[PHP-DEV] Bug #14645: Problems assigning by reference to globals

[jan]

From: [EMAIL PROTECTED]
Operating system: Linux
PHP version:  4.1.0
PHP Bug Type: Variables related
Bug description:  Problems assigning by reference to globals

Hi PHP-Developer-Community,

here are
- a description of the bug,
- a short test-script and
- the configuration I used:

A global variable within a function is assigned something by reference .
This value is only visible within this function - not in the main script,
not in functions calls by this function.

The same assignment by value works fine.

This sounds buggy, but it's getting worse! Things get really confusing if
you remove the comment before "#set_by_reference_GLOBALS();".

This function accesses the global variable over the $GLOBALS array. This
assignment works, BUT: After this function is called, the other function I
mentioned before that assigns by value does not work anymore for the rest
of the script-session! (It only works if this line is commented out)!!!

=[test.php]=

=[test.php]=

==[SYSTEM CONFIG]===
System: Linux 2.4.16-SMP
Apache: 1.3.22
PHP: 4.1.0
Configure Params:
./configure
  --prefix=/usr/local/php
  --with-mysql=/usr
  --with-apache=/usr/src/packages/SOURCES/apache_1.3.22
  --with-sybase-ct=/usr/local/freetds
==[SYSTEM CONFIG]===

=[php.ini]==
[PHP]
engine = On
short_open_tag = On
asp_tags = Off
precision=  14
y2k_compliance = Off
output_buffering = 4096
output_handler =
zlib.output_compression = Off
implicit_flush = Off
allow_call_time_pass_reference = Off
safe_mode = Off
safe_mode_gid = Off
safe_mode_include_dir =
safe_mode_exec_dir =
safe_mode_allowed_env_vars = PHP_
safe_mode_protected_env_vars = LD_LIBRARY_PATH
disable_functions =
highlight.string  = #CC
highlight.comment = #FF9900
highlight.keyword = #006600
highlight.bg  = #FF
highlight.default = #CC
highlight.html= #00
expose_php = On
max_execution_time = 30 ; Maximum execution time of each script, in
seconds
memory_limit = 8M  ; Maximum amount of memory a script may consume
(8MB)
error_reporting  =  E_ALL
display_errors = Off
display_startup_errors = Off
log_errors = On
track_errors = Off
warn_plus_overloading = Off
variables_order = "GPCS"
register_globals = Off
register_argc_argv = Off
post_max_size = 8M
pc_order = "GPC"
magic_quotes_gpc = Off
magic_quotes_runtime = Off
magic_quotes_sybase = Off
auto_prepend_file =
auto_append_file =
default_mimetype = "text/html"
doc_root =
user_dir =
extension_dir = ./
enable_dl = On
file_uploads = On
upload_max_filesize = 2M
allow_url_fopen = On
[... unimportant module-config follows ...]
=[php.ini]==

-- 
Edit bug report at: http://bugs.php.net/?id=14645&edit=1


-- 
PHP Development Mailing List 
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #14604 Updated: Empty packets as result of serialization

[jan]

ID: 14604
Updated by: jan
Reported By: [EMAIL PROTECTED]
Status: Analyzed
Bug Type: WDDX related
Operating System: Windows NT 4 sp 6a
PHP Version: 4.1.0
New Comment:

sander, can you also test for PHP 4.0.6 to verify if it broke during the 4.1 release 
circle?


Previous Comments:


[2001-12-22 15:20:20] [EMAIL PROTECTED]

Reproduced on Windows 2000 with 4.1.0 on Apache 1.3.22 (module).

Outpu:




[2001-12-19 10:45:21] [EMAIL PROTECTED]

at least for me your code works in the same way under 4.0.6 and 4.1.0 on FreeBSD 
4.4-STABLE (yes, I finally updated ;):

bash-2.05a$ ./php  test.php 
X-Powered-By: PHP/4.0.6
Content-type: text/html

bash-2.05a$ 

bash-2.05a$ ./php  test.php 
X-Powered-By: PHP/4.1.0
Content-type: text/html

bash-2.05a$ 


bash-2.05a$ cat test.php





[2001-12-19 09:48:07] [EMAIL PROTECTED]

$nazwisko = "Zgoda";
$imie = "Jarek";
$tmp = wddx_packet_start();
wddx_add_vars($tmp, $nazwisko, $imie);
$rs = wddx_packet_end($tmp);
echo $rs;

This code produces empty packet (just structure definition - no data) in 4.1.0, while 
it worked in 4.0.6





Edit this bug report at http://bugs.php.net/?id=14604&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #14663 Updated: Empty packets as result of serialization

[jan]

ID: 14663
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Summary: socket_read() doesn't function as documented
Old Status: Open
Status: Analyzed
Bug Type: Documentation problem
Old Operating System: FreeBSD 4.4
Operating System: Windows NT 4 sp 6a
PHP Version: 4.1.0
New Comment:

the manual page is marked as "EXPERIMENTAL".  so waiting for the extension to become 
stable wil be the best. fixing the example now just to change it again tomorrow is 
useless.

Previous Comments:


[2001-12-22 15:58:12] [EMAIL PROTECTED]

Making a documentation problem. Valid sample still have to be written.



[2001-12-22 15:50:49] [EMAIL PROTECTED]

Example code copy/pasted from http://www.php.net/manual/en/html/ref.sockets.html :

TCP/IP Connection\n";

/* Get the port for the WWW service. */
$service_port = getservbyname ('www', 'tcp');

/* Get the IP address for the target host. */
$address = gethostbyname ('www.php.net');

/* Create a TCP/IP socket. */
$socket = socket_create (AF_INET, SOCK_STREAM, 0);
if ($socket < 0) {
echo "socket_create() failed: reason: " . socket_strerror ($socket) . "\n";
} else {
"socket_create() successful: " . socket_strerror ($socket) . "\n";
}

echo "Attempting to connect to '$address' on port '$service_port'...";
$result = socket_connect ($socket, $address, $service_port);
if ($result < 0) {
echo "socket_connect() failed.\nReason: ($result) " . socket_strerror($result) . 
"\n";
} else {
echo "OK.\n";
}

$in = "HEAD / HTTP/1.0\r\n\r\n";
$out = '';

echo "Sending HTTP HEAD request...";
socket_write ($socket, $in, strlen ($in));
echo "OK.\n";

echo "Reading response:\n\n";
while (socket_read ($socket, $out, 2048)) {
echo $out;
}

echo "Closing socket...";
socket_close ($socket);
echo "OK.\n\n";
?>


When run :

TCP/IP Connection

Warning:  socket_strerror() expects parameter 1 to be long, resource given in 
/usr/home/aragon/squeakbot/test2 on line 18
Attempting to connect to '209.61.157.217' on port '80'...OK.
Sending HTTP HEAD request...OK.
Reading response:


Warning:  socket_read() expects parameter 2 to be long, string given in 
/usr/home/aragon/squeakbot/test2 on line 38
Closing socket...OK.


Regards,
Aragon






Edit this bug report at http://bugs.php.net/?id=14663&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

Re: [PHP-DEV] Bug #14120 Updated: libtool needs help.

[jan]

On Sun, Dec 30, 2001 at 04:58:16PM -0600, Larry Rosenman wrote:
> * Larry Rosenman <[EMAIL PROTECTED]> [011230 16:04]:
> > * Larry Rosenman <[EMAIL PROTECTED]> [011230 15:19]:
> > here is a patch that fixes it for me, but I don't know how to get this
> > into the configure.in stuff... (I can't find it).
> snip.
> 
> Ok, I found it, here are two patches, one to aclocal.m4, and the other
> to configure.  I believe they are equivalent, and fix my issues. 

configure is generated from aclocal.m4. Fixing aclocal.m4 isn't the right
place either as aclocal.m4 is generated by 'aclocal' which takes the
funtions from acinclude.m4 and the M4-files from /usr/share/aclocal/

Can you please find out which M4-file carryies these lines ?

  Jan

-- 
mailto: [EMAIL PROTECTED] weigon @ #php.de (IRCnet)
 http://jan.kneschke.de weigon @ #modlogan (openprojects)

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

Re: [PHP-DEV] Bug #14120 Updated: libtool needs help.

[jan]

On Sun, Dec 30, 2001 at 05:29:26PM -0600, Larry Rosenman wrote:
> > configure is generated from aclocal.m4. Fixing aclocal.m4 isn't the right
> > place either as aclocal.m4 is generated by 'aclocal' which takes the
> > funtions from acinclude.m4 and the M4-files from /usr/share/aclocal/
>
> Where are the /usr/share/aclocal files in the snap tarball?  I don't
> have autoconf installed here...

Ah, I see. Then it up to the packaging team to do this step.

> > Can you please find out which M4-file carryies these lines ?
> where are these in the snap tarball? 

nowhere. They ar part of the autoconf, automake, libtool packages which
generated the configure-file.
 
> Larry Rosenman http://www.lerctr.org/~ler

  Jan

-- 
mailto: [EMAIL PROTECTED] weigon @ #php.de (IRCnet)
 http://jan.kneschke.de weigon @ #modlogan (openprojects)

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

Re: [PHP-DEV] Bug #14120 Updated: libtool needs help.

[jan]

On Sun, Dec 30, 2001 at 05:52:36PM -0600, Larry Rosenman wrote:
> * [EMAIL PROTECTED] <[EMAIL PROTECTED]> [011230 17:49]:
> > On Sun, Dec 30, 2001 at 05:29:26PM -0600, Larry Rosenman wrote:
> > > > configure is generated from aclocal.m4. Fixing aclocal.m4 isn't the right
> > > > place either as aclocal.m4 is generated by 'aclocal' which takes the
> > > > funtions from acinclude.m4 and the M4-files from /usr/share/aclocal/
> > >
> > > Where are the /usr/share/aclocal files in the snap tarball?  I don't
> > > have autoconf installed here...
> > 
> > Ah, I see. Then it up to the packaging team to do this step.
>
> so, Am I out of the loop for now?  

More or less. Your patches should additionally be submitted to  the
autoconf/libtool team (depends on which M4-file is broken).

> Will my patches be considered? 

Perhaps.
 
> I'm trying to help the best I can. 

Can you please submit your patches to the autoconf and libtool list ?

They will ask you for the version of autoconf/automake/libtool you used. It
is best you generate teh configure scripts once on your own with your own
autoconf/automake/libtool setup.

> Larry Rosenman http://www.lerctr.org/~ler

  Jan

-- 
mailto: [EMAIL PROTECTED] weigon @ #php.de (IRCnet)
 http://jan.kneschke.de weigon @ #modlogan (openprojects)

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

Re: [PHP-DEV] Bug #14120 Updated: libtool needs help.

[jan]

On Sun, Dec 30, 2001 at 06:06:50PM -0600, Larry Rosenman wrote:
> * [EMAIL PROTECTED] <[EMAIL PROTECTED]> [011230 18:02]:
> > On Sun, Dec 30, 2001 at 05:52:36PM -0600, Larry Rosenman wrote:
> > > * [EMAIL PROTECTED] <[EMAIL PROTECTED]> [011230 17:49]:
> > > > On Sun, Dec 30, 2001 at 05:29:26PM -0600, Larry Rosenman wrote:
> > > > > > configure is generated from aclocal.m4. Fixing aclocal.m4 isn't the right
> > > > > > place either as aclocal.m4 is generated by 'aclocal' which takes the
> > > > > > funtions from acinclude.m4 and the M4-files from /usr/share/aclocal/
> > > > >
> > > > > Where are the /usr/share/aclocal files in the snap tarball?  I don't
> > > > > have autoconf installed here...
> > > > 
> > > > Ah, I see. Then it up to the packaging team to do this step.
> > >
> > > so, Am I out of the loop for now?  
> > 
> > More or less. Your patches should additionally be submitted to  the
> > autoconf/libtool team (depends on which M4-file is broken).
> Which versions do y'all use to create the tarball? 
> 
> the aclocal stuff looks local to PHP to me

As I've already said:

aclocal.m4 is generated by 'aclocal'. 

I've grepped for the first line of you fix:

$ grep "export_dynamic_flag_spec='-Bexport'" /usr/share/aclocal/*
/usr/share/aclocal/libtool.m4:export_dynamic_flag_spec='-Bexport'

Your fix has to go into the libtool package itself. Ask the libtool
developers for help, please.

http://www.gnu.org/software/libtool/

> > They will ask you for the version of autoconf/automake/libtool you used. It
> > is best you generate teh configure scripts once on your own with your own
> > autoconf/automake/libtool setup.
>
> where can I get the full setup you use?  

Perhaps the CVS requirement apply here, too:

http://www.php.net/anoncvs.php

> Larry Rosenman http://www.lerctr.org/~ler

  Jan

-- 
mailto: [EMAIL PROTECTED] weigon @ #php.de (IRCnet)
 http://jan.kneschke.de weigon @ #modlogan (openprojects)

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #14789 Updated: _SERVER["REQUEST_URI"] not fully given to Caudium 1.0.34

[jan]

ID: 14789
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Bogus
Bug Type: Other web server
Operating System: FreeBSD / Linux
PHP Version: 4.1.1
New Comment:

REQUEST_URI is an apache specific variable. no bug ->bogus.

Previous Comments:


[2002-01-01 09:56:13] [EMAIL PROTECTED]

When using PHP 4 with Caudium, the $REQUEST_URI / _SERVER["REQUEST_URI"] var is not 
full.

For example, look at this phpinfo(): http://aleph1.net/test.php3/toto/a=b&c=d

You get the following line :
_SERVER["REQUEST_URI"] /test.php3  

It should have been : 
_SERVER["REQUEST_URI"] /test.php3/toto/a=b&c=d





Edit this bug report at http://bugs.php.net/?id=14789&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #14807 Updated: core dump

[jan]

ID: 14807
Updated by: jan
Reported By: [EMAIL PROTECTED]
Status: Open
Bug Type: Unknown/Other Function
Operating System: 
PHP Version: 4.1.0
New Comment:

at least for me it doesn't (FreeBSD 4.4 PHP 4.1.0)
can you provide more Information about you environmation (OS, configure options e.g.) 
?


Previous Comments:


[2002-01-02 13:39:24] [EMAIL PROTECTED]

Running the following always causes a core dump:

 array ("0","0","0","0","0"),
2 => array ("0","0","0","0","0"),
3 => array ("0","0","0","0","0"),
4 => array ("0","0","0","0","0"),
5 => array ("0","0","0","0","0")
);

print_r($arr);
echo "\n\n\n\n";
$str = serialize($arr);
echo "serialized\n", $str, "\n\n";

function hex2bin($data) {
$len = strlen($data);
return pack("H" . $len, $data);
}

$enc = urlencode($str);
echo "urlencoded", $enc, "\n\n";
$gzd = gzcompress($enc);
//echo "gzcompressed (urlencoded)", $gzd, "\n\n";
$b64 = base64_encode($gzd);
echo "base64_encoded", $b64, "\n\n";
$b2h = bin2hex($enc);
echo "bin2hex (urlencoded)", $b2h, "\n\n";
$binary = base_convert($b2h, 16, 2);
echo $binary, "\n\n";
$conv = base_convert($binary, 2, 16);
echo $conv, "\n\n";
$h2b = hex2bin($conv);
echo $h2b, "\n\n";
$md = md5($str);
echo "md5", $md, "\n";

?>






Edit this bug report at http://bugs.php.net/?id=14807&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #12313 Updated: parent resolving and deep require's

[jan]

ID: 12313
User updated by: [EMAIL PROTECTED]
Reported By: [EMAIL PROTECTED]
Status: Closed
Bug Type: Scripting Engine problem
Operating System: Linux
PHP Version: 4.0.6
New Comment:

Works with php 4.1.1

Previous Comments:


[2002-01-05 16:46:54] [EMAIL PROTECTED]

No feedback. Closing.



[2001-12-14 14:41:28] [EMAIL PROTECTED]

Do you still have problem with 4.1.0?



[2001-07-23 05:18:35] [EMAIL PROTECTED]

PHP seems to have a problem if it tries to resolve the 
parent of an object that is place in an external file 
which is include over multiple inclusions.

http://jan.kneschke.de/projects/phpbugs/

is telling the whole story including demo code.






Edit this bug report at http://bugs.php.net/?id=12313&edit=1


-- 
PHP Development Mailing List 
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #14887 Updated: cann't configure use --with-oci8=ORACLE_HOME

[jan]

ID: 14887
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Bogus
Bug Type: OCI8 related
Operating System: dec unix 4.0.f
PHP Version: 4.1.0
New Comment:

1) ORACLE_HOME must be defined in your environment
2) you have to use $ORACLE_HOME to access it.

Previous Comments:


[2002-01-06 07:45:56] [EMAIL PROTECTED]

I cann't configure system when I use $./configure
--with-apache=../apache_1.3.22 --with-oci8=ORACLE_HOME
--enable-track-vars .System told me errors found. in
php_4.1.10/ext/oci8/oci8.c,
CALL_OCI_RETURN(connection, '''error.
Please tell what shall I do?
Thank you!






Edit this bug report at http://bugs.php.net/?id=14887&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #1515 Updated: improving php standalone mode

[jan]

ID: 1515
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Analyzed
Status: Closed
Bug Type: Feature/Change Request
Operating System: linux
PHP Version: 4.0
New Comment:

a cli sapi which will cause some more improvements in this area is about
to be commited. the original problem was also solved long ago. closing.

Previous Comments:


[2001-11-11 17:30:46] [EMAIL PROTECTED]

There isn't a one-letter param to turn off html_errors, however you can
pass '-d html_errors=0'.  Another option is to set it in the php.ini
file which is checked by the php binary.



[2001-11-11 14:21:32] [EMAIL PROTECTED]

IIRC this is implemented (-s or something)... can someone confirm this?



[2001-02-10 13:23:05] [EMAIL PROTECTED]

refiling against 4.0.



[2000-05-19 09:56:14] [EMAIL PROTECTED]

php -q is a step in the right direction
but this only disables http headers, 
not html markup in error messages

and php could be more clever when started
standalone (that is, if istty(stdin) is true
or certain environment variables are not set)







[1999-11-11 18:04:10] [EMAIL PROTECTED]

Try php -f or php -q. Is this what you are looking for? I often use it
for systems programming.



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/?id=1515


Edit this bug report at http://bugs.php.net/?id=1515&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #4587 Updated: Would like exit and die to return an int

[jan]

ID: 4587
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Closed
Bug Type: Feature/Change Request
Operating System: all
PHP Version: 4.04
New Comment:

fixed in CVS. closing.

Previous Comments:


[2001-03-02 12:47:33] [EMAIL PROTECTED]

Any update on this request?  Is there an existing way in the language
that I can force a non-zero return code to be passed back to the OS?



[2000-05-24 21:10:44] [EMAIL PROTECTED]

I think PHP is awesome.  I am a build engineer and am going to replace
all of my shell scripts and perl scripts with PHP.  One thing that is
currently hurting me is that exit and die do not return a value.

If I have a makefile with a target, and I run a PHP script to build that
target, I want the PHP script to return a value.  That way, make will
behave properly.

all: target1 target2

target1:
   php -q -f script1.php

target2:
   php -q -f script2.php

If script1.php does not execute successfully, I do not want script2.php
to run.

This problem isn't limited to make.  If I have stand alone scripts that
are called from other external programs, it is extremely important to be
able to pass back a return code.

If it causes too much of a problem to change the signature of the exit
and die functions, I'd be perfectly happy if there was a new function
that did exit with a return code,
ie 

int exit_rc(int value)

Thanks for your consideration and thanks for a great language!





Edit this bug report at http://bugs.php.net/?id=4587&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #5343 Updated: a way of implementing class methods

[jan]

ID: 5343
Updated by: jan
Reported By: [EMAIL PROTECTED]
Status: Open
Bug Type: Feature/Change Request
Operating System: 
PHP Version: 4.0.1
New Comment:

can you propose this on  [EMAIL PROTECTED] ?

Previous Comments:


[2000-07-03 17:10:59] [EMAIL PROTECTED]

How about:
function Foo::Bar() {
//implementation of method `Bar' in class `Foo'
}

ppl usually separate the functionality in different files, and with
classes this is only possible by forwarding calls to regular
functions,
or by making use of derivation, which is not the best practice.

IMHO that would imply small changes to the parser (to recognize `::'
syntax), what do you think?






Edit this bug report at http://bugs.php.net/?id=5343&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #6136 Updated: Exception Handling

[jan]

ID: 6136
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Closed
Bug Type: Feature/Change Request
Operating System: linux
PHP Version: 4.0 Latest CVS (13/08/2000)
New Comment:

this will come with ZendEngine2 (PHP5). closing.

Previous Comments:


[2000-08-13 10:21:44] [EMAIL PROTECTED]

Error handling: please please please add exception handling (some sort
of "try ... catch ... finally" as implemented in Java, C++, Delphi, ...)
to PHP.

It is simply annoying to check function results for errors and to do
error handling in if-statements.

Especially for the new MySQL-transactions such an exception-handling
would be great.





Edit this bug report at http://bugs.php.net/?id=6136&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #6380 Updated: Session handlers should be allowed in objects

[jan]

ID: 6380
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Closed
Bug Type: Feature/Change Request
Operating System: 
PHP Version: 4.0 Latest CVS (27/08/2000)
New Comment:

this is possible using
session_set_save_handler(array('class'=>'method'),...);
or do I misunderstand the problem here?

Previous Comments:


[2000-08-27 14:32:53] [EMAIL PROTECTED]

It would be nice (for encapsulation purposes) to allow user-defined
session handlers to be implemented as class methods, either through the
established array($this, "foo") syntax or via a session_set_object()
function.





Edit this bug report at http://bugs.php.net/?id=6380&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #6701 Updated: Use of http_proxy for fopen("http:// ...

[jan]

ID: 6701
Updated by: jan
Reported By: [EMAIL PROTECTED]
Status: Open
Bug Type: Feature/Change Request
Operating System: Customized RedHat 6.2
PHP Version: 4.0.0
New Comment:

although the request should remain open, this specific problem can be
solved using PEAR::Cache.

Previous Comments:


[2000-09-12 23:39:12] [EMAIL PROTECTED]

It would be nice if fopen("http://www.mywebsite.com";, "r"); used the
value in http_proxy as a proxy server instead of always fetching the
data directly.  Using CURL is an option, but its a lot more complex.  If
this functionality isn't too difficult to achieve, it would be much
appreciated.

For example, I have a PHP program that HTMLizes RFCs and displays them. 
It grabs the rfc###.txt directly from http://www.ietf.org/ but there's
no reason for it to refetch them every time if they can be cached by our
proxy server.

Note: blanking $http_proxy should disable the proxy.





Edit this bug report at http://bugs.php.net/?id=6701&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #7276 Updated: CVSup/rsync/easier patch access desired

[jan]

ID: 7276
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Closed
Bug Type: Feature/Change Request
Operating System: n/a
PHP Version: 4.0.3
New Comment:

no bug, see http://www.php.net/anoncvs.php

Previous Comments:


[2000-10-17 07:59:52] [EMAIL PROTECTED]

Hello,

a few days ago I downloaded 4.0.3 and now seem to have to download
4.0.3pl1. I'd rather not download everything and  then also re-configure
everything. I'd rather download a patch kit and apply that, keeping my
existing configuration, and run, or I'd like to cvsup a source tree and
start from there. Unfortunately you don't seem to support cvsup or the
like, and also don't offer a plain FTP service, less so one with patch
kits, or at least I couldn't find it.

A comment is appreciated since your CVS server also doesn't exactly
scream (may also be a network problem, of course).


Best Regards,
--Toni++





Edit this bug report at http://bugs.php.net/?id=7276&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #7480 Updated: next(), end(), prev(), reset() etc. don't return references

[jan]

ID: 7480
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Closed
Bug Type: Feature/Change Request
Operating System: RedHat Linux 7.0
PHP Version: 4.0.1pl2
New Comment:

dupe of 7578. 

Previous Comments:


[2000-10-26 09:26:40] [EMAIL PROTECTED]

next(), end(), prev(), reset() etc. don't return references

meaning this example code won't work as it's thought to do:







Edit this bug report at http://bugs.php.net/?id=7480&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #7766 Updated: not OO until automatic class destructor is added

[jan]

ID: 7766
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Closed
Bug Type: Feature/Change Request
Operating System: FreeBSD Apache
PHP Version: 4.0.3pl1
New Comment:

we'll have destructors with ZE2 (PHP5).  closing

Previous Comments:


[2000-11-12 04:49:46] [EMAIL PROTECTED]

Moved to feature requests




[2000-11-12 04:33:56] [EMAIL PROTECTED]

Then why add classes?  Are you implying to businesses that if they
invest in your features, you do not intend to complete them?



[2000-11-12 04:21:58] [EMAIL PROTECTED]

Then why add classes?  Are you implying to businesses that if they
invest in your features, you do not intend to complete them?



[2000-11-12 04:00:29] [EMAIL PROTECTED]

Nobody ever claimed that PHP was an OO language.



[2000-11-12 03:24:44] [EMAIL PROTECTED]

The lack of an automatic class destructor makes it impossible to make
code completely OO.

For example, we have a Table class which outputs an HTML  with
error checking.  It is impossible to detect the case where
Table::EndTable() is not called ( not issued) because class
destructors do not exist in the PHP4 language.

PLEASE CORRECT THIS GLARING DEFICIENCY ASAP!

I am a prolific commercial programming of $million products (Corel
Painter, Cool Page, etc).  PHP4 is a an incredible tool.  We are
considering purchasing the new compiler technology.  So please fix this
silly omission.  Please.  Please.





Edit this bug report at http://bugs.php.net/?id=7766&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #7480 Updated: next(), end(), prev(), reset() etc. don't return references

[jan]

ID: 7480
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Closed
Status: Bogus
Bug Type: Feature/Change Request
Operating System: RedHat Linux 7.0
PHP Version: 4.0.1pl2
New Comment:

dupes are bogus..

Previous Comments:


[2002-01-06 11:47:32] [EMAIL PROTECTED]

dupe of 7578. 



[2000-10-26 09:26:40] [EMAIL PROTECTED]

next(), end(), prev(), reset() etc. don't return references

meaning this example code won't work as it's thought to do:







Edit this bug report at http://bugs.php.net/?id=7480&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #7480 Updated: next(), end(), prev(), reset() etc. don't return references

[jan]

ID: 7480
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Bogus
Status: Duplicate
Bug Type: Feature/Change Request
Operating System: RedHat Linux 7.0
PHP Version: 4.0.1pl2
New Comment:

actually they are dublicates, silly me.

Previous Comments:


[2002-01-06 12:01:10] [EMAIL PROTECTED]

dupes are bogus..



[2002-01-06 11:47:32] [EMAIL PROTECTED]

dupe of 7578. 



[2000-10-26 09:26:40] [EMAIL PROTECTED]

next(), end(), prev(), reset() etc. don't return references

meaning this example code won't work as it's thought to do:







Edit this bug report at http://bugs.php.net/?id=7480&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #7899 Updated: Class selfdestruct

[jan]

ID: 7899
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Closed
Bug Type: Feature/Change Request
Operating System: any
PHP Version: 4.0 Latest CVS (21/11/2000)
New Comment:

we'll have destructors and more improved OO with ZE2 (PHP5). please
propose anything to [EMAIL PROTECTED] closing

Previous Comments:


[2000-11-21 04:41:15] [EMAIL PROTECTED]

Hiya 

I'd like to see a destructor for classes from within the class 

example:

$file = new file("somefile.ext);
// instert some long script here
$file->del_file();
at this point $file should be FALSE of NULL (set from within the
class)

some ideas:
unset($this);
or $this->Destruct();

Greetz,

Wico





Edit this bug report at http://bugs.php.net/?id=7899&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #6768 Updated: list() construct reference assignment

[jan]

ID: 6768
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Duplicate
Bug Type: Feature/Change Request
Operating System: Linux 2.2.14
PHP Version: 4.0.2
New Comment:

dupe of #7930 and marking it "dublicate"

Previous Comments:


[2000-09-15 05:28:18] [EMAIL PROTECTED]

I'm requesting that that the list() construct which is an assignment
construct, assign variable references. Like such..







Edit this bug report at http://bugs.php.net/?id=6768&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #7974 Updated: N/A

[jan]

ID: 7974
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Closed
Bug Type: Feature/Change Request
Operating System: N/A
PHP Version: 4.0.3pl1
New Comment:

please take part in the discussion on [EMAIL PROTECTED] about the
upcoming features in ZE2. PHP4 is not likely to see it.

Previous Comments:


[2000-11-25 20:02:41] [EMAIL PROTECTED]

Add multiple inheritance to Classes.





Edit this bug report at http://bugs.php.net/?id=7974&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #8058 Updated: Missing PDF functionality

[jan]

ID: 8058
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Summary: 
Old Status: Open
Status: Closed
Bug Type: Feature/Change Request
Operating System: linux 2.2.x
PHP Version: 4.0.3pl1
New Comment:

closing, much has been added since the request came in. reopen it if
something is still missing.

Previous Comments:


[2000-11-30 15:19:09] [EMAIL PROTECTED]

I'd like to see some new function into PDF support like:

- URL links functions (PDFLib do that)
- Form support so we can interface with FDF.

Thanks





Edit this bug report at http://bugs.php.net/?id=8058&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #8197 Updated: lack of socket() functions

[jan]

ID: 8197
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Closed
Bug Type: Feature/Change Request
Operating System: Windows
PHP Version: 4.0.3pl1
New Comment:

they should be now part of the windows distribution. closing

Previous Comments:


[2000-12-11 16:58:28] [EMAIL PROTECTED]

Please include a DLL (or built-in) for the socket() functions in a Win32
build.





Edit this bug report at http://bugs.php.net/?id=8197&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #8214 Updated: Request for Signal Handling functions

[jan]

ID: 8214
Updated by: jan
Reported By: [EMAIL PROTECTED]
Status: Open
Bug Type: Feature/Change Request
Operating System: *nix - CGI
PHP Version: 4.0 Latest CVS (12/12/2000)
New Comment:

doesen't pcnt do the job?

Previous Comments:


[2000-12-12 16:49:52] [EMAIL PROTECTED]

I would rather see:
posix_register_sig_func(char function, mixed signal)

And handle both "9" and "KILL" as valid signals.



[2000-12-12 16:35:41] [EMAIL PROTECTED]

A few additions to the posix module to allow registering functions when
certain signals are recieved would be cool. somthing like

posix_register_sig_func(char function,int signal);

where signal is HUP etc :)

would allow some long running systems in CGI mode with a graceous
shutdown. 


James





Edit this bug report at http://bugs.php.net/?id=8214&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #8599 Updated: a kind Exception handling

[jan]

ID: 8599
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Closed
Bug Type: Feature/Change Request
Operating System: 
PHP Version: 4.0.4
New Comment:

We'll have exception handling in PHP5, please follow the development of
the ZendEngine2 on [EMAIL PROTECTED] closing.

Previous Comments:


[2001-01-10 01:58:07] [EMAIL PROTECTED]

http://php.net/manual/function.set-error-handler.php
set_error_handler lets you register your own error handling function

however try...catch is also a much-needed feature



[2001-01-08 13:48:53] [EMAIL PROTECTED]

something like java's exception handling
or more kind of register_shutdown_function i.e
register_mysql_errorhandling_function

thnx for the attention





Edit this bug report at http://bugs.php.net/?id=8599&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #8681 Updated: Function arguments in 'global'

[jan]

ID: 8681
Updated by: jan
Reported By: [EMAIL PROTECTED]
Status: Open
Bug Type: Feature/Change Request
Operating System: Linux/FreeBSD
PHP Version: 4.0.4
New Comment:

so what is this anyway?

Previous Comments:


[2001-02-22 10:52:15] [EMAIL PROTECTED]

Not a bug.



[2001-01-12 13:11:42] [EMAIL PROTECTED]

function foo ($bar)
{
  global $bar;
  # ...
}

Opposite to php3, php4 doesn't warn you. This is now documented.





Edit this bug report at http://bugs.php.net/?id=8681&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #8863 Updated: function to get constant values by variable name

[jan]

ID: 8863
Updated by: jan
Reported By: [EMAIL PROTECTED]
Status: Open
Bug Type: Feature/Change Request
Operating System: Any
PHP Version: 4.0.4pl1
New Comment:

see get_defined_constants

Previous Comments:


[2001-01-23 12:21:28] [EMAIL PROTECTED]

I believe PHP lacks a function to get constant values by
name where the constant's name is stored in a string variable.

What I mean is something like

function get_constant($const_name)
{
eval ("\$myvar = " . $const_name . ";");
return $myvar;
}

but I really think that using eval() here is a waste.

Am I missing something?

andrew





Edit this bug report at http://bugs.php.net/?id=8863&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #8863 Updated: function to get constant values by variable name

[jan]

ID: 8863
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Closed
Bug Type: Feature/Change Request
Operating System: Any
PHP Version: 4.0.4pl1
New Comment:

and close it

Previous Comments:


[2002-01-06 13:00:36] [EMAIL PROTECTED]

see get_defined_constants



[2001-01-23 12:21:28] [EMAIL PROTECTED]

I believe PHP lacks a function to get constant values by
name where the constant's name is stored in a string variable.

What I mean is something like

function get_constant($const_name)
{
eval ("\$myvar = " . $const_name . ";");
return $myvar;
}

but I really think that using eval() here is a waste.

Am I missing something?

andrew





Edit this bug report at http://bugs.php.net/?id=8863&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #8974 Updated: Blank lines in includes cause session errors

[jan]

ID: 8974
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Bogus
Bug Type: Feature/Change Request
Operating System: All
PHP Version: 4.0.4pl1
New Comment:

technically it is an user error. to solve it either eliminate any
whitespace or call ob_start() before session_start(). the same thing
applies for header(). no bug. bogus

Previous Comments:


[2001-01-29 00:42:42] [EMAIL PROTECTED]

A PHP script with includes followed by commands which set headers (ie
session_start) will generate errors if include file(s) contain blank
lines. For example:



where functions.inc contains leading or trailing blank lines will result
in an error like:

Warning: Cannot send session cache limiter - headers already sent

This problem appears to be very common (according to a search on
google), and has no doubt frustrated many PHP users.

Unfortunately this behaviour is technically 'correct', although it
should probably be fixed anyway.






Edit this bug report at http://bugs.php.net/?id=8974&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #8978 Updated: Add a 'readonly' possibility to the session module

[jan]

ID: 8978
Updated by: jan
Reported By: Maxim Derkachev <[EMAIL PROTECTED]>
Old Status: Analyzed
Status: Closed
Bug Type: Feature/Change Request
Operating System: 
PHP Version: 4.0.4pl1
New Comment:

seems fixed then. reopen if I am wrong

Previous Comments:


[2001-05-16 04:17:12] Maxim Derkachev <[EMAIL PROTECTED]>

Forgot to include the batteries :)
After the patch above is applied, one could do:
session_start(SESS_READ_ONLY);
to start a readonly session. 
Functions that supposed to write the session data (core session
functions, not actual savehandler functions) will be disabled.
On the other page, if session_start() is called without the 
SESS_READ_ONLY flag, one could get the normal fully functional session,
which will write the data. That would allow to use session in framed
pages, when one frame is allowed to change the session data and another
frames only read the data, and in many other cases. E.g. for me the
feature has become inevitable when I needed to write a support chat,
which should read session variables, but should not change them and, the
most important, it should not save them, because a client could browse
other parts of the site  (and this could affect the sesson vars) while
he is chatting with the support. Without the readonly possibility, the
new session variables could be easily rewrited by the chat script with
outdated values.



[2001-05-16 04:02:23] Maxim Derkachev <[EMAIL PROTECTED]>

just made a patch against the current sources (session.c and
php_session.h).

*** php_session.h.orig  Tue May 15 15:16:50 2001
--- php_session.h   Tue May 15 15:23:26 2001
***
*** 96,100 
--- 96,103 
zend_bool define_sid;
zend_bool use_cookies;
+   int readonly;
  } php_ps_globals;
+
+ #define SESS_READONLY 1

  extern zend_module_entry session_module_entry;
*** session.c.orig  Tue May 15 15:16:04 2001
--- session.c   Wed May 16 11:54:31 2001
***
*** 526,529 
--- 526,533 
PLS_FETCH();

+   if (PS(readonly)) {
+   return;
+   }
+
if (!PG(register_globals)) {
if (!PS(http_session_vars)) {
***
*** 899,902 
--- 903,911 
zend_bool retval = SUCCESS;

+   if (PS(readonly)) {
+   php_error(E_WARNING, "Trying to destroy readonly
session");
+   return FAILURE;
+   }
+
if (PS(nr_open_sessions) == 0) {
php_error(E_WARNING, "Trying to destroy uninitialized
session");
***
*** 1265,1270 
--- 1274,1297 
  PHP_FUNCTION(session_start)
  {
+   pval **flag;
PSLS_FETCH();

+   if (ZEND_NUM_ARGS() > 1)
+   WRONG_PARAM_COUNT;
+
+   if (ZEND_NUM_ARGS() == 0 ) {
+   PS(readonly) = 0;
+   }
+   if (ZEND_NUM_ARGS() == 1 && zend_get_parameters_ex(1, &flag) !=
FAILURE) {
+   convert_to_long_ex(flag);
+   if (((int) ((*flag)->value.lval)) == SESS_READONLY) {
+   PS(readonly) = 1;
+   }
+   else {
+   PS(readonly) = 0;
+   }
+   }
+
+
php_session_start(PSLS_C);

***
*** 1314,1317 
--- 1341,1347 
PSLS_FETCH();

+   if (PS(readonly))
+   return;
+
if (PS(nr_open_sessions) == 0)
RETURN_FALSE;
***
*** 1353,1356 
--- 1383,1388 
PSLS_FETCH();

+   REGISTER_LONG_CONSTANT("SESS_READ_ONLY", SESS_READONLY,
CONST_CS);
+
php_rinit_session_globals(PSLS_C);

***
*** 1404,1407 
--- 1436,1440 
PS(module_number) = module_number;
REGISTER_INI_ENTRIES();
+   REGISTER_LONG_CONSTANT("SESS_READ_ONLY", SESS_READONLY,
CONST_CS);
return SUCCESS;
  }








[2001-01-29 06:21:31] Maxim Derkachev <[EMAIL PROTECTED]>

Just faced the fact that the possibility to call session 'readonly' 
should be added. 
For example, when somebody calls a framed pages where all 
frames are php scripts that needs session variables. But in this 
case only one of them should be allowed to write session state, 
because every frame would write session state in an unpredictable order,

and variables registered/changed in one frame could be overwritten 
by other frames, and that would definitely break an application. 
I suggest session_start could take an optional READONLY flag to 
disable write of the session data during the page shutdown.
The idea is 

[PHP-DEV] Bug #9195 Updated: Default class function arguments

[jan]

ID: 9195
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Analyzed
Bug Type: Feature/Change Request
Operating System: Linux
PHP Version: 4.0.2
New Comment:

please take part in the discussion about the upcoming ZendEngine2 on
[EMAIL PROTECTED] 

Previous Comments:


[2001-02-09 13:00:53] [EMAIL PROTECTED]

I'm requesting that the default function arguments for a class method be
able to reference the class variable members in it's definition. 

Here is an example of it's use:

class searchclass {

var $hits;
var $results;

function save($filename, $start = 0, $end = $this->hits){ /* <- this
returns Parse error: */
if($fp = @fopen($filename, "w")){
for($i = $start; $i < $end; $i++){
fwrite($fp, implode("|", $this->results[$i]));
}
}else{
return false;
}
}

function search(){
/* ... */
}

}





Edit this bug report at http://bugs.php.net/?id=9195&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #9250 Updated: Nested Functions are BAD, they make for inefficient debugging

[jan]

ID: 9250
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Bogus
Bug Type: Feature/Change Request
Operating System: Any
PHP Version: 4.0.4pl1
New Comment:

brackets are used for other controll structures as well, so disabling
nested functions do not solve the problem. ->bogus

Previous Comments:


[2001-02-14 04:10:46] [EMAIL PROTECTED]

If you have a missing close-bracket "}" somewhere in your code, the PHP
parser tells you that the error is at the last line.  If you have a
multi-hundred or thousand line file, this makes life a bitch.

Rasmus tells me that the reason for this behavior is because functions
can be nested, which means that:

function a 
{
  function b
  {
  } 
}
is OK. This is a silly feature.  Why do you have it, for scoping
reasons?  It means that you can't effectively tell where syntax errors
occur.

Suggestion: Either make nested functions turned off by default but you
can turn them on with a PHP global variable, like
$PHP_BIZARRO_NESTED_FUNCTIONS = 1;

or

have the parser syntax set up so that it marks where a possible syntax
error occurred (IOW it sees a possible nested function, but it could
also be a syntax error), and if it reaches the end of the file with a
missing "}" or two, that it displays the location where it parsed the
nested function.





Edit this bug report at http://bugs.php.net/?id=9250&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #9266 Updated: Unable to load 14 of php's extensions

[jan]

ID: 9266
Updated by: jan
Reported By: [EMAIL PROTECTED]
Status: Analyzed
Old Bug Type: Feature/Change Request
Bug Type: Documentation problem
Operating System: WinMe, Win98, Win2000
PHP Version: 4.0.4pl1
New Comment:

so better documentation is needed for extensions that rely on external
libraries.

Previous Comments:


[2001-02-14 20:07:11] [EMAIL PROTECTED]

Yes, a more detailed error message, or/and two more lines in the manual
could have saved me some time. And maybe it would for others too.

Thank you.



[2001-02-14 16:49:32] [EMAIL PROTECTED]

all of theese extensions depend on other DLLs 
being installed on your system
like, for example, sybase_ct client DLLs
(this is the native sybase call interface, _not_ the ODBC stuff)

maybe we could improve the error message?
(moved to feature request)



[2001-02-14 16:49:10] [EMAIL PROTECTED]

all of theese extensions depend on other DLLs 
being installed on your system
like, for example, sybase_ct client DLLs
(this is the native sybase call interface, _not_ the ODBC stuff)

maybe we could improve the error message?
(moved to feature request)



[2001-02-14 16:41:08] [EMAIL PROTECTED]

I spent almost 2 day trying to figure out how to load the sybase
extension under windows. I tried everything I could, but 14 of the
extensions would not load under WinMe, Win98, Win2000, with any
combination of apache, pws, iis. The extension_dir is just set all fine,
and the rest of the extensions load and function well.

I only wanted to get some data from a sybase db. I could actually solve
it with ODBC, but I still thought it's not the real way.

I discussed this problem on the local php mailing list, everyone knew
it, noone knew why. I searched through the php manual and the rest of
the web just to find a fraction of information that could be helpful
either way, but there was no.

The extensions that made me 2 horribe days:

php_curl.dll
php_fdf.dll
php_gettext.dll
php_ifx.dll
php_interbase.dll
php_ldap.dll
php_mssql65.dll
php_mssql70.dll
php_oci8.dll
php_openssl.dll
php_oracle.dll
php_sablot.dll
php_sybase_ct.dll
php_yaz.dll

A sample message:
PHP Warning:  Unable to load dynamic library
'c:\php\extensions\php_sybase_ct.dll'

My question is if it is possible to load these extensions some way, or
should I give up trying, becouse they will never load.

Thank you a lot.

I wish you nice developing.

Best regards:

Mike Myers





Edit this bug report at http://bugs.php.net/?id=9266&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #9558 Updated: array hanling

[jan]

ID: 9558
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Feedback
Bug Type: Feature/Change Request
Operating System: All OS
PHP Version: 4.0.4pl1
New Comment:

what about unlink($array['key']); $first=$array[0]; and
$last=$array[count($array)-1]. ?

Previous Comments:


[2001-03-05 06:32:06] [EMAIL PROTECTED]

There are no normal delete from array (by the key) there are no
possibility to get last, first element without altering the array itself
(or to examine the element), in general in PHP the array is the list
replacement (2 in 1)(for ex. in C++ we have to separate types/classes:
array and the list)  but there are no list functionality. If this will
cause performance loses then new type creation sugested.





Edit this bug report at http://bugs.php.net/?id=9558&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #1027 Updated: fprintf() please

[jan]

ID: 1027
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Duplicate
Bug Type: Feature/Change Request
PHP Version: 4.0
New Comment:

was duped by 14949 which is more descriptive ;)


Previous Comments:


[2001-07-22 12:27:39] [EMAIL PROTECTED]

Is fprintf going to be added or should this request be closed?



[2001-02-10 12:55:39] [EMAIL PROTECTED]

really moved to 4.0 this time.



[2001-02-10 12:40:17] [EMAIL PROTECTED]

still no fprintf. refiled against 4.0.



[1999-01-05 04:09:14] [EMAIL PROTECTED]

Quite often my PHP3-scripts take input from the user, and writes it into
a file. I can do

  fputs($fp,sprintf("%2d\n",$variable));

but it really looks ghastly. Fprintf would be nice to have.

I would also suggest that you take all the stdio-derived functions which
is dispersed throughout the manual (in Filesystem and Strings) and
collect them in a new section called "File I/O" or equivalent.






Edit this bug report at http://bugs.php.net/?id=1027&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #14952 Updated: $_GET and $HTTP_GET_VARS are separate arrays

[jan]

ID: 14952
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Bogus
Bug Type: Feature/Change Request
Operating System: Linux
PHP Version: 4.1.1
New Comment:

the $_* arrays are intended to be seperate from ther $HTTP_*_VARS
arrays.
The new arrays are so called "super-global". You don't hve to introduce
them nito function-scope with global. This is a feature no bug => Bogus


Previous Comments:


[2002-01-09 12:38:13] [EMAIL PROTECTED]

blah.php
";
$_GET['test'] = 1;
echo "";

print_r( $HTTP_GET_VARS );
print_r( $_GET );

?>

Results from blah.php
-
Array ( ) Array ( ) 
Array ( ) Array ( [test] => 1 )

Results from blah.php?test=test
-
Array ( [test] => test ) Array ( [test] => test ) 
Array ( [test] => test ) Array ( [test] => 1 ) 


Aside from the performance/memory implications of having two separate
arrays (ie: http://blah.com/blah.php?text=[shakespeare]&spell_check=1),
it's really annoying when trying to propagate information to a
moduralized script which normally takes data off the get-string.

Here is my patch:  $_GET =& $HTTP_GET_VARS;   It halves the memory
requirements of PHP's default arrays, and behaves much closer to what an
end-user would expect.  I am willing to bet that NOBODY wants
$_GET['blah'] = 'test'; echo $HTTP_GET_VARS['blah']; to fail in
mysterious ways.

Why is this not done in the first place?  I have checked on the mailing
lists and in the bug archives, but couldn't find any previous
discussion.  Send me an email at [EMAIL PROTECTED] if there is a good
reason for this, or if I am missing something important.





Edit this bug report at http://bugs.php.net/?id=14952&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #14785 Updated: cannot save data with session_register() in functions

[jan]

ID: 14785
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Closed
Status: Bogus
Bug Type: Session related
Operating System: Debian Linux
PHP Version: 4.1.0
New Comment:

no bugs can't be closed *nitpick* -> bogus


Previous Comments:


[2002-01-10 17:35:53] [EMAIL PROTECTED]

Indeed, not a bug, thx for explaining!. Please also note that you can
use this in 4.1.x and higher:
$_SESSION['var'] = 'somethinghere';

It registers the session var 'var', whcih can be accessed through your
whole script (in functions, or outside functions) as $_SESSION['var'].

Derick



[2002-01-10 17:31:56] [EMAIL PROTECTED]

This is not a Bug. All variables that you plan to register must be
global!

Reason:
The PHP-function session_register() only  'remember' the *names* of the
variables (not the content) it must save when the script terminates.
The session vars are read at script end. This happens outside of any
function or methode and therefor only global vars that have been set
have a value; all others are unset.



[2002-01-01 01:18:46] [EMAIL PROTECTED]

session_register() doesnt seem to save anything in functions, eg.

session_start();

function bob() {
$var = "somethinghere";
session_register("var");
}

bob();
echo session_encode();

the above only registers the var name, not the data in it.

if it helps theres a way around it, all you gota do is make the $var
global so instead of having $var have $GLOBALS['var'] and it works
fine.

might be little bug, but it stuffed me up 4 nights in a row.

oh and the way i compiled php was just with apt-get install php4 in
debian, (newbie here)





Edit this bug report at http://bugs.php.net/?id=14785&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #14998 Updated: Can't name a class 'Directory'

[jan]

ID: 14998
Updated by: jan
Old Summary: Can't name a class 'Directory'
Reported By: [EMAIL PROTECTED]
Status: Open
Old Bug Type: Class/Object related
Bug Type: Documentation problem
Operating System: Red Hat Linux 7.0
PHP Version: 4.0.5
New Comment:

See http://www.php.net/manual/en/function.get-declared-classes.php. It
is predefined. Should be pointed out more clearly, -> documentation
problem.


Previous Comments:


[2002-01-11 12:17:37] [EMAIL PROTECTED]

For example, given the code:



PHP gives the error: "Cannot redeclare class directory". Maybe the the
dir() function/class interferes??

Configuration:

'./configure' '--prefix=/usr' '--with-apxs=/usr/sbin/apxs'
'--with-exec-dir=/usr/bin' '--with-config-file-path=/etc/httpd'
'--with-regex=system' '--enable-debugger' '--enable-magic-quotes'
'--enable-sysvshm' '--with-dom' '--enable-force-cgi-redirect'
'--enable-sigchild' '--with-wddx' '--enable-inline-optimization'
'--with-gnu-ld' '--enable-bcmath' '--enable-crypt' '--with-xml'
'--with-sablot' '--enable-dbg=shared' '--with-dbg-profiler'

Thanks.






Edit this bug report at http://bugs.php.net/?id=14998&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #15158 Updated: Session get replaced when moving to other domain

[jan]

ID: 15158
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Feedback
Bug Type: Session related
Operating System: RH7.1
PHP Version: 4.1.1
New Comment:

Please check your session.cookie_domain setting (php.ini directive).


Previous Comments:


[2002-01-22 05:17:40] [EMAIL PROTECTED]

When I move from www.domain.com to secure.domain.com with the PHPSESID
in the URL the session from the previous domain continue to work in the
new one until the first inner link that does not have the PHPSESID is
clicked.

When such link is clicked ( or submitted ) the session_start/register
initiate new session to the domain and forget about the old session.

This behavior started after I upgraded to 4.1.1 from 4.0.6.

IMHO
In 4.0.6 the session was accepted for the domain ( as default ) after
PHPSESID was delivered but in 4.1.1 its not.





Edit this bug report at http://bugs.php.net/?id=15158&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #15162 Updated: (int) cast doesn't work properly

[jan]

ID: 15162
Updated by: jan
Old Summary: (int) cast doesn't work properly
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Bogus
Bug Type: Math related
Operating System: slackware linux (kernel 2.2.19)
PHP Version: 4.1.1
New Comment:

no bug in PHP, bogus


Previous Comments:


[2002-01-22 09:07:26] [EMAIL PROTECTED]

Or to give a shorter example:



which outputs:

  float(2)
  int(1)

This only happens when a subtraction is made. when (1.2 - 1) is
replaced by 0.2, this feature doesn't show up. 

Either I really overlooked something or this is a bug.

Anyway, I suggest you to calculate only with cents - which also allows
easier calculating. Formatting can be done later.

  $price = 3340;

then use number_format() to format.

Kind Regards,
  Daniel Lorch



[2002-01-22 08:45:33] [EMAIL PROTECTED]

i'm writing a little internet shop in php.
i found really bad bug, which can make a lot of losses to somebody...

may be i'm using cast not good way, but i call this bug anyway.

consider this script:

";

$price_dolars = floor($price);

echo "number of cents is: " . ($price - $price_dolars)*100 . "";
// we can see value of 44

echo "but as integer it would be: " . (int)(($price -
$price_dolars)*100) . "";
// but now it is 43

?>






Edit this bug report at http://bugs.php.net/?id=15162&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #15164 Updated: can php do a substraction? ;)

[jan]

ID: 15164
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Bogus
Bug Type: Math related
Operating System: linux 2.2.19
PHP Version: 4.1.1
New Comment:

floats are not supposed to be exact (per definition). do not use them
on currencies. calculate cents ant use (agein) number_format() to
display them prooperly. No bug in PHP, bogus.


Previous Comments:


[2002-01-22 09:17:32] [EMAIL PROTECTED]

my previous bug report was 15162. i wrote about (int) cast there.
another bug may be a reason of that.

i do: (267.57 - 267) * 100 
and i get: 57.
(i expected 57)
[may be i work too much...]

consider this:

";
?>






Edit this bug report at http://bugs.php.net/?id=15164&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #15187 Updated: PHP parses untrusted documents (or something like that ;P)

[jan]

ID: 15187
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Bogus
Bug Type: Scripting Engine problem
Operating System: Linux 2.4
PHP Version: 4.1.1
New Comment:

I call it a user error. no external sources are to be trusted with no
error/security/integrity check.


Previous Comments:


[2002-01-23 15:30:20] [EMAIL PROTECTED]

Hello,

I think there is a pretty large security issue in PHP. E.g. we have 2
servers, we call server 1 'embrace', and name server 2 'kossy'.

on embrace we create a file index.php:
http://kossy/issue.php";); ?>

on kossy we create a file issue.php:


well, if we now type: http://embrace/index.php?cmd=ls we see the 'ls'
output from kossy. This is the way it should be.. BUT, when we disable
PHP op kossy we get a rather nasty bug (imho).

If we type http://embrace/index.php?cmd=ls again (where kossy has no
PHP support) embrace includes the PHP source code from kossy and then
parses the file, which will give us the 'ls' result on embrace.

This can be used for numerous attacks. And I PHP should only include
parsed PHP or non-PHP files from external (http://) links.

Bye,
Tozz






Edit this bug report at http://bugs.php.net/?id=15187&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #15229 Updated: unable to load dynamic module

[jan]

ID: 15229
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Feedback
Bug Type: COM related
Operating System: win2000
PHP Version: 4.1.1
New Comment:

by saying "My php.ini is OK." you mean "Your extension_dir is set to
c:\PHP"?


Previous Comments:


[2002-01-25 17:23:24] [EMAIL PROTECTED]

Hi, 
   when I try to declare any .dll extension in php.ini on windows, PHP
give the message. 

Unable to load dynamic library 'c:\PHP/php_gd.dll' - The specified
procedure could not be found. in Unknown on line 0 

php_gd.dll is really in c:\PHP dir.
My php.ini is OK.

The same problems appears on 2 of my win2k machines.
Many others persons have the same problem without answers.
Maybe it's another destruction from the virus nimda.

Thanks

Jordi








Edit this bug report at http://bugs.php.net/?id=15229&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #15244 Updated: array_splice acts like array_splice

[jan]

ID: 15244
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Feedback
Bug Type: Arrays related
Operating System: Linux
PHP Version: 4.0.5
New Comment:

can you please provide the $array you are using?



Previous Comments:


[2002-01-27 09:53:41] [EMAIL PROTECTED]

both functions give back the same result:
array_slice($array,1,5);
is eaqual to:
array_splice($array,1,5);

but my phpDocumentation tells me that there should be a "small"
diverence





Edit this bug report at http://bugs.php.net/?id=15244&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #15244 Updated: array_splice acts like array_splice

[jan]

ID: 15244
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Analyzed
Old Bug Type: Arrays related
Bug Type: Documentation problem
Operating System: Linux
PHP Version: 4.0.5
Old Assigned To: 
Assigned To: jan
New Comment:

array_slice is used to extract values from an existing array without
modifying the orinial array. array_splice really extracts the array
elements and so modifies the array. however both return an array
containing the extracted elements, which _have_to_be_ equal. I call thi
a documentation problem, assigning to myself. thank you for reporting.


Previous Comments:


[2002-01-27 11:12:31] [EMAIL PROTECTED]

it's like this one :
$array = array( [0] => firstentry,[1] => secondtentry,[2] =>
thirdentry,[...etc...]);



[2002-01-27 09:57:17] [EMAIL PROTECTED]

can you please provide the $array you are using?




[2002-01-27 09:53:41] [EMAIL PROTECTED]

both functions give back the same result:
array_slice($array,1,5);
is eaqual to:
array_splice($array,1,5);

but my phpDocumentation tells me that there should be a "small"
diverence





Edit this bug report at http://bugs.php.net/?id=15244&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #15244 Updated: array_splice acts like array_splice

[jan]

ID: 15244
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Analyzed
Status: Closed
Bug Type: Documentation problem
Operating System: Linux
PHP Version: 4.0.5
Assigned To: jan
New Comment:

fixed in CVS. closing


Previous Comments:


[2002-01-27 11:23:18] [EMAIL PROTECTED]

array_slice is used to extract values from an existing array without
modifying the orinial array. array_splice really extracts the array
elements and so modifies the array. however both return an array
containing the extracted elements, which _have_to_be_ equal. I call thi
a documentation problem, assigning to myself. thank you for reporting.



[2002-01-27 11:12:31] [EMAIL PROTECTED]

it's like this one :
$array = array( [0] => firstentry,[1] => secondtentry,[2] =>
thirdentry,[...etc...]);



[2002-01-27 09:57:17] [EMAIL PROTECTED]

can you please provide the $array you are using?




[2002-01-27 09:53:41] [EMAIL PROTECTED]

both functions give back the same result:
array_slice($array,1,5);
is eaqual to:
array_splice($array,1,5);

but my phpDocumentation tells me that there should be a "small"
diverence





Edit this bug report at http://bugs.php.net/?id=15244&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #15260 Updated: file_exists doesn't work with 2G+ files

[jan]

ID: 15260
Updated by: jan
Old Summary: file_exists doesn't work with 2G+ files
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Feedback
Bug Type: *Directory/Filesystem functions
Operating System: Linux
PHP Version: 4.0.5
New Comment:

can you please test either PHP 4.1.1 (latest release) or a cvs snapshot
(snaps.php.net) ?


Previous Comments:


[2002-01-28 14:57:17] [EMAIL PROTECTED]

I have a 4G file in /opt/incoming called /opt/incoming/bigpackage.pck

I used 

if(file_exists("/opt/incoming/bigpackage.pck"))
echo "Exists";
else
echo "Not exists";

and it always returns me "Not exists".

I suspect the file_exists function has problem with 2G+ files.







Edit this bug report at http://bugs.php.net/?id=15260&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #15260 Updated: file_exists doesn't work with 2G+ files

[jan]

ID: 15260
Updated by: jan
Old Summary: file_exists doesn't work with 2G+ files
Reported By: [EMAIL PROTECTED]
Old Status: Bogus
Status: Feedback
Bug Type: *Directory/Filesystem functions
Operating System: Linux
PHP Version: 4.0.5
New Comment:

how can he have a file of 4GB if his filesystem does not support it?


Previous Comments:


[2002-01-28 15:00:17] [EMAIL PROTECTED]

This is a limitation of the filesystem (and linux
kernel without LFS support), not php.



[2002-01-28 14:59:43] [EMAIL PROTECTED]

can you please test either PHP 4.1.1 (latest release) or a cvs snapshot
(snaps.php.net) ?



[2002-01-28 14:57:17] [EMAIL PROTECTED]

I have a 4G file in /opt/incoming called /opt/incoming/bigpackage.pck

I used 

if(file_exists("/opt/incoming/bigpackage.pck"))
echo "Exists";
else
echo "Not exists";

and it always returns me "Not exists".

I suspect the file_exists function has problem with 2G+ files.







Edit this bug report at http://bugs.php.net/?id=15260&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #15260 Updated: file_exists doesn't work with 2G+ files

[jan]

ID: 15260
Updated by: jan
Old Summary: file_exists doesn't work with 2G+ files
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Feedback
Bug Type: *Directory/Filesystem functions
Operating System: Linux
PHP Version: 4.0.5
New Comment:

and then again, can you please test with a more recent PHP release?


Previous Comments:


[2002-01-28 15:15:54] [EMAIL PROTECTED]

Where can I (and how do I) get the source code to file_exists function?



[2002-01-28 15:15:00] [EMAIL PROTECTED]

Actually, I do have a 4Gig file on the file system. (Since we used the
64bit file operation for our software on all our file  I/O) 

[root@ECTWO root] ls -la /opt/incoming
total 4399256
drwxrwxr-x 2   4096 Jan 28 13:35
drwxr-xr-x 5 root root 4096 Jan 28 10:44
-rw-rw-r-- 1   4499238984 May 10 2001 BigPackage.fpk.pck



[2002-01-28 15:02:33] [EMAIL PROTECTED]

how can he have a file of 4GB if his filesystem does not support it?



[2002-01-28 15:00:17] [EMAIL PROTECTED]

This is a limitation of the filesystem (and linux
kernel without LFS support), not php.



[2002-01-28 14:59:43] [EMAIL PROTECTED]

can you please test either PHP 4.1.1 (latest release) or a cvs snapshot
(snaps.php.net) ?



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/?id=15260


Edit this bug report at http://bugs.php.net/?id=15260&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

Re: [PHP-DEV] PECL/PHPDoc

[jan]

On Wed, Jan 30, 2002 at 09:07:57AM +0100, Sebastian Bergmann wrote:
>   'lo,
> 
>   could someone please fix, if possible, the PHPDoc extension from PECL
>   so that it compiles in ZTS mode?

Can you provide some error messages ?
 
>   Thanks,
> Sebastian

  Jan

-- 
mailto: [EMAIL PROTECTED] weigon @ #php.de (IRCnet)
 http://jan.kneschke.de weigon @ #modlogan (openprojects)

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #15336 Updated: New mysql function: mysql_real_connect

[jan]

ID: 15336
Updated by: [EMAIL PROTECTED]
Reported By: [EMAIL PROTECTED]
Status: Open
Bug Type: Feature/Change Request
Operating System: any
PHP Version: 4.1.1
Old Assigned To: 
Assigned To: zak
New Comment:

assigning it to zak ;)


Previous Comments:


[2002-02-02 03:57:21] [EMAIL PROTECTED]

Hey Georg,

I will be back to PHP stuff in a few days and will be working on the
MySQL extension. There should be support for selecting the database on
mysql_connect, and may be support for mysql_options. There is a
proposal in the Dev list archives. I don't currently have a copy to
send.



[2002-02-01 18:47:14] [EMAIL PROTECTED]

Since version 3.22.x we have a mysql_real_connect instead 
of mysql_connect + mysql_db_select in the mysql C-API.

If you come from C, its annoying, to use the (deprecated) 
functions mysql_connect + mysql_select_db.

Its ok for BC, but mysql_real_connect should be art of the 
state.

Georg






Edit this bug report at http://bugs.php.net/?id=15336&edit=1


-- 
PHP Development Mailing List 
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #15359 Updated: php crashes if temp dir does not exist

[jan]

 ID:   15359
 Updated by:   [EMAIL PROTECTED]
 Reported By:  [EMAIL PROTECTED]
 Status:   Open
 Bug Type: Reproducible crash
 Operating System: winxp pro
 PHP Version:  4.1.1
 New Comment:

Yasuo Ohgaki fixed this in CVS some days ago


Previous Comments:


[2002-02-03 16:17:30] [EMAIL PROTECTED]

Oh.. The apache error log says:
Premature end of script headers.

Not a very good error message imo.



[2002-02-03 16:16:29] [EMAIL PROTECTED]

When the directory for temporary files (sessions) set in php.ini does
not exist, PHP crashes.

This is on 4.1.1 WindowsXP Pro with the CGI version.




-- 
Edit this bug report at http://bugs.php.net/?id=15359&edit=1


-- 
PHP Development Mailing List 
To unsubscribe, visit: http://www.php.net/unsub.php

[PHP-DEV] Bug #15392 Updated: configration

[jan]

 ID:   15392
 Updated by:   [EMAIL PROTECTED]
 Reported By:  [EMAIL PROTECTED]
 Status:   Bogus
 Bug Type: *Configuration Issues
 Operating System: window98
 PHP Version:  4.1.1
 New Comment:

The bug system is not the appropriate forum for asking support
questions. For a list of a range of more appropriate places to ask
for help using PHP, please visit http://www.php.net/support.php


Previous Comments:


[2002-02-05 15:58:55] [EMAIL PROTECTED]

Please go to http://www.php.net/support.php for support;
look for the mailing list '[EMAIL PROTECTED]' and
sign up. The bug database is not a support forum.


Torben



[2002-02-05 15:57:53] [EMAIL PROTECTED]

The bug system is not the appropriate forum for asking support
questions. For a list of a range of more appropriate places to ask
for help using PHP, please visit http://www.php.net/support.php



[2002-02-05 15:53:26] [EMAIL PROTECTED]

dear sir,
i am ASP programmer and now starting PHP.I have installed window98 and
server is PWS.
how can i config my system for PHP and PWS  on window98.
please send me the slotion and program  files if necessary.
bye
amir
address 672 shad bagh lahore 54900 pakistan




-- 
Edit this bug report at http://bugs.php.net/?id=15392&edit=1


-- 
PHP Development Mailing List 
To unsubscribe, visit: http://www.php.net/unsub.php

[PHP-DEV] Bug #15407 Updated: The " == " op does not work

[jan]

 ID:   15407
 Updated by:   [EMAIL PROTECTED]
-Summary:  The " == " op does not work
 Reported By:  [EMAIL PROTECTED]
-Status:   Open
+Status:   Bogus
 Bug Type: Variables related
 Operating System: Windows 2000
 PHP Version:  4.1.1
 New Comment:

= is no comparison operator but an assignment operator. since a $var =
"bla" is not very error prone of course it "works" but that is
logically not what you want. additionally the == operator on arrays is
not veryhelpful. I suggest, you read the manual sections about
"Assignments", "Comparision", "Arrays" and "Array Functions". No bug ->
bogus 


Previous Comments:


[2002-02-06 16:00:13] [EMAIL PROTECTED]

I'm using Apache and IIS 5.

When I use the comparison operator ( == ), my code fails. But when I
use a single ( = ), it goes fine. I'm using the plain config that comes
with Apache and PHP, but added a line to have .incs treated in the
server.

This is the lame code:

if ($myrow == mysql_fetch_array($Recset))
{
  $gIDClient = $HTTP_POST_VARS["IDClient"];
}

The assigment never gets touched and no error messages are issued. 
Weird, ain't it?
Thanks.




-- 
Edit this bug report at http://bugs.php.net/?id=15407&edit=1


-- 
PHP Development Mailing List 
To unsubscribe, visit: http://www.php.net/unsub.php

[PHP-DEV] Bug #10307 Updated: PHP refusing to get temporary file name from an HTTP PUT request.

[jan]

ID: 10307
Updated by: jan
Reported By: [EMAIL PROTECTED]
Status: Open
Bug Type: HTTP related
Operating system: 
PHP Version: 4.0.4pl1
Assigned To: 
Comments:

perhaps you can add a sample script for other to reproduce the misbehaviour.

Previous Comments:
---

[2001-04-12 13:17:57] [EMAIL PROTECTED]
I am using PHP 4.04pl1 on a FreeBSD 4.2 based system with Apache 1.3.19. I have a page 
that is "accepting" HTTP PUT requests and is supposed to save the file in a designated 
area, however PHP is not picking up the temporary filename while the script runs. All 
other variables required are set, but not $PHP_PUT_FILENAME. I have tried all 
recommendations in the docs, mailing list and IRC channels. None have worked. I do 
notice that the protocol level is HTTP 1.0 PUT not HTTP 1.1 PUT, I have not been able 
to determine if that is the issue.

---



ATTENTION! Do NOT reply to this email!
To reply, use the web interface found at http://bugs.php.net/?id=10307&edit=2


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #11835 Updated: Include and Require does not work inner a class definition.

[jan]

ID: 11835
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old-Status: Open
Status: Closed
Bug Type: Scripting Engine problem
Operating system: 
PHP Version: 4.0.4pl1
Assigned To: 
Comments:

include is not allowed there, use the constructor instead
class {
function class ()   {
include('2.php'); //same with require
}
}



Previous Comments:
---

[2001-07-02 09:14:57] [EMAIL PROTECTED]

Require nor Include works inner a class file.

This is the file where I define the class:
= 1.php =

=  end  =

And this is the file where I have content.
= 2.php =
a++;
}
?>
=  end  =

This does not work.
The error is:

Parse error:  parse error, expecting `T_OLD_FUNCTION' or `T_FUNCTION' or `T_VAR' or 
`'}'' 


---



ATTENTION! Do NOT reply to this email!
To reply, use the web interface found at http://bugs.php.net/?id=11835&edit=2


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #12313: parent resolving and deep require's

[jan]

From: [EMAIL PROTECTED]
Operating system: Linux
PHP version:  4.0.6
PHP Bug Type: Scripting Engine problem
Bug description:  parent resolving and deep require's

PHP seems to have a problem if it tries to resolve the 
parent of an object that is place in an external file 
which is include over multiple inclusions.

http://jan.kneschke.de/projects/phpbugs/

is telling the whole story including demo code.

-- 
Edit bug report at: http://bugs.php.net/?id=12313&edit=1


-- 
PHP Development Mailing List 
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] PHP 4.0 Bug #10176: configurable cgi parameter separators

[jan]

From: [EMAIL PROTECTED]
Operating system: Any
PHP version:  4.0.4pl1
PHP Bug Type: Feature/Change Request
Bug description:  configurable cgi parameter separators

It would be nice to have configurable cgi parameter separators for http get requests, 
because the commonly used & sign doesn't fullfill the HTML specification.

read more on this at: http://validator.w3.org/docs/errors.html#bad-entity

It would be nice if this could be configured on a directory/server/virtual server 
basis in .htaccess or httpd.conf files maybe in the following way:

php_value cgi_get_separator ;

for ; as an example


-- 
Edit Bug report at: http://bugs.php.net/?id=10176&edit=1



-- 
PHP Development Mailing List 
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #14018: make ip2long and long2ip handle unsigned longs correctly

[jan]

From: [EMAIL PROTECTED]
Operating system: Linux
PHP version:  4.0.6
PHP Bug Type: Network related
Bug description:  make ip2long and long2ip handle unsigned longs correctly

Actually the my fix for this problem is on the PHP side, 
not in the PHP-code it self:


function jk_ip2long ($ip) {

preg_match("#^([0-9]{1,3})\.([0-9]{1,3})\.([0-9]{1,3})\.([0-9]{1,3})$#", 
$ip, $a);

$long = 
  $a[1] << 24 | 
  $a[2] << 16 |
  $a[3] <<  8 |
  $a[4] <<  0;

## handling negative urls
if($long < 0) $long += pow(2,32); 

return $long;
}

## long2ip

function jk_long2ip ($long) {
## handle long value which should be signed in PHP but 
are unsigned

if ($long > 0x7fff) {
## although this code seems to be useless, it does 
what we want
$l = ($long - 0x7fff) - 1;
$long = -(0x7fff - $l) - 1;
}


$ip3 = sprintf("%d.%d.%d.%d",
   $long >> 24 & 0xff,
   $long >> 16 & 0xff,
   $long >>  8 & 0xff,
   $long >>  0 & 0xff
   );
return $ip3;
}

You can see a working example at

http://jan.kneschke.de/projects/phpbugs/ip2long.php


-- 
Edit bug report at: http://bugs.php.net/?id=14018&edit=1


-- 
PHP Development Mailing List 
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #14160 Updated: associative arrays - cannot use exclamation mark in key!

[jan]

ID: 14160
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Feedback
Status: Open
Bug Type: Arrays related
Operating System: Windows 2000
PHP Version: 4.0.6
New Comment:

Can't reproduce this on FreeBSD 4.2-Stable and PHP 4.1.0RC3 as cgi or RC2 as mod_php. 
maybe a windows thingie ?


Previous Comments:


[2001-11-21 10:22:24] [EMAIL PROTECTED]

Sorry, of course I meant

http://phpuk.org/~james/php-4.1.0RC3-win32.zip





[2001-11-21 10:20:43] [EMAIL PROTECTED]

Works for me:
[chroot] mfischer@ficken:~/src/php/bugtest$ php

X-Powered-By: PHP/4.0.6-dev
Content-type: text/html

¡Regístrate ya gratis![chroot] mfischer@ficken:~/src/php/bugtest$ 

Hmm probably dumb question, are you really running 4.0.6?

Can you try latest RC

http://www.php.net/~zeev/php-4.1.0RC3.tar.gz

Feedback.



[2001-11-21 10:15:34] [EMAIL PROTECTED]

$translations=Array();
$translations["Register Now for Free!"]="¡Regístrate ya gratis!";
echo $translations["Register Now for Free!"];


will print nothing, or an error about undefined index, depending on your error 
settings in php.ini. As I said before, removing the exclamation mark in the key fixes 
the problem.



[2001-11-21 10:08:25] [EMAIL PROTECTED]

Works for me on 4.0.6 , Linux.

Please give the shortest reproduceable script.

Feedback.



[2001-11-21 10:04:02] [EMAIL PROTECTED]

if you have an array where the key is a string like
$arrayname["a string here!"]. You will not be able to pull the value for that key from 
the array. Removing the exclamation mark in the key fixes this bug.





Edit this bug report at http://bugs.php.net/?id=14160&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Re: [PEAR-DEV] PHPDoc Development Status

[jan]

On Thu, Nov 22, 2001 at 04:06:35PM +0100, Andre Gildemeister wrote:
> hi,
> 
> I have pursued the development of PHPDoc very interestedly.
> After the incorporation into the CVS Repository: php4/pear/PHPDoc I have
> heard nothing more new...
> 
> - in the meantime was something to heard about a rewrite in C, is that true?

I've written a PHP-2-XML converter which is sitting directly on the
Zend-parser. This gives us more speed and more accuracy.

It has been written because Ulf Wendel convinced me to do it by delivering a
few pizzas. 

If someone wants to test it, download it from http://weigon.dyndns.org/

It is a php-extension which requires some patches to the Zend engine. If
someone tests it, don't give up to early. The ext has been written for
4.0.4pl1 or the like and won't compile with the latest php as the Zend core
has changed.

Currently missing is the PHPDoc part (the php code) which can handle XML.

> - shouldn't PHPDoc become an official extension either? (ext/phpdoc)

phpdoc itself or just this extension ?

> comments etc. welcome.
> 
> -Andre Gildemeister

  Jan

-- 
mailto: [EMAIL PROTECTED] weigon @ #php.de (IRCnet)
 http://jan.kneschke.de weigon @ #modlogan (openprojects)

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Re: [PEAR-DEV] PHPDoc Development Status

[jan]

On Fri, Nov 23, 2001 at 09:19:15AM +0100, Ol wrote:
> On Thu, 22 Nov 2001 22:17:18 +0100
> [EMAIL PROTECTED] wrote:
> 
> > I've written a PHP-2-XML converter which is sitting directly on the
> > Zend-parser. This gives us more speed and more accuracy.
> 
> Difficult to do better in fact ... ;)

Right.

> > If someone wants to test it, download it from http://weigon.dyndns.org/
> 
> Yeap it works fine :)
> Some minor bugfixes i've find tonight (diff -u at bottom)

Fine.

> > It is a php-extension which requires some patches to the Zend engine. If
> > someone tests it, don't give up to early. The ext has been written for
> > 4.0.4pl1 or the like and won't compile with the latest php as the Zend
> > core has changed.
> 
> At least it still work with 4.0.6

Very good.

> > Currently missing is the PHPDoc part (the php code) which can handle
> > XML.
> 
> What to you thing about a double output for this code :
> - on a side can generate standart HTML output like 'classical' phpdoc
> - on the other a docbook output.

Why not. Perfect. 

PHP -> HTML (highlight_string)
PHP -> XML  (ext/phpdoc)
PHP -> SGML (ext/phpdoc + sgml)

> It could be a good way to obtain more easily pear's enduser doc. Even if
> the docbook output surely need corrections it could be anyway a nice
> starting point... 

Sure. Doing the SGML in PHP based on the XML could simplify the development
a lot. Perhaps XSLT can be usefull in these cases.

PHP -> XML -> [XSLT] -> HTML
PHP -> XML -> [XSLT] -> DocBook
PHP -> XML -> PHPDoc

This would be nice.

> Olivier Courtin

  Jan

-- 
mailto: [EMAIL PROTECTED] weigon @ #php.de (IRCnet)
 http://jan.kneschke.de weigon @ #modlogan (openprojects)

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

Re: [PHP-DEV] Re: [PEAR-DEV] PHPDoc Development Status

[jan]

On Sun, Nov 25, 2001 at 02:16:38AM +0100, Stig S. Bakken wrote:
> Just this extension.  Hurry up and apply for a CVS account at
> http://php.net/cvs-php.php :-)

done.

>  - Stig

  Jan

-- 
mailto: [EMAIL PROTECTED] weigon @ #php.de (IRCnet)
 http://jan.kneschke.de weigon @ #modlogan (openprojects)

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #14282: wrong strtotime calculation with timezones -0100 and +0100

[jan]

From: [EMAIL PROTECTED]
Operating system: any
PHP version:  4.1.0
PHP Bug Type: Date/time related
Bug description:  wrong strtotime calculation with timezones -0100 and +0100

This code:


produces this output:
Thu, 29 Nov 2001 15:24:21 +0100
Sun, 25 Nov 2001 12:24:21 +0100

This is the fix:
--- parsedate.y~Thu Nov 15 15:33:41 2001
+++ parsedate.y Thu Nov 29 15:28:52 2001
@@ -255,7 +255,7 @@
yySeconds = $5;
yyMeridian = MER24;
yyHaveZone++;
-   if ($6 < -100 || $6 > 100) {
+   if ($6 <= -100 || $6 >= 100) {
yyTimezone =  -$6 % 100 + (-$6 / 100) * 60;
} else {
yyTimezone =  -$6 * 60;
-- 
Edit bug report at: http://bugs.php.net/?id=14282&edit=1


-- 
PHP Development Mailing List 
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #14364 Updated: Unable to clear out an individual variable in a session

[jan]

ID: 14364
Updated by: jan
Reported By: [EMAIL PROTECTED]
Status: Open
Bug Type: Session related
Operating System: FreeBSD 4.2
PHP Version: 4.1.0
New Comment:

Can you provide the shortest sample script, that reproduce that bug?

Previous Comments:


[2001-12-06 14:40:30] [EMAIL PROTECTED]

There appears to be an issue with unsetting variables in a session. I have an object 
variable that is stored in the session and when I attempt to unset the value via 
unset(object) and session_unregister(object) and test if(isset(object)) to see if it 
is set in a seperate script the test passes.  In PHP 4.05 I did not encounter this 
problem.





Edit this bug report at http://bugs.php.net/?id=14364&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

Re: [PHP-DEV] Re: [PHP-DOC] cvs: php4 /ext/phpdoc .cvsignore Makefile.in config.m4 libs.mk php_phpdoc.h phpdoc.c

[jan]

On Wed, Dec 12, 2001 at 11:51:00AM +0100, Sterling Hughes wrote:
> > Jan Kneschke wrote:
> > > ostborn Wed Dec 12 04:46:27 2001 EDT
> > >
> > >   Added files:
> > > /php4/ext/phpdoc.cvsignore Makefile.in config.m4 libs.mk
> > >     php_phpdoc.h phpdoc.c
> > 
> >   Some questions, Jan:
> > 
> > 1.) Why /php4/ext/ and not /pear/PECL/?
> >
> 
> I think this belongs in PECL personally...

I was asked to commit it under ext/phpdoc if a read the mails about 'the
future of PHPDoc'  correctly. 
 
> > 2.) Does this version compile with current CVS? I haven't tried yet,
> > but I see an old module entry declaration.
> > 
> then it doesn't compile...

Any pointer for me what has to be changed ?

> > 3.) What's the relation between PHPDoc and IPv6? :-)
> > (php_info_print_table_row(2, "PHPDoc Version", "IPv6" );)
> > 
> 
> Its just to confuse newbies??? :)

Yes :) 

> >   Anyhow: Thanks for your work and I'm happy seeing this finally on
> >   cvs.php.net!
> > 
> Agreed, just where is up to debate :)

Debate on that and move it around if you want. 

BTW: What is PECL ?

> -Sterling

  Jan

-- 
mailto: [EMAIL PROTECTED] weigon @ #php.de (IRCnet)
 http://jan.kneschke.de weigon @ #modlogan (openprojects)

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

Re: [PHP-DEV] Re: [PHP-DOC] cvs: php4 /ext/phpdoc .cvsignore Makefile.in config.m4 libs.mk php_phpdoc.h phpdoc.c

[jan]

On Wed, Dec 12, 2001 at 01:08:36PM +0100, Sterling Hughes wrote:
> > I was asked to commit it under ext/phpdoc if a read the mails about 'the
> > future of PHPDoc'  correctly. 
> >  
> 
> Didn't read them...  It doesn't make sense in ext/phpdoc however
> (imho), as:
> 
> a) its  not something that is widely used

ext/phpdoc + ext/xslt => printing colourfull PHP, highlighht_string (HTML),
...

> b) it has a very specific, local use when it is used (ie, it doesn't
> help anyone *build* a site, just generate documentation for their
> code).

It depends on what is will be used for. The XML output is not directly
connected to phpdoc.

> c) it requires a pear package to be useful anyway...

currently, yes.

In my opinion it should be moved to PECL as all extensions should live
there. Not just ext/phpdoc.

> > > > 2.) Does this version compile with current CVS? I haven't tried yet,
> > > > but I see an old module entry declaration.
> > > > 
> > > then it doesn't compile...
> > 
> > Any pointer for me what has to be changed ?
> 
> check out README.EXTENSIONS...

Will read them.
 
> > BTW: What is PECL ?
> > 
> Its for pear C extensions and the like

Fine.
 
> -Sterling

  Jan

-- 
mailto: [EMAIL PROTECTED] weigon @ #php.de (IRCnet)
 http://jan.kneschke.de weigon @ #modlogan (openprojects)

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #11719 Updated: Clarification of $foo = new Class and $foo = &new Class

[jan]

ID: 11719
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Closed
Bug Type: Documentation problem
Operating System: Any
PHP Version: 4.0 Latest CVS (2001-06-27)
New Comment:

fixed in CVS.

Previous Comments:


[2001-06-27 01:06:40] [EMAIL PROTECTED]

The "References" section of the documentation needs more clarification of the 
difference between:

$foo = new Class;
and
$foo = &new Class;

and when or if each should be used.






Edit this bug report at http://bugs.php.net/?id=11719&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #11937 Updated: Clarification of $foo = new Class and $foo = &new Class

[jan]

ID: 11937
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Summary: Bug in array handling
Old Status: Assigned
Status: Closed
Bug Type: Documentation problem
Old Operating System: Windows 98
Operating System: Any
Old PHP Version: 4.0.5
PHP Version: 4.0 Latest CVS (2001-06-27)
Old Assigned To: jeroen
Assigned To: 
New Comment:

fixed in CVS. 

Previous Comments:


[2001-07-07 09:10:37] [EMAIL PROTECTED]

Should be more clear in the docs. It is only implicitely there.

Assigning to myself.



[2001-07-06 15:20:01] [EMAIL PROTECTED]

If you unset an element in the array, it does not make the whole array shift so it 
starts from index zero again. That's normal behaviour.



[2001-07-06 14:52:11] [EMAIL PROTECTED]

";
//The following dump will produce:
//array(1) { [1]=> string(4) "Lisa" } 
//This means that the first element has the index 1 
//But it should have index 0
var_dump ($array);
?>

If you have more than two elements, and you free one of them, you can just prevent 
this bug, by resorting the array with any sort function. But with only two elements, 
this won't work, and if you clear the first of these, you will get the same result as 
I have got, there was one element,  but with the wrong index.





Edit this bug report at http://bugs.php.net/?id=11937&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #12863 Updated: What does pg_freeresult return?

[jan]

ID: 12863
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Summary: what does pg_freeresult return?
Old Status: Analyzed
Status: Closed
Bug Type: Documentation problem
Old Operating System: 
Operating System: Any
Old PHP Version: 4.0.6
PHP Version: 4.0 Latest CVS (2001-06-27)
New Comment:

fixed in CVS

Previous Comments:


[2001-10-19 08:21:58] [EMAIL PROTECTED]

I think (according to the source) it returns TRUE on success, FALSE on error.

Can someone confirm this, and add it to the docs?



[2001-08-20 12:54:06] [EMAIL PROTECTED]

The doc page for pg_freeresult does not state what the significance of the int return 
value is.





Edit this bug report at http://bugs.php.net/?id=12863&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #13911 Updated: Documentation on ini_set could use some reorganizing

[jan]

ID: 13911
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Summary: Documentation on ini_set could use some reorganizing
Status: Open
Bug Type: Documentation problem
Operating System: Any
PHP Version: 4.0.6
New Comment:

they are sortet in the way phpinfo() shows them. If we change it here IMHO phpinfo 
needs to be changed as well. what about reorganizing the phpinfo()output to be sorted 
alphabetically.

Previous Comments:


[2001-11-02 16:44:05] [EMAIL PROTECTED]

The table that describes the configuration options on 
http://www.php.net/manual/en/function.ini-set.php suffers from a small but annoying 
usability problem: the options aren't sorted in any logical order.  It would maybe 
make sense to have the configuraion options listed in alphabetical order to make it 
easier to scan the list in order to find a certain option.

Some nice source code formatting to the pages would be nice, too.  Check out the code 
for the table that lists the options and you'll see what I mean.

If anyone else doesn't want to take the time to do it, I could fix the table myself 
and submit it.





Edit this bug report at http://bugs.php.net/?id=13911&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #13967 Updated: unset() on a static variable causes it to forget it is static

[jan]

ID: 13967
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Closed
Bug Type: Documentation problem
Operating System: Debian Linux
PHP Version: 4.0.6
New Comment:

fixed in CVS

Previous Comments:


[2001-11-07 18:29:43] [EMAIL PROTECTED]

Like with global, static is a keyword linking a variable specially. unset() will 
completely destroy the variable with all links it might have. This is expected 
behaviour, or at least a known side-effect. With the current way of handling variables 
internally, this is not going to change.

This isn't stressen enough in the manual, so changing to documentation problem.



[2001-11-06 21:49:10] [EMAIL PROTECTED]

Doing unset() on a static variable causes it to forget the fact that it is static for 
the rest of the function.

e.g.



should give 123 then 456. It gives 123 and 123.

Replacing unset($a) with $a = null, or removing the unset() entirely has the desired 
result of giving 123 then 456.





Edit this bug report at http://bugs.php.net/?id=13967&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] location of "ext/phpdoc"

[jan]

Can someone please make decision where the phpdoc extension should be
located at ?

I just wanted to apply some patches to make it work with the latest php and
failed I have only karma for php4/ext/phpdoc which has been moved to
pear/PECL/phpdoc.

But I don't have enough karma for pear/PECL/phpdoc.

  Jan

-- 
mailto: [EMAIL PROTECTED] weigon @ #php.de (IRCnet)
 http://jan.kneschke.de weigon @ #modlogan (openprojects)

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #14591: uasort mixes array with equal elements

[jan]

From: [EMAIL PROTECTED]
Operating system: Linux/Windows
PHP version:  4.1.0
PHP Bug Type: Arrays related
Bug description:  uasort mixes array with equal elements

If you call uasort with a function that always returns 0 (elements equal)
you get a totally screwed array returned.



returns:

array(6) {
  [0]=>
  string(1) "h"
  [1]=>
  string(1) "s"
  [2]=>
  string(1) "i"
  [3]=>
  string(1) "c"
  [4]=>
  string(1) "q"
  [5]=>
  string(1) "m"
}
array(6) {
  [1]=>
  string(1) "s"
  [2]=>
  string(1) "i"
  [3]=>
  string(1) "c"
  [4]=>
  string(1) "q"
  [5]=>
  string(1) "m"
  [0]=>
  string(1) "h"
}

-- 
Edit bug report at: http://bugs.php.net/?id=14591&edit=1


-- 
PHP Development Mailing List 
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #14591 Updated: uasort mixes array with equal elements

[jan]

ID: 14591
User updated by: [EMAIL PROTECTED]
Reported By: [EMAIL PROTECTED]
Status: Bogus
Bug Type: Arrays related
Operating System: Linux/Windows
PHP Version: 4.1.0
New Comment:

Hm, I'd rather expect that an array keeps untouched if there is no need to sort it.

Previous Comments:


[2001-12-18 19:24:23] [EMAIL PROTECTED]

and the problem is ...?

by returning all zeros you made perfectly clear
that you don't care about the order of elements
in your array ;)





[2001-12-18 17:03:19] [EMAIL PROTECTED]

If you call uasort with a function that always returns 0 (elements equal) you get a 
totally screwed array returned.



returns:

array(6) {
  [0]=>
  string(1) "h"
  [1]=>
  string(1) "s"
  [2]=>
  string(1) "i"
  [3]=>
  string(1) "c"
  [4]=>
  string(1) "q"
  [5]=>
  string(1) "m"
}
array(6) {
  [1]=>
  string(1) "s"
  [2]=>
  string(1) "i"
  [3]=>
  string(1) "c"
  [4]=>
  string(1) "q"
  [5]=>
  string(1) "m"
  [0]=>
  string(1) "h"
}






Edit this bug report at http://bugs.php.net/?id=14591&edit=1


-- 
PHP Development Mailing List 
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #14604 Updated: Empty packets as result of serialization

[jan]

ID: 14604
Updated by: jan
Reported By: [EMAIL PROTECTED]
Status: Open
Bug Type: WDDX related
Operating System: Windows NT 4 sp 6a
PHP Version: 4.1.0
New Comment:

at least for me your code works in the same way under 4.0.6 and 4.1.0 on FreeBSD 
4.4-STABLE (yes, I finally updated ;):

bash-2.05a$ ./php  test.php 
X-Powered-By: PHP/4.0.6
Content-type: text/html

bash-2.05a$ 

bash-2.05a$ ./php  test.php 
X-Powered-By: PHP/4.1.0
Content-type: text/html

bash-2.05a$ 


bash-2.05a$ cat test.php



Previous Comments:


[2001-12-19 09:48:07] [EMAIL PROTECTED]

$nazwisko = "Zgoda";
$imie = "Jarek";
$tmp = wddx_packet_start();
wddx_add_vars($tmp, $nazwisko, $imie);
$rs = wddx_packet_end($tmp);
echo $rs;

This code produces empty packet (just structure definition - no data) in 4.1.0, while 
it worked in 4.0.6





Edit this bug report at http://bugs.php.net/?id=14604&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #14605 Updated: compare string with integer 0 returns t

[jan]

ID: 14605
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Summary: compare string with integer 0 returns true
Old Status: Open
Status: Bogus
Bug Type: Scripting Engine problem
Old Operating System: Linux
Operating System: Windows NT 4 sp 6a
PHP Version: 4.1.0
New Comment:

this i due to the autoconversation of variables in PHP. Your string is converted into 
an integer during the comparison.The integer is 0, since no fitting value can be 
found. 
So 0 equals 0.

Previous Comments:


[2001-12-19 11:11:09] [EMAIL PROTECTED]


// This outputs 'the string' == 0





Edit this bug report at http://bugs.php.net/?id=14605&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

Re: [PHP-DEV] Building / using the LDAP extension

[jan]

On Thu, Feb 21, 2002 at 02:14:35AM -0700, Venkat Raghavan wrote:
> Again, it's we folks porting PHP onto NetWare.
> 
> Is it necessary that the LDAP extension has to be a separate binary?

No.

> Can we build it into PHPLIB itself? If so, how do we enable LDAP,
> considering we can't run the CONFIGURE script on NetWare (--with-ldap)?

$ ./configure --with-ldap

enables the check for the ldap-libraries (libldap.so, liblber.so) and the
ldap headers. 

If those libs aren't installed in the default path you can specify them with
the --with-ldap switch:

$ ./configure --with-ldap=/opt/novell/

if they are located in

/opt/novell/lib/
/opt/novell/include/

> Venkat

  Jan

-- 
mailto: [EMAIL PROTECTED] weigon @ #php.de (IRCnet)
 http://jan.kneschke.de weigon @ #modlogan (openprojects)

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, visit: http://www.php.net/unsub.php

Re: [PHP-DEV] Building / using the LDAP extension

[jan]

On Thu, Feb 21, 2002 at 02:42:16AM -0700, Venkat Raghavan wrote:
> Thanx.
> 
> What I am asking is how do I 'enable' LDAP in my PHP binary, provided I
> can't run the CONFIGURE script. Also, what is the "default path" for the
> LDAP libraries?

You can not run the ./configure script of the php source package ? That
means that you can't even compile php ?

There is no real default path for the ldap-libs as they are vendor supplied.

OpenLDAP installs to /usr/lib/libldap.so 
Sun installs to /usr/lib/libldap.so on Solaris
NetScape installs to /usr/local/netscape/ ...
Novell ?

  Jan

-- 
mailto: [EMAIL PROTECTED] weigon @ #php.de (IRCnet)
 http://jan.kneschke.de weigon @ #modlogan (openprojects)

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, visit: http://www.php.net/unsub.php

Re: [PHP-DEV] Building / using the LDAP extension

[jan]

On Thu, Feb 21, 2002 at 03:44:07AM -0700, Venkat Raghavan wrote:
> I am not talking about compiling PHP on an Unix platform. What we are
> doing is porting PHP onto NetWare.
> 
> We have PHP built along with LDAP, since we want it as part of the
> standard package. So, we just want to know how to make PHP 'aware' that
> LDAP functionality is built in, since we're not using the standard
> configure script way. And we don't need to set the path also since that
> is already taken care of in our build process.
>
> I hope this is clear.

Ok, this is a different situation.

Let's summarize:

- PHP is already built successfully on your NetWare

Something that is not clear yet:
- (a) the ldap-extension is compiled into the php-binary itself,
- (b) the ldap-extension is compiled standalone as a seperated binary
- (c) I'm missed to the whole point again.

for (a) use  check if php is aware of it ldap support
for (b) load the extension with dl() and call  afterwards

> - Venkat

  Jan

-- 
mailto: [EMAIL PROTECTED] weigon @ #php.de (IRCnet)
 http://jan.kneschke.de weigon @ #modlogan (openprojects)

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, visit: http://www.php.net/unsub.php

Re: [PHP-DEV] Building / using the LDAP extension

[jan]

On Thu, Feb 21, 2002 at 04:36:13AM -0700, Venkat Raghavan wrote:
> OK, finally, we're on track.

Fine. :)
 
> Out of the options you gave, the correct one is (a). And when I do
> phpinfo(), I don't see LDAP. That's why I asked what else I need to do
> in the code / build process to 'enable' LDAP.

Just make sure that ext/ldap is really compiled and linked into the php
binary.

Please check check scripts in ./build/ if there is some extra magic applied.

> Thanx again,
> Venkat

  Jan

-- 
mailto: [EMAIL PROTECTED] weigon @ #php.de (IRCnet)
 http://jan.kneschke.de weigon @ #modlogan (openprojects)

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, visit: http://www.php.net/unsub.php

[PHP-DEV] Bug #12871: updating build-system to autoconf 2.5x

[jan]

From: [EMAIL PROTECTED]
Operating system: 
PHP version:  4.0CVS-2001-08-21
PHP Bug Type: *Compile Issues
Bug description:  updating build-system to autoconf 2.5x

After updating autoconf to 2.52 I've noticed that php is 
working with this autoconf version yet.

Even 'autoupdate' didn't helped to fix it.

I've taken a look at the generated shell code and noticed 
that some 'if-then-else' had an empty else-tree (else fi) 
which caused several unexpected shell exits. 
  These empty else-trees were the result of the following 
commands:
  AC_CHECK_FUNC(yp_get_default_domain,
 [php_no_nsl_checks=yes],[ ])
 ^^^

That [ ] is the problem. If it is replaced by [] the else 
tree won't be build.

This applies to nearly all macros. The only exception is 
AC_DEFINE which requires a [ ] as third parameter if it is 
set at all.

Another issue was the PHP_SUBST in the acinclude.m4. I 
don't know what the five PHP_SUBST are doing there, but 
they will have to go into the configure.in the satisfy 
autoconf.

(acinclude.m4:930 [somewhere there])

PHP_SUBST(EXT_SUBDIRS)
PHP_SUBST(EXT_STATIC)
PHP_SUBST(EXT_SHARED)
PHP_SUBST(EXT_LIBS)
PHP_SUBST(EXT_LTLIBS)

I've uploaded the corresponding patch to 
http://jan.kneschke.de/projects/phpbugs/php4-20010821.diff

The patch is just a proof of concepts and doesn't fix 
every glitches, but it provides an working example.


-- 
Edit bug report at: http://bugs.php.net/?id=12871&edit=1


-- 
PHP Development Mailing List 
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

Re: [PHP-DEV] Buildproblems in latest CVS

[jan]

On Thu, Aug 09, 2001 at 02:40:19AM +0200, Anders Johannsen wrote:
> I'am experiencing problems building PHP from the latest cvs. The errors
> occur when running 'buildconf'. The box is a Debian/unstable. Autoconf,
> automake and libtool versions seems to suffice.
> 
> aj@uranos% ./buildconf
> buildconf: checking installation...
> buildconf: autoconf version 2.52 (ok)
> buildconf: automake version 1.4-p4 (ok)
> buildconf: libtool version 1.4b

Please check if your auto* combination works with the patch provided in:

[PHP-DEV] Bug #12871: updating build-system to autoconf 2.5x

  Jan

-- 
  => http://www.kneschke.de <==> http://www.random-downloads.com <=
   mailto: [EMAIL PROTECTED]

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

Re: [PHP-DEV] php and autoconf 2.50

[jan]

On Sat, Aug 11, 2001 at 09:36:09AM +0200, Sascha Schumann wrote:
> On Fri, 10 Aug 2001, Daniel Adams wrote:
> 
> > Has anyone got the cvs source to work with autoconf 2.50?
> 
> Noone as far as I know has worked on that so far; anyone who
> is interested in this functionality is encouraged to provide
> patches which addresses the remaining issues.

Done -> [PHP-DEV] Bug #12871: updating build-system to autoconf 2.5x

> - Sascha     Experience IRCG

  Jan

-- 
  => http://www.kneschke.de <==> http://www.random-downloads.com <=
   mailto: [EMAIL PROTECTED]

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #12964: direct access to an array fails but walkthrough by foreach works

[jan]

From: [EMAIL PROTECTED]
Operating system: Linux
PHP version:  4.0.6
PHP Bug Type: Scripting Engine problem
Bug description:  direct access to an array fails but walkthrough by foreach works

It seems that I've reached one of the borders of PHP/Zend.

I'm coding a qt-like environment called pasta which 
results in a quite large class-tree. This is a problem for 
PHP as it looses a some parts of the class:

The code:
print_r($this->sec_global);
print "incpath: 
".$this->sec_global["includepath"]."\n";
foreach($this->sec_global as $k => $v) {
print "$k -- $v";
}

results in the following output:

Array
(
["debug_level"] => 0
["incremental"] => 0
["enable_resolver"] => 1
["outputdir"] =>
["include"] =>
["includepath"] => ../etc/
)
 
 
Warning:  Undefined index:  includepath in 
/pages/projects/modlogan/mlaconfiggen.php on line 262
 
incpath:
"debug_level" -- 0
"incremental" -- 0
"enable_resolver" -- 1
"outputdir" --
"include" --
"includepath" -- ../etc/


As you can see a print_r and foreach display the same 
result. The array is filled with these data like I've 
expected it. But a direct access to the array results in a 
'undefined index' which looks very odd to me.

This problem exists for other class variables of this 
class too:

Undefined property: processorplugin ...

This happens in the default branch of a switch.

The full code can be found at 
http://jan.kneschke.de/showsource.php?page=%2Fprojects%2Fmodlogan%2Fmlaconfiggen.php

The output is at
http://jan.kneschke.de/projects/modlogan/mlaconfiggen.php

(click 3 times on the '-->', fill in some data into the 
4th page and click on '-->' again. This will result in the 
described behaviour which has been reproduced with 
PHP4.0.6 and PHP4.0.7rc1)


-- 
Edit bug report at: http://bugs.php.net/?id=12964&edit=1


-- 
PHP Development Mailing List 
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Re: Bug #12964: direct access to an array fails but walkthrough by foreach works

[jan]

On Sun, Aug 26, 2001 at 10:26:20AM -, PHP Bug Database wrote:
> From: [EMAIL PROTECTED]
> Operating system: Linux
> PHP version:  4.0.6
> PHP Bug Type: Scripting Engine problem
> Bug description:  direct access to an array fails but walkthrough by foreach works

As I havn't set a password for the bugreport I can't close it. 

Please set this bug to 'bogus'. 

The key of the array elements was "includepath" (including the "). :(
The brown paperbag is mine.

> -- 
> Edit bug report at: http://bugs.php.net/?id=12964&edit=1
> 

  Jan

-- 
  => http://www.kneschke.de <==> http://www.random-downloads.com <=
   mailto: [EMAIL PROTECTED]

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #13040 Updated: German Translation of Docs of open_basedir wrong

[jan]

ID: 13040
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Closed
Bug Type: Documentation problem
Operating System: 
PHP Version: 4.0.6
New Comment:

fixed

Previous Comments:


[2001-08-29 13:52:45] [EMAIL PROTECTED]

Limit the files that can be opened by PHP to the specified directory-tree. 

should be translated to:

Beschränkt die Dateien, die von PHP geöffnet werden können, auf Dateien im angegebenen 
Verzeichnisbaum.


The wrong German Translation back-translated would be:

Limits the number of files, that can be opened by PHP in a given directory.






Edit this bug report at http://bugs.php.net/?id=13040&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #13440 Updated: PHP

[jan]

ID: 13440
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Bogus
Bug Type: Scripting Engine problem
Operating System: Windows
PHP Version: 4.0.6
New Comment:

No Bug. Please ask such questions on [EMAIL PROTECTED]

Previous Comments:


[2001-09-25 14:42:06] [EMAIL PROTECTED]

How i make a random banners in PHP ?





Edit this bug report at http://bugs.php.net/?id=13440&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

[PHP-DEV] Bug #13442 Updated: Php

[jan]

ID: 13442
Updated by: jan
Reported By: [EMAIL PROTECTED]
Old Status: Open
Status: Bogus
Bug Type: Scripting Engine problem
Operating System: Windows
PHP Version: 4.0.6
New Comment:

This is no support form. For questions on the development with PHP ask 
[EMAIL PROTECTED] or a support forum at your favour. (www.phpbuilder.com e.g.)



Previous Comments:


[2001-09-25 15:34:55] [EMAIL PROTECTED]

How i make a random banner's system on PHP ?





Edit this bug report at http://bugs.php.net/?id=13442&edit=1


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

Re: [PHP-DEV] Compile error in xml-rpc (HEAD)

[Jan Schneider]

Zitat von Magnus Määttä <[EMAIL PROTECTED]>:

> Hi!
> 
> gcc -I/opt/DEV/php/php4/ext/xmlrpc/libxmlrpc -DVERSION=0.50 -Iext/xmlrpc/
> -I/opt/DEV/php/php4/ext/xmlrpc/ -DPHP_ATOM_INC
> -I/opt/DEV/php/php4/include -I/opt/DEV/php/php4/main -I/opt/DEV/php/php4
> -I/opt/DEV/php/php4/Zend -I/usr/include/libxml2 -I/usr/X11R6/include
> -I/usr/include/freetype2 -I/usr/include/imap -I/usr//include/mysql
> -I/usr/include/pspell -I/usr/qt/3/include -I/usr//include 
> -I/opt/DEV/php/php4/TSRM  -g -Wall  -c
> /opt/DEV/php/php4/ext/xmlrpc/xmlrpc-epi-php.c -o
> ext/xmlrpc/xmlrpc-epi-php.o  && echo > ext/xmlrpc/xmlrpc-epi-php.lo
> /opt/DEV/php/php4/ext/xmlrpc/xmlrpc-epi-php.c: In function
> `set_zval_xmlrpc_type':
> /opt/DEV/php/php4/ext/xmlrpc/xmlrpc-epi-php.c:1348: structure has no
> member named `properties'
> /opt/DEV/php/php4/ext/xmlrpc/xmlrpc-epi-php.c:1349: structure has no
> member named `properties'
> make: *** [ext/xmlrpc/xmlrpc-epi-php.lo] Error 1
> 
> config.nice
> #! /bin/sh
> #
> # Created by configure
> 
> './configure' \
> '--disable-all' \
> '--prefix=/usr/local/www/php4cli' \
> '--enable-debug' \
> '--enable-sysvshm' \
> '--enable-aggregate' \
> '--enable-xslt' \
> '--enable-bcmath' \
> '--enable-calendar' \
> '--enable-dbase' \
> '--enable-dio' \
> '--enable-dbx' \
> '--enable-exif' \
> '--enable-filepro' \
> '--enable-ftp' \
> '--enable-gd-native-ttf' \
> '--enable-sockets' \
> '--enable-yp' \
> '--enable-enable-trans-sid' \
> '--enable-pcntl' \
> '--enable-inline-optimization' \
> '--enable-cli' \
> '--enable-rpath' \
> '--enable-short-tags' \
> '--enable-ipv6' \
> '--enable-ctype' \
> '--enable-overload' \
> '--enable-posix' \
> '--enable-session' \
> '--enable-tokenizer' \
> '--enable-xml' \
> '--enable-libtool-lock' \
> '--with-xslt-sablot=/usr/' \
> '--with-expat-dir=/usr' \
> '--with-xmlrpc' \
> '--with-ming=/usr' \
> '--with-swf=/usr' \
> '--with-dom=/usr' \
> '--with-qtdom' \
> '--with-recode' \
> '--with-mysql=/usr/' \
> '--with-openssl' \
> '--with-zlib-dir=/usr' \
> '--with-jpeg-dir' \
> '--with-tiff-dir' \
> '--with-db' \
> '--with-gdbm' \
> '--with-dom' \
> '--with-dom-xslt' \
> '--with-gd' \
> '--with-pdflib' \
> '--with-png-dir' \
> '--with-xpm-dir' \
> '--with-freetype-dir' \
> '--with-ttf' \
> '--with-t1lib' \
> '--with-gettext' \
> '--with-gmp' \
> '--with-ldap' \
> '--with-mcrypt' \
> '--with-mhash' \
> '--with-pspell' \
> '--with-readline' \
> '--with-bz2' \
> '--with-snmp' \
> '--with-curl' \
> '--with-imap' \
> '--with-imap-ssl' \
> '--with-pcre-regex' \
> '--with-pear' \
> "$@"

I get this error too, compiling latest CVS HEAD with ZE2.

The attached patch fixed it for me.

Jan.

--
http://www.horde.org - The Horde Project
http://www.ammma.de - discover your knowledge
http://www.tip4all.de - Deine private Tippgemeinschaft
Index: ext/xmlrpc/xmlrpc-epi-php.c
===
RCS file: /repository/php4/ext/xmlrpc/xmlrpc-epi-php.c,v
retrieving revision 1.26
diff -u -r1.26 xmlrpc-epi-php.c
--- ext/xmlrpc/xmlrpc-epi-php.c 31 Dec 2002 16:08:09 -  1.26
+++ ext/xmlrpc/xmlrpc-epi-php.c 10 Jan 2003 02:54:18 -
@@ -1345,8 +1345,8 @@
   ztimestamp->value.lval = timestamp;
 
   convert_to_object(value);
-  if(SUCCESS == zend_hash_update(value->value.obj.properties, 
OBJECT_TYPE_ATTR, sizeof(OBJECT_TYPE_ATTR), (void *) &type, sizeof(zval *), NULL)) {
- bSuccess = zend_hash_update(value->value.obj.properties, 
OBJECT_VALUE_TS_ATTR, sizeof(OBJECT_VALUE_TS_ATTR), (void *) &ztimestamp, sizeof(zval 
*), NULL);
+  if(SUCCESS == zend_hash_update(Z_OBJPROP_P(value), 
+OBJECT_TYPE_ATTR, sizeof(OBJECT_TYPE_ATTR), (void *) &type, sizeof(zval *), NULL)) {
+ bSuccess = zend_hash_update(Z_OBJPROP_P(value), 
+OBJECT_VALUE_TS_ATTR, sizeof(OBJECT_VALUE_TS_ATTR), (void *) &ztimestamp, sizeof(zval 
+*), NULL);
   }
}
XMLRPC_CleanupValue(v);

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, visit: http://www.php.net/unsub.php

[PHP-DEV] domxml behaviour changed/broken?

[Jan Schneider]

Hi,

since when can the dump_node() method only be applied to DomDocument nodes
and not to DomNodes as it has always been the case and is still documented
in the manual?

Running PHP 4.3.1-dev cli.

Jan.

--
http://www.horde.org - The Horde Project
http://www.ammma.de - discover your knowledge
http://www.tip4all.de - Deine private Tippgemeinschaft

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, visit: http://www.php.net/unsub.php

Re: [PHP-DEV] Current HEAD segfaults with Horde/CHORA

[Jan Schneider]

Zitat von Derick Rethans <[EMAIL PROTECTED]>:

> Looks like a recursive thing here so I wonder if it's really a bug
> in PHP, but perhaps an assumption in Horde that doesn't work with ZE2?

Correct, Horde doesn't work with ZE2 out of the box.

Jan.

--
http://www.horde.org - The Horde Project
http://www.ammma.de - discover your knowledge
http://www.tip4all.de - Deine private Tippgemeinschaft

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, visit: http://www.php.net/unsub.php

[PHP-DEV] compile error on HEAD

[Jan Schneider]

/home/jan/software/php5/ext/standard/basic_functions.c: In function
`php_simple_ini_parser_cb':
/home/jan/software/php5/ext/standard/basic_functions.c:2829:
`ZEND_INI_PARSER_POP_ENTRY' undeclared (first use in this function)
/home/jan/software/php5/ext/standard/basic_functions.c:2829: (Each
undeclared identifier is reported only once
/home/jan/software/php5/ext/standard/basic_functions.c:2829: for each
function it appears in.)

Jan.

--
http://www.horde.org - The Horde Project
http://www.ammma.de - discover your knowledge
http://www.tip4all.de - Deine private Tippgemeinschaft

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, visit: http://www.php.net/unsub.php

Re: [PHP-DEV] mb_string overloading and binary data...

[Jan Schneider]

Zitat von David Brown <[EMAIL PROTECTED]>:

> This is kind of a user-space question, but I'm hoping that it concerns
> enough of the PHP infrastructure (conceptually) that this is the right
> place to post it.

This should rather be asked on the i18n mailing list if not on php-general.
But as you already asked:
 
> I've got an application that processes both textual and binary data. The
> domain of the application isn't really relevant, but it makes liberal
> use of string and pcre function.
> 
> Ignoring the fact (?) that php_pcre doesn't seem to be mb-aware...
> Say I were to turn on the mb_string overload support, effectively
> replacing strlen, etc. with equivalent multibyte functions. Is there any
> way then to still get the exact size, in bytes, of a 'string' of binary
> data?

If you write the application, why do you want to depend on a certain php.ini
setting? I won't use overloading at all but call the mbstring functions
directly where appropriate and the "original" string function else.

Please respond to one of the above mentioned lists.

Jan.

--
http://www.horde.org - The Horde Project
http://www.ammma.de - discover your knowledge
http://www.tip4all.de - Deine private Tippgemeinschaft

-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, visit: http://www.php.net/unsub.php

[PHP-DEV] CVS Account Request: akelavlk

[Jan Sunavec]

I want add DotPHP layer into PHP. See webcomp.sourceforge.net Or CVS repository on 
sourceforge.net/projects/webcomp

-- 
PHP Development Mailing List 
To unsubscribe, visit: http://www.php.net/unsub.php

Re: cvs: php4 /ext/standard basic_functions.c

[Jan Schneider]

Sascha Schumann wrote:
sas		Tue Apr  1 05:02:29 2003 EDT

  Modified files:  
/php4/ext/standard	basic_functions.c 
  Log:
  Fix segfault in register_shutdown_function
  
  
Index: php4/ext/standard/basic_functions.c
diff -u php4/ext/standard/basic_functions.c:1.599 php4/ext/standard/basic_functions.c:1.600
--- php4/ext/standard/basic_functions.c:1.599	Tue Apr  1 00:30:32 2003
+++ php4/ext/standard/basic_functions.c	Tue Apr  1 05:02:29 2003
@@ -17,7 +17,7 @@
+--+
  */
 
-/* $Id: basic_functions.c,v 1.599 2003/04/01 05:30:32 sebastian Exp $ */
+/* $Id: basic_functions.c,v 1.600 2003/04/01 10:02:29 sas Exp $ */
 
 #include "php.h"
 #include "php_streams.h"
@@ -2171,6 +2171,7 @@
 	if (zend_get_parameters_array(ht, shutdown_function_entry.arg_count, shutdown_function_entry.arguments) == FAILURE) {
 		RETURN_FALSE;
 	}
+	convert_to_string(shutdown_function_entry.arguments[0]);
 	if (!BG(user_shutdown_function_names)) {
 		ALLOC_HASHTABLE(BG(user_shutdown_function_names));
 		zend_hash_init(BG(user_shutdown_function_names), 0, NULL, (void (*)(void *)) user_shutdown_function_dtor, 0);


I didn't try HEAD yet, but at least in PHP_4_3 it broke registration of 
object methods a la:

register_shutdown_function(array($my_object, 'method_name'));

Jan.