclare gc() in session.inc"-problem.
Is there anyone that will fix this problem?


Previous Comments:


[2003-04-08 20:31:31] dave at toad dot net

The problem does not appear to be that php defines a gc() function that
phplib is clobbering, but that phplib's session.inc file actually tries
to define gc() twice.

I was able to eliminate the error message and problems by commenting
out the first gc() declaration in session.inc, as follows:

// Garbage collection
//
// Destroy all session data older than this
// COMMENTED THIS FUNCTION OUT DCT 4/8/03
/* function gc() {
$this->that->ac_gc($this->gc_time, $this->name);
} */

Would be interested in anyone's opinion as to whether this was terribly
foolish, or an accurate fix to the problem.



[2003-02-10 16:29:42] edwin-php dot net at mavetju dot org

I found it to be an old line in my php.ini:

auto_prepend = /usr/local/apache/php/prepend.php3

Update your php.ini and you'll see it goes away again.

Edwin, bitten by this problem too :-)



[2003-02-04 15:30:53] skerdog at hotmail dot com

SURELY someone knows how to fix this annoying error?  It was first
reported over 2 months ago and there are no postings on this problem
report showing a resolution.  Well?



[2003-01-02 07:42:13] phpbugs at dewet dot org

This one also just hit me upgrading from 4.2.3 to 4.3.0 on 
Gentoo Linux.  Since I cannot seem to find a solution 
online, I will try and rename the PHPLIB functions.



[2002-12-06 21:56:44] craigm at ace dot com dot au

Has anybody got a solution for this problem. The same thing is
happening to some of our websites. Is there some setting in php.ini
file that can be set to get around this problem.

Thanks,

Craig Marchant



The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
http://bugs.php.net/20473

-- 
Edit this bug report at http://bugs.php.net/?id=20473&edit=1

#26297 [Opn->Bgs]: Function redeclare

[daniel at cyberprune dot com]

 ID:   26297
 User updated by:  daniel at cyberprune dot com
 Reported By:  daniel at cyberprune dot com
-Status:   Open
+Status:   Bogus
 Bug Type: Scripting Engine problem
 Operating System: Linux
 PHP Version:  4.3.4
 New Comment:

Found problem


Previous Comments:


[2003-11-18 04:52:07] daniel at cyberprune dot com

Description:

Fatal error: Cannot redeclare get_total_topics() (previously declared
in
/vhost/rollo/rollo.cyberprune.com/portal/exoops/modules/newbb_plus/functions.php:2)
in
/vhost/rollo/rollo.cyberprune.com/portal/exoops/modules/newbb_plus/functions.php
on line 2


Reproduce code:
---
query($sql)) {
return(_ERROR);
}

if (!$myrow = $db->fetch_array($result)) {
return(_ERROR);
}

return($myrow['total']); }

Expected result:

Function to work






-- 
Edit this bug report at http://bugs.php.net/?id=26297&edit=1

#26297 [NEW]: Function redeclare

[daniel at cyberprune dot com]

From: daniel at cyberprune dot com
Operating system: Linux
PHP version:  4.3.4
PHP Bug Type: Scripting Engine problem
Bug description:  Function redeclare

Description:

Fatal error: Cannot redeclare get_total_topics() (previously declared in
/vhost/rollo/rollo.cyberprune.com/portal/exoops/modules/newbb_plus/functions.php:2)
in
/vhost/rollo/rollo.cyberprune.com/portal/exoops/modules/newbb_plus/functions.php
on line 2


Reproduce code:
---
query($sql)) {
return(_ERROR);
}

if (!$myrow = $db->fetch_array($result)) {
return(_ERROR);
}

return($myrow['total']); }

Expected result:

Function to work


-- 
Edit bug report at http://bugs.php.net/?id=26297&edit=1
-- 
Try a CVS snapshot (php4):  http://bugs.php.net/fix.php?id=26297&r=trysnapshot4
Try a CVS snapshot (php5):  http://bugs.php.net/fix.php?id=26297&r=trysnapshot5
Fixed in CVS:   http://bugs.php.net/fix.php?id=26297&r=fixedcvs
Fixed in release:   http://bugs.php.net/fix.php?id=26297&r=alreadyfixed
Need backtrace: http://bugs.php.net/fix.php?id=26297&r=needtrace
Try newer version:  http://bugs.php.net/fix.php?id=26297&r=oldversion
Not developer issue:http://bugs.php.net/fix.php?id=26297&r=support
Expected behavior:  http://bugs.php.net/fix.php?id=26297&r=notwrong
Not enough info:http://bugs.php.net/fix.php?id=26297&r=notenoughinfo
Submitted twice:http://bugs.php.net/fix.php?id=26297&r=submittedtwice
register_globals:   http://bugs.php.net/fix.php?id=26297&r=globals
PHP 3 support discontinued: http://bugs.php.net/fix.php?id=26297&r=php3
Daylight Savings:   http://bugs.php.net/fix.php?id=26297&r=dst
IIS Stability:  http://bugs.php.net/fix.php?id=26297&r=isapi
Install GNU Sed:http://bugs.php.net/fix.php?id=26297&r=gnused
Floating point limitations: http://bugs.php.net/fix.php?id=26297&r=float

#25623 [Fbk->Opn]: Memory leak in function load()

[maoy at fjii dot com]

 ID:   25623
 User updated by:  maoy at fjii dot com
 Reported By:  maoy at fjii dot com
-Status:   Feedback
+Status:   Open
 Bug Type: OCI8 related
 Operating System: Windows XP
 PHP Version:  4.3.3
 New Comment:

when I replaced ORACLE 8.1.7 against ORACLE 9.2, the situation
disappeared.


Previous Comments:


[2003-11-17 21:50:41] [EMAIL PROTECTED]

And do answer Thies' question: Does _PHP_ report any leaks when
compiled with --enable-debug (on linux)




[2003-11-17 21:49:49] [EMAIL PROTECTED]

You can find debug enabled win32 builds here:

   http://www.thebrainroom.net/php-snaps




[2003-10-23 23:03:52] maoy at fjii dot com

I had a try to run the following code on windows and linux, it cause
memory leak on both platform. The size of memory leak depends on the
size of CLOB-field's value and record count.
load(); 
  ecco $result;
}
?>



[2003-10-09 20:08:47] maoy at fjii dot com

I'm not able to compile a debug-build, but it's undoubted that the
function cause memory leak. Please help me as soon as possible.

I am looking forward to reply. Thank you very much.



[2003-09-24 05:41:54] [EMAIL PROTECTED]

sorry, if the leak does not show on unix there's not much 
i can do as i don't run windows. 
 
can you compile a debug-build and see if php actually 
complains about mem-leaks? 
 



The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
http://bugs.php.net/25623

-- 
Edit this bug report at http://bugs.php.net/?id=25623&edit=1

#26254 [Fbk->Opn]: Inaccuracy in imagecopyresized

[rc at opelgt dot org]

 ID:   26254
 User updated by:  rc at opelgt dot org
 Reported By:  rc at opelgt dot org
-Status:   Feedback
+Status:   Open
 Bug Type: GD related
 Operating System: Mac OS X 10.2.8
 PHP Version:  4.3.4
 New Comment:

I used the downloadable version of www.entropy.ch with 
an installer. So I didnt change something of that. Here 
is what phpinfo tells about the configuration command:

'./configure' '--prefix=/usr/local/php' '--with-config-
file-path=/usr/local/php/lib' '--with-apxs' '--with-
iconv=/usr/local/php' '--with-openssl=/usr' '--with-
zlib=/usr' '--with-mysql' '--with-pgsql=/Users/marc/
cvs/entropy/php-module/build/postgresql-build' '--
enable-cgi' '--with-gd' '--with-png-dir=/usr/local/php' 
'--with-freetype-dir=/usr/local/php' '--with-t1lib=/
usr/local/php' '--with-jpeg-dir=/usr/local/php' '--
with-tiff-dir=/usr/local/php' '--with-curl=/usr/local/
php' '--with-mcal=/usr/local/php' '--with-mcrypt=/usr/
local/php' '--with-mhash=/usr/local/php' '--with-
pdflib=/usr/local/php' '--with-imap=../imap-2002d' '--
with-imap-ssl=/usr' '--with-expat-dir=/usr/local/php' 
'--with-gettext=/usr/local/php' '--with-dom=/usr/local/
php' '--with-dom-xslt=/usr/local/php' '--with-dom-
exslt=/usr/local/php' '--with-xslt-sablot=/usr/local/
php' '--with-mssql=/usr/local/php' '--enable-xslt' '--
with-ldap' '--with-xmlrpc' '--with-xml' '--with-iodbc' 
'--enable-sockets' '--enable-dbx' '--enable-dbase' '--
enable-trans-sid' '--enable-exif' '--enable-wddx' '--
enable-ftp' '--enable-mbstring'


Previous Comments:


[2003-11-17 22:48:20] [EMAIL PROTECTED]

What was the configure line you used? 




[2003-11-17 08:22:47] rc at opelgt dot org

When the destination image is created with 
imagecreatetruecolor the behaviour seems to be ok.
So the bug is obviously for a palette based image.



[2003-11-17 07:48:31] rc at opelgt dot org

The bug is still existing in 4.3.4 R1.
So in PHP5 it is solved?



[2003-11-16 13:13:10] [EMAIL PROTECTED]

Thank you for your bug report. This issue has already been fixed
in the latest released version of PHP, which you can download at 
http://www.php.net/downloads.php





[2003-11-16 11:43:37] rc at opelgt dot org

The upper left corner was placed correctly but the 
height was one pixel too small.
I have updated to 4.3.4 R1 and it seems to be ok now.



The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
http://bugs.php.net/26254

-- 
Edit this bug report at http://bugs.php.net/?id=26254&edit=1

#26257 [Csd]: Iterators leak memory

[helly]

 ID:   26257
 Updated by:   [EMAIL PROTECTED]
 Reported By:  adam at trachtenberg dot com
 Status:   Closed
 Bug Type: Zend Engine 2 problem
 Operating System: *
 PHP Version:  5CVS-2003-11-14 (dev)
 Assigned To:  helly
 New Comment:

Actually the problem had nothing to do with iterators. The real problem
was list() assigning to properties.


Previous Comments:


[2003-11-18 02:20:24] [EMAIL PROTECTED]

This bug has been fixed in CVS.

In case this was a PHP problem, snapshots of the sources are packaged
every three hours; this change will be in the next snapshot. You can
grab the snapshot 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 in short time.
 
Thank you for the report, and for helping us make PHP better.





[2003-11-14 18:49:45] adam at trachtenberg dot com

Description:

Iterators leak memory when assigning the results of 
each() to object properties using a list(). When the 
results are stored in local variables and then reassign, 
no leak occurs.

Reproduce code:
---
class ai implements Iterator {

private $array;

function __construct() {
$this->array = array('foo', 'bar', 'baz');
}

function rewind() {
// empty
}

function hasMore() {
if (list($this->key, $this->current) = each($this->array)) {
#   if (list($key, $current) = each($this->array)) {
#   $this->key = $key;
#   $this->current = $current;
return true;
} else {
return false;
}
}

function key() {
return 'key';
}

function current() {
return 'current';
}

function next() {
// empty
}
}

class a implements IteratorAggregate {

public function getIterator() {
return new ai();
}
}

$array = new a();
foreach ($array as $property => $value) {
print "$property: $value\n";
}

Expected result:

No memory leaks. Activate the commented code to see a 
working example.

Actual result:
--
/usr/local/cvs/php/php-src/Zend/zend_API.c(1011) :  
Freeing 0x00BC8AF8 (16 bytes), script=./
iterator_leak.php
Last leak repeated 2 times
/usr/local/cvs/php/php-src/Zend/zend_execute.c(3241) :  
Freeing 0x00BC8300 (4 bytes), script=./iterator_leak.php
/usr/local/cvs/php/php-src/Zend/zend_variables.c(137) : 
Actual location (location was relayed)
Last leak repeated 2 times
/usr/local/cvs/php/php-src/Zend/zend_execute.c(3238) :  
Freeing 0x00BC82B0 (16 bytes), script=./
iterator_leak.php
Last leak repeated 2 times
=== Total 9 memory leaks detected ===





-- 
Edit this bug report at http://bugs.php.net/?id=26257&edit=1

#26257 [Opn->Csd]: Iterators leak memory

[helly]

 ID:   26257
 Updated by:   [EMAIL PROTECTED]
 Reported By:  adam at trachtenberg dot com
-Status:   Open
+Status:   Closed
 Bug Type: Zend Engine 2 problem
 Operating System: *
 PHP Version:  5CVS-2003-11-14 (dev)
-Assigned To:  
+Assigned To:  helly
 New Comment:

This bug has been fixed in CVS.

In case this was a PHP problem, snapshots of the sources are packaged
every three hours; this change will be in the next snapshot. You can
grab the snapshot 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 in short time.
 
Thank you for the report, and for helping us make PHP better.




Previous Comments:


[2003-11-14 18:49:45] adam at trachtenberg dot com

Description:

Iterators leak memory when assigning the results of 
each() to object properties using a list(). When the 
results are stored in local variables and then reassign, 
no leak occurs.

Reproduce code:
---
class ai implements Iterator {

private $array;

function __construct() {
$this->array = array('foo', 'bar', 'baz');
}

function rewind() {
// empty
}

function hasMore() {
if (list($this->key, $this->current) = each($this->array)) {
#   if (list($key, $current) = each($this->array)) {
#   $this->key = $key;
#   $this->current = $current;
return true;
} else {
return false;
}
}

function key() {
return 'key';
}

function current() {
return 'current';
}

function next() {
// empty
}
}

class a implements IteratorAggregate {

public function getIterator() {
return new ai();
}
}

$array = new a();
foreach ($array as $property => $value) {
print "$property: $value\n";
}

Expected result:

No memory leaks. Activate the commented code to see a 
working example.

Actual result:
--
/usr/local/cvs/php/php-src/Zend/zend_API.c(1011) :  
Freeing 0x00BC8AF8 (16 bytes), script=./
iterator_leak.php
Last leak repeated 2 times
/usr/local/cvs/php/php-src/Zend/zend_execute.c(3241) :  
Freeing 0x00BC8300 (4 bytes), script=./iterator_leak.php
/usr/local/cvs/php/php-src/Zend/zend_variables.c(137) : 
Actual location (location was relayed)
Last leak repeated 2 times
/usr/local/cvs/php/php-src/Zend/zend_execute.c(3238) :  
Freeing 0x00BC82B0 (16 bytes), script=./
iterator_leak.php
Last leak repeated 2 times
=== Total 9 memory leaks detected ===





-- 
Edit this bug report at http://bugs.php.net/?id=26257&edit=1

#26263 [Opn]: Apache doesn't start up wiht LoadModule phpX_module .. in config

[peoyli at bredband dot net]

 ID:   26263
 User updated by:  peoyli at bredband dot net
 Reported By:  peoyli at bredband dot net
 Status:   Open
 Bug Type: Apache2 related
 Operating System: OpenBSD 3.4
 PHP Version:  Irrelevant
 New Comment:

PHP5 + MySQL 4.1 core dumps when calling mysql related functions
(mysql_connect ...). I will try the internal support for mysql 4.0.16
or older.


Previous Comments:


[2003-11-16 15:52:47] peoyli at bredband dot net

One other note:

I DID get it running using the internal mysql support (no path given to
--with-mysql) after some hassle. I also now have PHP 5 running together
with mysql 4.1.0 alpha.

The problem seems to have something to do with mysql 4.0.16 (haven't
tried earlier versions), although, as I mentioned before, there is not
problem with mysql 4.0.16 by itself.



[2003-11-16 15:48:40] peoyli at bredband dot net

no output using 'apachectl configtest'. Seems like httpd crasches as
soon as the LoadModule line is appended to the configuration.



[2003-11-16 13:15:23] [EMAIL PROTECTED]

Have you tried running:
apachectl configtest ?



[2003-11-15 13:50:14] peoyli at bredband dot net

the apachectl command gives me no error at all, it just doesn't start
apache, and if I happened to have a previous started httpd (without the
LoadModule line for PHP), I can't use apachectl to stop it either.



[2003-11-15 13:34:04] [EMAIL PROTECTED]

What error does it give?



The remainder of the comments for this report are too long. To view
the rest of the comments, please view the bug report online at
http://bugs.php.net/26263

-- 
Edit this bug report at http://bugs.php.net/?id=26263&edit=1