#41236 [NEW]: regression with SSL connections since fix for #39571
From: judas dot iscariote at gmail dot com
Operating system: linux 32/64bit, Win32
PHP version: 5CVS-2007-04-30 (CVS)
PHP Bug Type: Streams related
Bug description: regression with SSL connections since fix for #39571
Description:
Since PHP 5.2.1, our application that has jabber stopped working.all
previuos PHP version works fine, this regression is caused by the fix for
bug #39571,reverting xp_ssl.c to the version shipped with 5.2.0 solves the
problem.
Reproduce code:
---
./configure --with-openssl
");
fwrite($j, "");
$data = '';
// Wait for a response until timeout is reached
$start = time();
$data = '';
$timeout = 10;
$wait = false;
do {
$read = trim(fread($j, 4096));
$data .= $read;
// ok.. this check aint part of the original code
// that breaks,but check if we timed out of reached eof..anyway..
$foo = stream_get_meta_data($j);
if(feof($j) || $foo['timed_out'] || $foo['eof']) break;
// yeah, we know aint nice, but works in **any** other PHP version.
} while (time() <= $start + $timeout && ($wait || $data == '' || $read !=
'' || (substr(rtrim($data), -1) != '>')));
fclose($j);
var_dump($data);
?>
Expected result:
response from the server var_dump'ed as in (all) previous versions
Actual result:
--
Fatal error: Maximum execution time of **30** seconds exceeded...an xdebug
session reveals that the script hang, in a **single** fread() call
an strace shows something hangs permanently with EAGAIN signal.. (probably
the loop in xp_ssl.c from line 398 to 410), note that checking if has
reached EOF or if the stream times out does not help either.
--
Edit bug report at http://bugs.php.net/?id=41236&edit=1
--
Try a CVS snapshot (PHP 4.4):
http://bugs.php.net/fix.php?id=41236&r=trysnapshot44
Try a CVS snapshot (PHP 5.2):
http://bugs.php.net/fix.php?id=41236&r=trysnapshot52
Try a CVS snapshot (PHP 6.0):
http://bugs.php.net/fix.php?id=41236&r=trysnapshot60
Fixed in CVS: http://bugs.php.net/fix.php?id=41236&r=fixedcvs
Fixed in release:
http://bugs.php.net/fix.php?id=41236&r=alreadyfixed
Need backtrace: http://bugs.php.net/fix.php?id=41236&r=needtrace
Need Reproduce Script:http://bugs.php.net/fix.php?id=41236&r=needscript
Try newer version:http://bugs.php.net/fix.php?id=41236&r=oldversion
Not developer issue: http://bugs.php.net/fix.php?id=41236&r=support
Expected behavior:http://bugs.php.net/fix.php?id=41236&r=notwrong
Not enough info:
http://bugs.php.net/fix.php?id=41236&r=notenoughinfo
Submitted twice:
http://bugs.php.net/fix.php?id=41236&r=submittedtwice
register_globals: http://bugs.php.net/fix.php?id=41236&r=globals
PHP 3 support discontinued: http://bugs.php.net/fix.php?id=41236&r=php3
Daylight Savings: http://bugs.php.net/fix.php?id=41236&r=dst
IIS Stability:http://bugs.php.net/fix.php?id=41236&r=isapi
Install GNU Sed: http://bugs.php.net/fix.php?id=41236&r=gnused
Floating point limitations: http://bugs.php.net/fix.php?id=41236&r=float
No Zend Extensions: http://bugs.php.net/fix.php?id=41236&r=nozend
MySQL Configuration Error:http://bugs.php.net/fix.php?id=41236&r=mysqlcfg
#41231 [Opn->Bgs]: mysql lib configure error in x86_64 builds - with patch
ID: 41231 Updated by: [EMAIL PROTECTED] Reported By: hillct-php-bugreport at cthill dot org -Status: Open +Status: Bogus Bug Type: Compile Failure Operating System: Lunux (fc4-fc6) x86_64 PHP Version: 5CVS-2007-04-29 (snap) New Comment: You should be using --with-libdir=lib64 in your configure line. No bug here. Previous Comments: [2007-04-29 21:40:58] hillct-php-bugreport at cthill dot org Description: The configure script fails to search for the mysql libraries in the proper location for some (Fedora) x86_64 environments. There is no way to override this behavior on the configure command line. This error is 100% reproducible in FC4 - FC6 on x86_64 hardware. The solution is a one line fix in the configure script. More properly it should probably be applied to whatever m4 macro is used to construct the script, but that doesn't seem to be available in the snapshot. The patch that corrects this bug is available here: http://www.pastebin.ca/465020 Reproduce code: --- configure --with-zlib --with-mysql --enable-mbstring --enable-fastcgi and --enable-discard-path --no-create --no-recursion Expected result: checking for MySQL support... yes checking for specified location of the MySQL UNIX socket... no checking for MySQL UNIX socket location... /var/lib/mysql/mysql.sock checking for mysql_close in -lmysqlclient... yes This result is achieved by applying the patch located here: http://www.pastebin.ca/465020 Actual result: -- checking for specified location of the MySQL UNIX socket... no checking for MySQL UNIX socket location... /var/lib/mysql/mysql.sock configure: error: Cannot find libmysqlclient under /usr. Note that the MySQL client library is not bundled anymore! -- Edit this bug report at http://bugs.php.net/?id=41231&edit=1
#24956 [Com]: GDlib crash running stable code
ID: 24956 Comment by: mail1883947 at handymail dot org Reported By: danielsabsay at pacbell dot net Status: Feedback Bug Type: GD related Operating System: MacOS 10.2.6 PHP Version: 4.3.2 Assigned To: pajoye New Comment: It still crashes. I compiled libgd 2.0.34 and php from http://snaps.php.net/php5.2-latest.tar.gz by myself using the --enable-maintainer-zts configure flag. The error happens on every page when the gd is accessed by two threads (probably this is the reason the crash happens in different places); when I access only one image it works well. This example is enough to cause the crash: http://it.php.net/manual/en/ref.image.php#60410 #0 0x2b4117be2450 in TT_RunIns () from /usr/lib/libfreetype.so.6 #1 0x2b4117bdb3af in ?? () from /usr/lib/libfreetype.so.6 #2 0x2b4117bd1979 in FT_New_Size () from /usr/lib/libfreetype.so.6 #3 0x2b4117984c58 in gdImageStringFTEx (im=0xf7e7a0, brect=0x438012d0, fg=2, fontlist=, ptsize=8, angle=-0.26179938779914941, x=0, y=54, string=0xdced58 "80", strex=0x0) at gdft.c:976 #4 0x2b4117984f8b in gdImageStringFT (im=0x41, brect=0x41, fg=62, fontlist=0x3f , ptsize=512, angle=63302.914951680381, x=0, y=0, string=0xdced58 "80") at gdft.c:811 #5 0x2b41170be6d9 in php_imagettftext_common (ht=, return_value=0xdced08, return_value_ptr=, this_ptr=, return_value_used=, tsrm_ls=0xd9ae40, mode=0, extended=0) at /opt/php5.2-200704300630/ext/gd/gd.c:4260 #6 0x2b41170be768 in zif_imagettftext (ht=65, return_value=0x41, return_value_ptr=0x3e, this_ptr=0x3f, return_value_used=0, tsrm_ls=0x0) at /opt/php5.2-200704300630/ext/gd/gd.c:4164 #7 0x2b41173232fc in zend_do_fcall_common_helper_SPEC (execute_data=0x43802950, tsrm_ls=0xd9ae40) at /opt/php5.2-200704300630/Zend/zend_vm_execute.h:200 #8 0x2b411731098f in execute (op_array=0xc9eed0, tsrm_ls=0xd9ae40) at /opt/php5.2-200704300630/Zend/zend_vm_execute.h:92 #9 0x2b4117322cf7 in zend_do_fcall_common_helper_SPEC (execute_data=0x43802e00, tsrm_ls=0xd9ae40) at /opt/php5.2-200704300630/Zend/zend_vm_execute.h:234 #10 0x2b411731098f in execute (op_array=0xdcbef8, tsrm_ls=0xd9ae40) at /opt/php5.2-200704300630/Zend/zend_vm_execute.h:92 #11 0x2b41172ee13a in zend_execute_scripts (type=8, tsrm_ls=0xd9ae40, retval=0x3e, file_count=3) at /opt/php5.2-200704300630/Zend/zend.c:1134 #12 0x2b41172a329b in php_execute_script (primary_file=0x43805350, tsrm_ls=0xd9ae40) at /opt/php5.2-200704300630/main/main.c:1794 Previous Comments: [2007-04-29 21:14:23] [EMAIL PROTECTED] Please try using this CVS snapshot: http://snaps.php.net/php5.2-latest.tar.gz For Windows: http://snaps.php.net/win32/php5.2-win32-latest.zip "Same error under debian GNU/Linux_x64 with php 5.2.1:" Please try using the last 5.2 snapshot and RC. If you still have this problem: Do you compile php yourself? Which GD do you use? Bundled or external? Do you have a script? Please note that we do not support debian package but only install done from the official sources (that's true for the external libgd too, debian's libgd being everything but clean). [2007-04-29 21:05:16] mail1878743 at handymail dot org Same error under debian GNU/Linux_x64 with php 5.2.1: #0 0x2ac7 in ?? () #1 0x2ac71ca19123 in ?? () from /usr/lib/libfreetype.so.6 #2 0x2ac71ca191e0 in FT_Done_Face () from /usr/lib/libfreetype.so.6 #3 0x2ac71c8cef6d in ?? () from /usr/lib/libgd.so.2 #4 0x2ac71c8cd65d in gdCacheDelete () from /usr/lib/libgd.so.2 #5 0x2ac71c8cf5a6 in gdFontCacheShutdown () from /usr/lib/libgd.so.2 #6 0x2ac71c0115c9 in zm_deactivate_gd (type=16158400, module_number=16224416, tsrm_ls=0x1522e20) at ext/gd/gd.c:1303 #7 0x2ac71c23d4cf in module_registry_cleanup (module=, tsrm_ls=0x1522e20) at Zend/zend_API.c:1945 #8 0x2ac71c246b58 in zend_hash_apply (ht=0x2ac71c7abec0, apply_func=0x2ac71c23d4b0 , tsrm_ls=0xd8ca40) at Zend/zend_hash.c:673 #9 0x2ac71c23b0b0 in zend_deactivate_modules (tsrm_ls=0xd8ca40) at Zend/zend.c:839 #10 0x2ac71c1f2495 in php_request_shutdown (dummy=) at main/main.c:1293 [2003-08-10 21:49:58] [EMAIL PROTECTED] No feedback was provided. The bug is being suspended because we assume that you are no longer experiencing the problem. If this is not the case and you are able to provide the information that was requested earlier, please do so and change the status of the bug back to "Open". Thank you. [2003-08-06 09:23:10] [EMAIL PROTECTED] Please try using this CVS snapshot: http://snaps.php.net/php4-STABLE-latest.tar.gz For Windows: http://snaps.php.net/win3
#40943 [NoF->Opn]: Function str_ireplace() crashes Apache server
ID: 40943 User updated by: dzoukr at dzoukr dot cz Reported By: dzoukr at dzoukr dot cz -Status: No Feedback +Status: Open Bug Type: Reproducible crash Operating System: Windows XP SP2 PHP Version: 5.2.1 New Comment: Status changed to open. That`s all I can do wit it ;) Previous Comments: [2007-04-29 21:21:54] qrf_coil at yahoo dot fr Good to hear, obviously it's a real bug.. could someone change the status of this bug ? [2007-04-19 20:40:56] registry at ivide dot com I am having this problem as well on php 5.2.1. It does not happen on another server I have running php 5.1.6. I had it happen trying to str_ireplace with an array of characters to be replaced with "". I am using XP Sp2, Apache 1.3.37, and Php stable 5.2.1. The apache logs show "zend_mm_heap corrupted" and the browser says "The connection to the server was interrupted". [2007-04-16 10:25:34] qrf_coil at yahoo dot fr I've experienced the same problem on a production server of a customer, quiet annoying, even using str_replace instead will correct the problem The server reported : ERROR The requested URL could not be retrieved While trying to retrieve the URL: The following error was encountered: * Zero Sized Reply Server used is a Unix with PHP 5.2.1-0.dotdeb.1 with Suhosin-Patch 0.9.6.2 (cli) (built: Feb 9 2007 04:46:32) [2007-04-06 01:00:00] php-bugs at lists dot php dot net No feedback was provided for this bug for over a week, so it is being suspended automatically. If you are able to provide the information that was originally requested, please do so and change the status of the bug back to "Open". [2007-03-29 09:20:44] dzoukr at dzoukr dot cz OK. Anyway, thanks for looking at it. Maybe it is some crazy combination of Apache settings and PHP. ;) As I said - I`m using Wamp server and I made almost no configuration changes, so who knows. Now I`m using str_replace() and no problem occured. 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/40943 -- Edit this bug report at http://bugs.php.net/?id=40943&edit=1
#41237 [NEW]: XOR problem
From: victorepand at gmail dot com Operating system: Linux PHP version: 5.2.1 PHP Bug Type: Math related Bug description: XOR problem Description: I am finding a difference between the same bitwise arithmetic from one server to the next when using PHP. What's more, this bitwise arithmetic is necessary for the PHP script to run, so as a result it will only function on one server, but not the other. Here is an example I am using to demonstrate this: if ((43814 ^ -4738698913)!=-443704711) print "incorrect result"; else print "correct result"; The (^) operator is an XOR bitwise arithmetic function as shown here: http://us2.php.net/manual/en/language.operators.bitwise.php and I am required to use numbers like the ones shown. On one server, I have tried both PHP 4.4.0 and PHP 5.1.0RC1 and the math works correctly for both (the correct answer as shown above is -443704711). But on another server, I have tried the same math with both PHP 4.4.6 and PHP 5.2.1, and it does not work correctly with either version of PHP! The result I get at that server is: -2147439834. I have no idea what could be the problem, but I can show you the PHP Info for both servers and perhaps you can detect what might be the difference? Here is the PHP Info for the server that works correctly using PHP 5.1.0RC1: http://www.buycellularphones.info/cron/special/info.php Here is the PHP Info for the other server using PHP 5.2.1 that does not work correctly: http://www.customdesignpostcards.com/cron/special/info.php Reproduce code: --- if ((43814 ^ -4738698913)!=-443704711) print "incorrect result"; else print "correct result"; Expected result: correct result Actual result: -- incorrect result -- Edit bug report at http://bugs.php.net/?id=41237&edit=1 -- Try a CVS snapshot (PHP 4.4): http://bugs.php.net/fix.php?id=41237&r=trysnapshot44 Try a CVS snapshot (PHP 5.2): http://bugs.php.net/fix.php?id=41237&r=trysnapshot52 Try a CVS snapshot (PHP 6.0): http://bugs.php.net/fix.php?id=41237&r=trysnapshot60 Fixed in CVS: http://bugs.php.net/fix.php?id=41237&r=fixedcvs Fixed in release: http://bugs.php.net/fix.php?id=41237&r=alreadyfixed Need backtrace: http://bugs.php.net/fix.php?id=41237&r=needtrace Need Reproduce Script:http://bugs.php.net/fix.php?id=41237&r=needscript Try newer version:http://bugs.php.net/fix.php?id=41237&r=oldversion Not developer issue: http://bugs.php.net/fix.php?id=41237&r=support Expected behavior:http://bugs.php.net/fix.php?id=41237&r=notwrong Not enough info: http://bugs.php.net/fix.php?id=41237&r=notenoughinfo Submitted twice: http://bugs.php.net/fix.php?id=41237&r=submittedtwice register_globals: http://bugs.php.net/fix.php?id=41237&r=globals PHP 3 support discontinued: http://bugs.php.net/fix.php?id=41237&r=php3 Daylight Savings: http://bugs.php.net/fix.php?id=41237&r=dst IIS Stability:http://bugs.php.net/fix.php?id=41237&r=isapi Install GNU Sed: http://bugs.php.net/fix.php?id=41237&r=gnused Floating point limitations: http://bugs.php.net/fix.php?id=41237&r=float No Zend Extensions: http://bugs.php.net/fix.php?id=41237&r=nozend MySQL Configuration Error:http://bugs.php.net/fix.php?id=41237&r=mysqlcfg
#41235 [Com]: ereg_replace does not replace all occurrences
ID: 41235 Comment by: robinv at gmail dot com Reported By: alexey at gmail dot com Status: Open Bug Type: *Search functions Operating System: linux PHP Version: 4.4.6 New Comment: Not a bug: The pattern matches '&qs=&q' and replaces it with '&qs=&q'. It matches '&si=0&q' and replaces it with '&si=0&q'. It matches '&nh=25' and replaces it with '&nh=25' and it matches '&lk=1&' replacing it with '&lk=1&'. Which is exactly what you've told it to do. Previous Comments: [2007-04-30 06:47:34] alexey at gmail dot com Description: ereg_replace() does not replace all occurrences of a regular expression as the following example shows. The pattern &([^;][^;][^;][^;][^;]) is replaced with &\\1 in a string http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 The output is http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 which indicates for example that the last & was not replaced. Reproduce code: --- http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0";; $pattern="&([^;][^;][^;][^;][^;])"; $data=ereg_replace($pattern, "&\\1", $data); echo $data."\n"; // http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 $data=ereg_replace($pattern, "&\\1", $data); echo $data."\n"; // http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 ?> Expected result: http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 Actual result: -- http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 -- Edit this bug report at http://bugs.php.net/?id=41235&edit=1
#24956 [Fbk->Bgs]: GDlib crash running stable code
ID: 24956 Updated by: [EMAIL PROTECTED] Reported By: danielsabsay at pacbell dot net -Status: Feedback +Status: Bogus Bug Type: GD related Operating System: MacOS 10.2.6 PHP Version: 4.3.2 Assigned To: pajoye New Comment: "It still crashes. I compiled libgd 2.0.34 and php from http://snaps.php.net/php5.2-latest.tar.gz by myself using the --enable-maintainer-zts configure flag." Please use the bundled library. It contains more fixes and features. It also has a better thread support. To compile php using the bundled version, do: "./configure" "--with-gd" ... instead of "--with-gd=/opt/" It is clearly one of the thread safety fixed in 5.2.1 and 5.2.2 and only available in the bundled library. GD 2.1.0 will have them, but that does not change the golden rule: always use the bundled library :) Cannot be fixed in php, it is not a php bug per se > bogus. Previous Comments: [2007-04-30 08:29:22] mail1883947 at handymail dot org It still crashes. I compiled libgd 2.0.34 and php from http://snaps.php.net/php5.2-latest.tar.gz by myself using the --enable-maintainer-zts configure flag. The error happens on every page when the gd is accessed by two threads (probably this is the reason the crash happens in different places); when I access only one image it works well. This example is enough to cause the crash: http://it.php.net/manual/en/ref.image.php#60410 #0 0x2b4117be2450 in TT_RunIns () from /usr/lib/libfreetype.so.6 #1 0x2b4117bdb3af in ?? () from /usr/lib/libfreetype.so.6 #2 0x2b4117bd1979 in FT_New_Size () from /usr/lib/libfreetype.so.6 #3 0x2b4117984c58 in gdImageStringFTEx (im=0xf7e7a0, brect=0x438012d0, fg=2, fontlist=, ptsize=8, angle=-0.26179938779914941, x=0, y=54, string=0xdced58 "80", strex=0x0) at gdft.c:976 #4 0x2b4117984f8b in gdImageStringFT (im=0x41, brect=0x41, fg=62, fontlist=0x3f , ptsize=512, angle=63302.914951680381, x=0, y=0, string=0xdced58 "80") at gdft.c:811 #5 0x2b41170be6d9 in php_imagettftext_common (ht=, return_value=0xdced08, return_value_ptr=, this_ptr=, return_value_used=, tsrm_ls=0xd9ae40, mode=0, extended=0) at /opt/php5.2-200704300630/ext/gd/gd.c:4260 #6 0x2b41170be768 in zif_imagettftext (ht=65, return_value=0x41, return_value_ptr=0x3e, this_ptr=0x3f, return_value_used=0, tsrm_ls=0x0) at /opt/php5.2-200704300630/ext/gd/gd.c:4164 #7 0x2b41173232fc in zend_do_fcall_common_helper_SPEC (execute_data=0x43802950, tsrm_ls=0xd9ae40) at /opt/php5.2-200704300630/Zend/zend_vm_execute.h:200 #8 0x2b411731098f in execute (op_array=0xc9eed0, tsrm_ls=0xd9ae40) at /opt/php5.2-200704300630/Zend/zend_vm_execute.h:92 #9 0x2b4117322cf7 in zend_do_fcall_common_helper_SPEC (execute_data=0x43802e00, tsrm_ls=0xd9ae40) at /opt/php5.2-200704300630/Zend/zend_vm_execute.h:234 #10 0x2b411731098f in execute (op_array=0xdcbef8, tsrm_ls=0xd9ae40) at /opt/php5.2-200704300630/Zend/zend_vm_execute.h:92 #11 0x2b41172ee13a in zend_execute_scripts (type=8, tsrm_ls=0xd9ae40, retval=0x3e, file_count=3) at /opt/php5.2-200704300630/Zend/zend.c:1134 #12 0x2b41172a329b in php_execute_script (primary_file=0x43805350, tsrm_ls=0xd9ae40) at /opt/php5.2-200704300630/main/main.c:1794 [2007-04-29 21:14:23] [EMAIL PROTECTED] Please try using this CVS snapshot: http://snaps.php.net/php5.2-latest.tar.gz For Windows: http://snaps.php.net/win32/php5.2-win32-latest.zip "Same error under debian GNU/Linux_x64 with php 5.2.1:" Please try using the last 5.2 snapshot and RC. If you still have this problem: Do you compile php yourself? Which GD do you use? Bundled or external? Do you have a script? Please note that we do not support debian package but only install done from the official sources (that's true for the external libgd too, debian's libgd being everything but clean). [2007-04-29 21:05:16] mail1878743 at handymail dot org Same error under debian GNU/Linux_x64 with php 5.2.1: #0 0x2ac7 in ?? () #1 0x2ac71ca19123 in ?? () from /usr/lib/libfreetype.so.6 #2 0x2ac71ca191e0 in FT_Done_Face () from /usr/lib/libfreetype.so.6 #3 0x2ac71c8cef6d in ?? () from /usr/lib/libgd.so.2 #4 0x2ac71c8cd65d in gdCacheDelete () from /usr/lib/libgd.so.2 #5 0x2ac71c8cf5a6 in gdFontCacheShutdown () from /usr/lib/libgd.so.2 #6 0x2ac71c0115c9 in zm_deactivate_gd (type=16158400, module_number=16224416, tsrm_ls=0x1522e20) at ext/gd/gd.c:1303 #7 0x2ac71c23d4cf in module_registry_cleanup (module=, tsrm_ls=0x1522e20) at Zend/zend_API.c:1945 #8 0x2ac71c246b58 in zend_hash_apply (ht=0x2ac71c7abec0, apply_func=0x2ac71c23d4b0 , tsrm_ls=0xd8ca40) at Zend/zend_hash.c:673 #9 0x2ac71c23b0b0 in zend_de
#41238 [NEW]: PSpell fails to open dictionaries when server is under load
From: spam02 at pornel dot net Operating system: Debian stable PHP version: 5.2.1 PHP Bug Type: Pspell related Bug description: PSpell fails to open dictionaries when server is under load Description: Problem reported in #33095 might be valid. I can reproduce the same problem, every time, on PHP/5.2.1, Linux/2.6.8, Apache/1.3.33, Aspell/0.60.3 It appears that PHP process which uses PSpell must be the only process on the server (even images loading concurently affect it.), otherwise PSpell fails to open dictionary. Failure seems to be determined per process, because PSpell will either work first time or always fail for a certain process (when you retry in a loop, even with a delay, you always get failure). I've tried different invocation methods, pspell_clear_session, clearstatcache, different dictionaries - nothing helps. Reproduce code: --- Try pspell on pages with and without images. Different browsers have different patterns for requesting images, so for example failure rate is higher in Firefox than Opera (I'm not mad, really). Actual result: -- Warning: pspell_new_config() [function.pspell-new-config]: PSPELL couldn't open the dictionary. reason: No word lists can be found for the language "en". in ... -- Edit bug report at http://bugs.php.net/?id=41238&edit=1 -- Try a CVS snapshot (PHP 4.4): http://bugs.php.net/fix.php?id=41238&r=trysnapshot44 Try a CVS snapshot (PHP 5.2): http://bugs.php.net/fix.php?id=41238&r=trysnapshot52 Try a CVS snapshot (PHP 6.0): http://bugs.php.net/fix.php?id=41238&r=trysnapshot60 Fixed in CVS: http://bugs.php.net/fix.php?id=41238&r=fixedcvs Fixed in release: http://bugs.php.net/fix.php?id=41238&r=alreadyfixed Need backtrace: http://bugs.php.net/fix.php?id=41238&r=needtrace Need Reproduce Script:http://bugs.php.net/fix.php?id=41238&r=needscript Try newer version:http://bugs.php.net/fix.php?id=41238&r=oldversion Not developer issue: http://bugs.php.net/fix.php?id=41238&r=support Expected behavior:http://bugs.php.net/fix.php?id=41238&r=notwrong Not enough info: http://bugs.php.net/fix.php?id=41238&r=notenoughinfo Submitted twice: http://bugs.php.net/fix.php?id=41238&r=submittedtwice register_globals: http://bugs.php.net/fix.php?id=41238&r=globals PHP 3 support discontinued: http://bugs.php.net/fix.php?id=41238&r=php3 Daylight Savings: http://bugs.php.net/fix.php?id=41238&r=dst IIS Stability:http://bugs.php.net/fix.php?id=41238&r=isapi Install GNU Sed: http://bugs.php.net/fix.php?id=41238&r=gnused Floating point limitations: http://bugs.php.net/fix.php?id=41238&r=float No Zend Extensions: http://bugs.php.net/fix.php?id=41238&r=nozend MySQL Configuration Error:http://bugs.php.net/fix.php?id=41238&r=mysqlcfg
#41239 [NEW]: Special chars not handled as alpha numeric
From: rene dot stabingis at gmail dot com
Operating system: Linux 2.6.13.2 (Debian 3.1)
PHP version: 5.2.1
PHP Bug Type: *Regular Expressions
Bug description: Special chars not handled as alpha numeric
Description:
This code should delete all non-alphanumeric signs (except spaces).
If you have latin special chars like "ü", on the linux machine they are
deleted. On a win environment using same php version they are kept as
alnum. The deb machine runs with ISO-8859-1 charset.
Reproduce code:
---
$somePhrase = 'i don\'t know why, but many people know über';
echo preg_replace('/[^[:alnum:][:space:]]/', '', $somePhrase);
Expected result:
i dont know why but many people know über
Actual result:
--
i dont know why but many people know ber
--
Edit bug report at http://bugs.php.net/?id=41239&edit=1
--
Try a CVS snapshot (PHP 4.4):
http://bugs.php.net/fix.php?id=41239&r=trysnapshot44
Try a CVS snapshot (PHP 5.2):
http://bugs.php.net/fix.php?id=41239&r=trysnapshot52
Try a CVS snapshot (PHP 6.0):
http://bugs.php.net/fix.php?id=41239&r=trysnapshot60
Fixed in CVS: http://bugs.php.net/fix.php?id=41239&r=fixedcvs
Fixed in release:
http://bugs.php.net/fix.php?id=41239&r=alreadyfixed
Need backtrace: http://bugs.php.net/fix.php?id=41239&r=needtrace
Need Reproduce Script:http://bugs.php.net/fix.php?id=41239&r=needscript
Try newer version:http://bugs.php.net/fix.php?id=41239&r=oldversion
Not developer issue: http://bugs.php.net/fix.php?id=41239&r=support
Expected behavior:http://bugs.php.net/fix.php?id=41239&r=notwrong
Not enough info:
http://bugs.php.net/fix.php?id=41239&r=notenoughinfo
Submitted twice:
http://bugs.php.net/fix.php?id=41239&r=submittedtwice
register_globals: http://bugs.php.net/fix.php?id=41239&r=globals
PHP 3 support discontinued: http://bugs.php.net/fix.php?id=41239&r=php3
Daylight Savings: http://bugs.php.net/fix.php?id=41239&r=dst
IIS Stability:http://bugs.php.net/fix.php?id=41239&r=isapi
Install GNU Sed: http://bugs.php.net/fix.php?id=41239&r=gnused
Floating point limitations: http://bugs.php.net/fix.php?id=41239&r=float
No Zend Extensions: http://bugs.php.net/fix.php?id=41239&r=nozend
MySQL Configuration Error:http://bugs.php.net/fix.php?id=41239&r=mysqlcfg
#41239 [Opn->Bgs]: Special chars not handled as alpha numeric
ID: 41239
Updated by: [EMAIL PROTECTED]
Reported By: rene dot stabingis at gmail dot com
-Status: Open
+Status: Bogus
Bug Type: *Regular Expressions
Operating System: Linux 2.6.13.2 (Debian 3.1)
PHP Version: 5.2.1
New Comment:
You need to use the correct locale, see setlocale() documentation.
Previous Comments:
[2007-04-30 10:21:25] rene dot stabingis at gmail dot com
Description:
This code should delete all non-alphanumeric signs (except spaces).
If you have latin special chars like "ü", on the linux machine they are
deleted. On a win environment using same php version they are kept as
alnum. The deb machine runs with ISO-8859-1 charset.
Reproduce code:
---
$somePhrase = 'i don\'t know why, but many people know über';
echo preg_replace('/[^[:alnum:][:space:]]/', '', $somePhrase);
Expected result:
i dont know why but many people know über
Actual result:
--
i dont know why but many people know ber
--
Edit this bug report at http://bugs.php.net/?id=41239&edit=1
#41235 [Opn->Csd]: ereg_replace does not replace all occurrences
ID: 41235 User updated by: alexey at gmail dot com Reported By: alexey at gmail dot com -Status: Open +Status: Closed Bug Type: *Search functions Operating System: linux PHP Version: 4.4.6 New Comment: Oh well. Thanks for your explanation. Closed. Previous Comments: [2007-04-30 09:33:18] robinv at gmail dot com Not a bug: The pattern matches '&qs=&q' and replaces it with '&qs=&q'. It matches '&si=0&q' and replaces it with '&si=0&q'. It matches '&nh=25' and replaces it with '&nh=25' and it matches '&lk=1&' replacing it with '&lk=1&'. Which is exactly what you've told it to do. [2007-04-30 06:47:34] alexey at gmail dot com Description: ereg_replace() does not replace all occurrences of a regular expression as the following example shows. The pattern &([^;][^;][^;][^;][^;]) is replaced with &\\1 in a string http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 The output is http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 which indicates for example that the last & was not replaced. Reproduce code: --- http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0";; $pattern="&([^;][^;][^;][^;][^;])"; $data=ereg_replace($pattern, "&\\1", $data); echo $data."\n"; // http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 $data=ereg_replace($pattern, "&\\1", $data); echo $data."\n"; // http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 ?> Expected result: http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 Actual result: -- http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 -- Edit this bug report at http://bugs.php.net/?id=41235&edit=1
#41235 [Csd]: ereg_replace does not replace all occurrences
ID: 41235 User updated by: alexey at gmail dot com Reported By: alexey at gmail dot com Status: Closed Bug Type: *Search functions Operating System: linux PHP Version: 4.4.6 New Comment: The PHP manual does not specify the replacement order. What if we replace it from right to the left? This will give us a different answer. Also, the definition of a regular expression does not impose the evaluation order. Therefore, the right answer is to replace each occurrence of & in the original string with a &. Previous Comments: [2007-04-30 11:05:12] alexey at gmail dot com Oh well. Thanks for your explanation. Closed. [2007-04-30 09:33:18] robinv at gmail dot com Not a bug: The pattern matches '&qs=&q' and replaces it with '&qs=&q'. It matches '&si=0&q' and replaces it with '&si=0&q'. It matches '&nh=25' and replaces it with '&nh=25' and it matches '&lk=1&' replacing it with '&lk=1&'. Which is exactly what you've told it to do. [2007-04-30 06:47:34] alexey at gmail dot com Description: ereg_replace() does not replace all occurrences of a regular expression as the following example shows. The pattern &([^;][^;][^;][^;][^;]) is replaced with &\\1 in a string http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 The output is http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 which indicates for example that the last & was not replaced. Reproduce code: --- http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0";; $pattern="&([^;][^;][^;][^;][^;])"; $data=ereg_replace($pattern, "&\\1", $data); echo $data."\n"; // http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 $data=ereg_replace($pattern, "&\\1", $data); echo $data."\n"; // http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 ?> Expected result: http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 Actual result: -- http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 -- Edit this bug report at http://bugs.php.net/?id=41235&edit=1
#41239 [Bgs]: Special chars not handled as alpha numeric
ID: 41239
User updated by: rene dot stabingis at gmail dot com
Reported By: rene dot stabingis at gmail dot com
Status: Bogus
Bug Type: *Regular Expressions
Operating System: Linux 2.6.13.2 (Debian 3.1)
PHP Version: 5.2.1
New Comment:
I'm sorry but setlocale(LC_ALL, 'de_DE'); has no effect. Same with an
explicit set of the german or german/european charset. (btw same
behaviour on a freebsd4 machine)
Until now i have only seen problems with this on ubuntu machines
running with utf8 standard charset
Previous Comments:
[2007-04-30 10:46:50] [EMAIL PROTECTED]
You need to use the correct locale, see setlocale() documentation.
[2007-04-30 10:21:25] rene dot stabingis at gmail dot com
Description:
This code should delete all non-alphanumeric signs (except spaces).
If you have latin special chars like "ü", on the linux machine they are
deleted. On a win environment using same php version they are kept as
alnum. The deb machine runs with ISO-8859-1 charset.
Reproduce code:
---
$somePhrase = 'i don\'t know why, but many people know über';
echo preg_replace('/[^[:alnum:][:space:]]/', '', $somePhrase);
Expected result:
i dont know why but many people know über
Actual result:
--
i dont know why but many people know ber
--
Edit this bug report at http://bugs.php.net/?id=41239&edit=1
#41234 [Opn->Bgs]: Additional Notes
ID: 41234 Updated by: [EMAIL PROTECTED] Reported By: cpriest at warpmail dot net -Status: Open +Status: Bogus Bug Type: MySQLi related Operating System: Windows Vista Business PHP Version: 5.2.1 New Comment: You probably have another libmysql.dll or you have mixed up several PHP installs. The mysqli extension works fine. Previous Comments: [2007-04-30 05:02:00] cpriest at warpmail dot net I have verified the following: - libmysql.dll is located in c:\php - c:\php is in the environment PATH variable (rebooted afterwards) - php_mysql.dll and php_mysqli.dll are both in .\ext and are being loaded (mysqli load fails) I have also tried the following: - Apache 2.2.x - Copying the libmysql.dll to c:\windows\system32 - Copying the libmysql.dll from my MySQL 4.1.x installation into c:\php All result in the same error "Procedure could not be found" while loading php_mysqli.dll I'm reasonably certain that this is related to my Windows Vista Business install, I have successfully installed this environment many times before on Windows XP and Linux installations with no problems. [2007-04-30 04:56:16] cpriest at warpmail dot net Description: Unable to get php_mysqli.dll to load properly on Windows Vista Business with Apache 2.0.59 or 2.2.x. Reproduce code: --- Install Apache 2.0.59 + PHP 5.2.1 Try loading the php_mysqli.dll Expected result: Able to load extension Actual result: -- Cannot load, get error "Procedure could not be found." -- Edit this bug report at http://bugs.php.net/?id=41234&edit=1
#41240 [NEW]: numbers bugs
From: webmaster at cashoclic dot fr Operating system: windowsxp PHP version: 5.2.1 PHP Bug Type: *Math Functions Bug description: numbers bugs Description: 100-99.99=0.015 Here is the bug i don't understand why Reproduce code: --- echo 100-99.99; Expected result: 0.01 Actual result: -- 0.015 -- Edit bug report at http://bugs.php.net/?id=41240&edit=1 -- Try a CVS snapshot (PHP 4.4): http://bugs.php.net/fix.php?id=41240&r=trysnapshot44 Try a CVS snapshot (PHP 5.2): http://bugs.php.net/fix.php?id=41240&r=trysnapshot52 Try a CVS snapshot (PHP 6.0): http://bugs.php.net/fix.php?id=41240&r=trysnapshot60 Fixed in CVS: http://bugs.php.net/fix.php?id=41240&r=fixedcvs Fixed in release: http://bugs.php.net/fix.php?id=41240&r=alreadyfixed Need backtrace: http://bugs.php.net/fix.php?id=41240&r=needtrace Need Reproduce Script:http://bugs.php.net/fix.php?id=41240&r=needscript Try newer version:http://bugs.php.net/fix.php?id=41240&r=oldversion Not developer issue: http://bugs.php.net/fix.php?id=41240&r=support Expected behavior:http://bugs.php.net/fix.php?id=41240&r=notwrong Not enough info: http://bugs.php.net/fix.php?id=41240&r=notenoughinfo Submitted twice: http://bugs.php.net/fix.php?id=41240&r=submittedtwice register_globals: http://bugs.php.net/fix.php?id=41240&r=globals PHP 3 support discontinued: http://bugs.php.net/fix.php?id=41240&r=php3 Daylight Savings: http://bugs.php.net/fix.php?id=41240&r=dst IIS Stability:http://bugs.php.net/fix.php?id=41240&r=isapi Install GNU Sed: http://bugs.php.net/fix.php?id=41240&r=gnused Floating point limitations: http://bugs.php.net/fix.php?id=41240&r=float No Zend Extensions: http://bugs.php.net/fix.php?id=41240&r=nozend MySQL Configuration Error:http://bugs.php.net/fix.php?id=41240&r=mysqlcfg
#41240 [Opn->Bgs]: numbers bugs
ID: 41240 Updated by: [EMAIL PROTECTED] Reported By: webmaster at cashoclic dot fr -Status: Open +Status: Bogus Bug Type: *Math Functions Operating System: windowsxp PHP Version: 5.2.1 New Comment: Floating point values have a limited precision. Hence a value might not have the same string representation after any processing. That also includes writing a floating point value in your script and directly printing it without any mathematical operations. If you would like to know more about "floats" and what IEEE 754 is read this: http://docs.sun.com/source/806-3568/ncg_goldberg.html Thank you for your interest in PHP. Previous Comments: [2007-04-30 11:50:10] webmaster at cashoclic dot fr Description: 100-99.99=0.015 Here is the bug i don't understand why Reproduce code: --- echo 100-99.99; Expected result: 0.01 Actual result: -- 0.015 -- Edit this bug report at http://bugs.php.net/?id=41240&edit=1
#41231 [Bgs->Opn]: x86_64 config library path search documentation problem
ID: 41231 User updated by: hillct-php-bugreport at cthill dot org -Summary: mysql lib configure error in x86_64 builds - with patch Reported By: hillct-php-bugreport at cthill dot org -Status: Bogus +Status: Open -Bug Type: Compile Failure +Bug Type: *Configuration Issues Operating System: Lunux (fc4-fc6) x86_64 PHP Version: 5CVS-2007-04-29 (snap) New Comment: The documentation indicates that --with-libdir should be used to specify a full path, which is a case tested before this bug was submitted. The documentation should be updated to reflect use of the libdir argument with relative paths or path fragments. Having consulted many supposed expert php developers, who have all told me the configure script is broken in this way (one having given me the patch as a solution), it seems the documentation needs to be clarified on this point. Previous Comments: [2007-04-30 07:38:05] [EMAIL PROTECTED] You should be using --with-libdir=lib64 in your configure line. No bug here. [2007-04-29 21:40:58] hillct-php-bugreport at cthill dot org Description: The configure script fails to search for the mysql libraries in the proper location for some (Fedora) x86_64 environments. There is no way to override this behavior on the configure command line. This error is 100% reproducible in FC4 - FC6 on x86_64 hardware. The solution is a one line fix in the configure script. More properly it should probably be applied to whatever m4 macro is used to construct the script, but that doesn't seem to be available in the snapshot. The patch that corrects this bug is available here: http://www.pastebin.ca/465020 Reproduce code: --- configure --with-zlib --with-mysql --enable-mbstring --enable-fastcgi and --enable-discard-path --no-create --no-recursion Expected result: checking for MySQL support... yes checking for specified location of the MySQL UNIX socket... no checking for MySQL UNIX socket location... /var/lib/mysql/mysql.sock checking for mysql_close in -lmysqlclient... yes This result is achieved by applying the patch located here: http://www.pastebin.ca/465020 Actual result: -- checking for specified location of the MySQL UNIX socket... no checking for MySQL UNIX socket location... /var/lib/mysql/mysql.sock configure: error: Cannot find libmysqlclient under /usr. Note that the MySQL client library is not bundled anymore! -- Edit this bug report at http://bugs.php.net/?id=41231&edit=1
#41240 [Bgs]: numbers bugs
ID: 41240 User updated by: webmaster at cashoclic dot fr Reported By: webmaster at cashoclic dot fr Status: Bogus Bug Type: *Math Functions Operating System: windowsxp PHP Version: 5.2.1 New Comment: i'm french and i havent understand your message. what can i do for the problem ? Because if i do $a=100 and $b=99.99 and $c=$a-$b $c=0.015 Previous Comments: [2007-04-30 11:53:43] [EMAIL PROTECTED] Floating point values have a limited precision. Hence a value might not have the same string representation after any processing. That also includes writing a floating point value in your script and directly printing it without any mathematical operations. If you would like to know more about "floats" and what IEEE 754 is read this: http://docs.sun.com/source/806-3568/ncg_goldberg.html Thank you for your interest in PHP. [2007-04-30 11:50:10] webmaster at cashoclic dot fr Description: 100-99.99=0.015 Here is the bug i don't understand why Reproduce code: --- echo 100-99.99; Expected result: 0.01 Actual result: -- 0.015 -- Edit this bug report at http://bugs.php.net/?id=41240&edit=1
#41241 [NEW]: Realization array type. SQL 2003, S091
From: develar at gmail dot com
Operating system: Windows XP SP2
PHP version: 5.2.1
PHP Bug Type: PDO related
Bug description: Realization array type. SQL 2003, S091
Description:
SQL 2003, S091.
Reproduce code:
---
PostgreSQL 8.2
select proargnames from pg_catalog.pg_proc order by proargnames limit 1
Expected result:
array (
0 =>
array (
'proargnames' =>
'{"","","",id,uri,time_create,priority_base_time_create,title,time_modify,status,priority_base_title,type_name,leaf,version,author_name,author_surname,author_patronymic,privilege_edit,privilege_delete,privilege_grant}',
),
)
Actual result:
--
array (
0 =>
array (
'proargnames' =>
'{"","","",id,uri,time_create,priority_base_time_create,title,time_modify,array
(
0 =>
array (
'proargnames' =>
array (
0 => '',
1 => '',
2 => '',
3 => 'id',
4 => 'uri',
5 => 'time_create',
6 => 'priority_base_time_create',
7 => 'title',
8 => 'time_modify',
9 => 'status',
10 => 'priority_base_title',
11 => 'type_name',
12 => 'leaf',
13 => 'version',
14 => 'author_name',
15 => 'author_surname',
16 => 'author_patronymic',
17 => 'privilege_edit',
18 => 'privilege_delete',
19 => 'privilege_grant',
),
),
)}',
),
)
--
Edit bug report at http://bugs.php.net/?id=41241&edit=1
--
Try a CVS snapshot (PHP 4.4):
http://bugs.php.net/fix.php?id=41241&r=trysnapshot44
Try a CVS snapshot (PHP 5.2):
http://bugs.php.net/fix.php?id=41241&r=trysnapshot52
Try a CVS snapshot (PHP 6.0):
http://bugs.php.net/fix.php?id=41241&r=trysnapshot60
Fixed in CVS: http://bugs.php.net/fix.php?id=41241&r=fixedcvs
Fixed in release:
http://bugs.php.net/fix.php?id=41241&r=alreadyfixed
Need backtrace: http://bugs.php.net/fix.php?id=41241&r=needtrace
Need Reproduce Script:http://bugs.php.net/fix.php?id=41241&r=needscript
Try newer version:http://bugs.php.net/fix.php?id=41241&r=oldversion
Not developer issue: http://bugs.php.net/fix.php?id=41241&r=support
Expected behavior:http://bugs.php.net/fix.php?id=41241&r=notwrong
Not enough info:
http://bugs.php.net/fix.php?id=41241&r=notenoughinfo
Submitted twice:
http://bugs.php.net/fix.php?id=41241&r=submittedtwice
register_globals: http://bugs.php.net/fix.php?id=41241&r=globals
PHP 3 support discontinued: http://bugs.php.net/fix.php?id=41241&r=php3
Daylight Savings: http://bugs.php.net/fix.php?id=41241&r=dst
IIS Stability:http://bugs.php.net/fix.php?id=41241&r=isapi
Install GNU Sed: http://bugs.php.net/fix.php?id=41241&r=gnused
Floating point limitations: http://bugs.php.net/fix.php?id=41241&r=float
No Zend Extensions: http://bugs.php.net/fix.php?id=41241&r=nozend
MySQL Configuration Error:http://bugs.php.net/fix.php?id=41241&r=mysqlcfg
#41229 [Opn->Fbk]: PHP as fastcgi module crashes
ID: 41229
Updated by: [EMAIL PROTECTED]
Reported By: admin at torrent dot lt
-Status: Open
+Status: Feedback
Bug Type: CGI related
Operating System: gentoo
PHP Version: 5.2.1
New Comment:
And no core files anywhere? find / -name 'core*' :)
Previous Comments:
[2007-04-29 23:21:32] admin at torrent dot lt
The ulimit thing was already done last Thursday, I have set these
variables as told. Nothing happened, the website is working fine, but
the traffic and the number of users are less than more ten times than in
a day.
"bin-path" => "/usr/bin/php-cgi", "max-procs" => 25, idle-timeout" =>
9,
"bin-environment" => ("PHP_FCGI_CHILDREN" => "1",
"PHP_FCGI_MAX_REQUESTS" => "10"
[2007-04-29 21:09:58] [EMAIL PROTECTED]
Start lighttpd without it forking. And only have 1 child for PHP
started. You don't need to "su lighttpd", just do the ulimit thing as
root.
[2007-04-29 20:02:39] admin at torrent dot lt
I have done the ulimit last week, but under uid0/gid0 I have set ulimit
-c unlimited under lighttpd user: su lighttpd ; ulimit -c unlimited,
what's more where I could expect the core file to be dumped? and how to
make it done?
I have just got the 500 error (after setting the ulimit) I reckon' that
before stating the 500 internal error the webserver is starting to act
extremely slow, the webpages hardly open's and so on. I have 1000
queries to mysql per sec.
Will be waiting for your answers.
I've got these errors.
http://p.defau.lt/?bNHoX3oyKfPWS80YDxIjAg
[2007-04-29 19:21:18] [EMAIL PROTECTED]
You need to come up with a gdb backtrace. Sometimes getting one is
tricky in this kind of configuration but if you can get a core file
generated...try starting lighttpd after setting "ulimit -c unlimited"
and you should get the core file.
[2007-04-29 18:55:52] admin at torrent dot lt
Description:
Greetings,
moving from debian PHP 5.2.0-8 (cgi-fcgi) (built: Dec 17 2006
19:57:41)
to gentoo PHP 5.1.6-pl11-gentoo (cgi-fcgi) (built: Apr 29 2007
12:43:38) as well as 5.2.1 produces me php segfaultion. Web server
(lighttpd) produces me an 500 error code, within only restart of it,
helps to solve a problem. Was concerning the bug of xcache (accelerator)
as well as trying to downgrade to earlier versions. After monitoring the
system for 2-3 days no answers were solved while php keept crashing with
or without additional extensions, and with one or other version (tried
PHP 4 too). Lastly, I am willing to get a help from you. Will be waiting
for your answer.
Reproduce code:
---
http://forum.lighttpd.net/topic/5981#7322
more:
2007-04-25 15:03:01: (mod_fastcgi.c.1711) connect failed: No such file
or directory on
unix:/var/run/lighttpd/lighttpd-fastcgi-php-315519.socket-17
Expected result:
examination of the bug
Actual result:
--
http://rafb.net/p/BegtLJ55.html
That's only what I could get.
--
Edit this bug report at http://bugs.php.net/?id=41229&edit=1
#41241 [Opn]: Realization array type. SQL 2003, S091
ID: 41241
User updated by: develar at gmail dot com
Reported By: develar at gmail dot com
Status: Open
Bug Type: PDO related
Operating System: Windows XP SP2
PHP Version: 5.2.1
New Comment:
"Actual result" and "Expected result" reverse, excuse me :).
Previous Comments:
[2007-04-30 13:01:28] develar at gmail dot com
Description:
SQL 2003, S091.
Reproduce code:
---
PostgreSQL 8.2
select proargnames from pg_catalog.pg_proc order by proargnames limit 1
Expected result:
array (
0 =>
array (
'proargnames' =>
'{"","","",id,uri,time_create,priority_base_time_create,title,time_modify,status,priority_base_title,type_name,leaf,version,author_name,author_surname,author_patronymic,privilege_edit,privilege_delete,privilege_grant}',
),
)
Actual result:
--
array (
0 =>
array (
'proargnames' =>
'{"","","",id,uri,time_create,priority_base_time_create,title,time_modify,array
(
0 =>
array (
'proargnames' =>
array (
0 => '',
1 => '',
2 => '',
3 => 'id',
4 => 'uri',
5 => 'time_create',
6 => 'priority_base_time_create',
7 => 'title',
8 => 'time_modify',
9 => 'status',
10 => 'priority_base_title',
11 => 'type_name',
12 => 'leaf',
13 => 'version',
14 => 'author_name',
15 => 'author_surname',
16 => 'author_patronymic',
17 => 'privilege_edit',
18 => 'privilege_delete',
19 => 'privilege_grant',
),
),
)}',
),
)
--
Edit this bug report at http://bugs.php.net/?id=41241&edit=1
#41241 [Opn->Fbk]: Realization array type. SQL 2003, S091
ID: 41241
Updated by: [EMAIL PROTECTED]
Reported By: develar at gmail dot com
-Status: Open
+Status: Feedback
Bug Type: PDO related
Operating System: Windows XP SP2
PHP Version: 5.2.1
New Comment:
Not enough information was provided for us to be able
to handle this bug. Please re-read the instructions at
http://bugs.php.net/how-to-report.php
If you can provide more information, feel free to add it
to this bug and change the status back to "Open".
Thank you for your interest in PHP.
Previous Comments:
[2007-04-30 13:06:53] develar at gmail dot com
"Actual result" and "Expected result" reverse, excuse me :).
[2007-04-30 13:01:28] develar at gmail dot com
Description:
SQL 2003, S091.
Reproduce code:
---
PostgreSQL 8.2
select proargnames from pg_catalog.pg_proc order by proargnames limit 1
Expected result:
array (
0 =>
array (
'proargnames' =>
'{"","","",id,uri,time_create,priority_base_time_create,title,time_modify,status,priority_base_title,type_name,leaf,version,author_name,author_surname,author_patronymic,privilege_edit,privilege_delete,privilege_grant}',
),
)
Actual result:
--
array (
0 =>
array (
'proargnames' =>
'{"","","",id,uri,time_create,priority_base_time_create,title,time_modify,array
(
0 =>
array (
'proargnames' =>
array (
0 => '',
1 => '',
2 => '',
3 => 'id',
4 => 'uri',
5 => 'time_create',
6 => 'priority_base_time_create',
7 => 'title',
8 => 'time_modify',
9 => 'status',
10 => 'priority_base_title',
11 => 'type_name',
12 => 'leaf',
13 => 'version',
14 => 'author_name',
15 => 'author_surname',
16 => 'author_patronymic',
17 => 'privilege_edit',
18 => 'privilege_delete',
19 => 'privilege_grant',
),
),
)}',
),
)
--
Edit this bug report at http://bugs.php.net/?id=41241&edit=1
#41237 [Opn->Fbk]: XOR problem
ID: 41237 Updated by: [EMAIL PROTECTED] Reported By: victorepand at gmail dot com -Status: Open +Status: Feedback Bug Type: Math related Operating System: Linux PHP Version: 5.2.1 New Comment: Please try using this CVS snapshot: http://snaps.php.net/php5.2-latest.tar.gz For Windows: http://snaps.php.net/win32/php5.2-win32-latest.zip Previous Comments: [2007-04-30 09:15:07] victorepand at gmail dot com Description: I am finding a difference between the same bitwise arithmetic from one server to the next when using PHP. What's more, this bitwise arithmetic is necessary for the PHP script to run, so as a result it will only function on one server, but not the other. Here is an example I am using to demonstrate this: if ((43814 ^ -4738698913)!=-443704711) print "incorrect result"; else print "correct result"; The (^) operator is an XOR bitwise arithmetic function as shown here: http://us2.php.net/manual/en/language.operators.bitwise.php and I am required to use numbers like the ones shown. On one server, I have tried both PHP 4.4.0 and PHP 5.1.0RC1 and the math works correctly for both (the correct answer as shown above is -443704711). But on another server, I have tried the same math with both PHP 4.4.6 and PHP 5.2.1, and it does not work correctly with either version of PHP! The result I get at that server is: -2147439834. I have no idea what could be the problem, but I can show you the PHP Info for both servers and perhaps you can detect what might be the difference? Here is the PHP Info for the server that works correctly using PHP 5.1.0RC1: http://www.buycellularphones.info/cron/special/info.php Here is the PHP Info for the other server using PHP 5.2.1 that does not work correctly: http://www.customdesignpostcards.com/cron/special/info.php Reproduce code: --- if ((43814 ^ -4738698913)!=-443704711) print "incorrect result"; else print "correct result"; Expected result: correct result Actual result: -- incorrect result -- Edit this bug report at http://bugs.php.net/?id=41237&edit=1
#41238 [Opn->Bgs]: PSpell fails to open dictionaries when server is under load
ID: 41238 Updated by: [EMAIL PROTECTED] Reported By: spam02 at pornel dot net -Status: Open +Status: Bogus Bug Type: Pspell related Operating System: Debian stable PHP Version: 5.2.1 New Comment: Pspell library problems should be reported to Pspell developers. Previous Comments: [2007-04-30 09:45:45] spam02 at pornel dot net Description: Problem reported in #33095 might be valid. I can reproduce the same problem, every time, on PHP/5.2.1, Linux/2.6.8, Apache/1.3.33, Aspell/0.60.3 It appears that PHP process which uses PSpell must be the only process on the server (even images loading concurently affect it.), otherwise PSpell fails to open dictionary. Failure seems to be determined per process, because PSpell will either work first time or always fail for a certain process (when you retry in a loop, even with a delay, you always get failure). I've tried different invocation methods, pspell_clear_session, clearstatcache, different dictionaries - nothing helps. Reproduce code: --- Try pspell on pages with and without images. Different browsers have different patterns for requesting images, so for example failure rate is higher in Firefox than Opera (I'm not mad, really). Actual result: -- Warning: pspell_new_config() [function.pspell-new-config]: PSPELL couldn't open the dictionary. reason: No word lists can be found for the language "en". in ... -- Edit this bug report at http://bugs.php.net/?id=41238&edit=1
#40943 [Opn->Bgs]: Function str_ireplace() crashes Apache server
ID: 40943 Updated by: [EMAIL PROTECTED] Reported By: dzoukr at dzoukr dot cz -Status: Open +Status: Bogus Bug Type: Reproducible crash Operating System: Windows XP SP2 PHP Version: 5.2.1 New Comment: >I am having this problem as well on php 5.2.1. I asked the original reporter to try the snapshot month ago, obviously you did not bother doing it, otherwise you would have discovered that the problem has been fixed months ago. Previous Comments: [2007-04-30 08:43:11] dzoukr at dzoukr dot cz Status changed to open. That`s all I can do wit it ;) [2007-04-29 21:21:54] qrf_coil at yahoo dot fr Good to hear, obviously it's a real bug.. could someone change the status of this bug ? [2007-04-19 20:40:56] registry at ivide dot com I am having this problem as well on php 5.2.1. It does not happen on another server I have running php 5.1.6. I had it happen trying to str_ireplace with an array of characters to be replaced with "". I am using XP Sp2, Apache 1.3.37, and Php stable 5.2.1. The apache logs show "zend_mm_heap corrupted" and the browser says "The connection to the server was interrupted". [2007-04-16 10:25:34] qrf_coil at yahoo dot fr I've experienced the same problem on a production server of a customer, quiet annoying, even using str_replace instead will correct the problem The server reported : ERROR The requested URL could not be retrieved While trying to retrieve the URL: The following error was encountered: * Zero Sized Reply Server used is a Unix with PHP 5.2.1-0.dotdeb.1 with Suhosin-Patch 0.9.6.2 (cli) (built: Feb 9 2007 04:46:32) [2007-04-06 01:00:00] php-bugs at lists dot php dot net No feedback was provided for this bug for over a week, so it is being suspended automatically. If you are able to provide the information that was originally requested, please do so and change the status of the bug back to "Open". 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/40943 -- Edit this bug report at http://bugs.php.net/?id=40943&edit=1
#41235 [Csd->Bgs]: ereg_replace does not replace all occurrences
ID: 41235 Updated by: [EMAIL PROTECTED] Reported By: alexey at gmail dot com -Status: Closed +Status: Bogus Bug Type: *Search functions Operating System: linux PHP Version: 4.4.6 Previous Comments: [2007-04-30 11:15:53] alexey at gmail dot com The PHP manual does not specify the replacement order. What if we replace it from right to the left? This will give us a different answer. Also, the definition of a regular expression does not impose the evaluation order. Therefore, the right answer is to replace each occurrence of & in the original string with a &. [2007-04-30 11:05:12] alexey at gmail dot com Oh well. Thanks for your explanation. Closed. [2007-04-30 09:33:18] robinv at gmail dot com Not a bug: The pattern matches '&qs=&q' and replaces it with '&qs=&q'. It matches '&si=0&q' and replaces it with '&si=0&q'. It matches '&nh=25' and replaces it with '&nh=25' and it matches '&lk=1&' replacing it with '&lk=1&'. Which is exactly what you've told it to do. [2007-04-30 06:47:34] alexey at gmail dot com Description: ereg_replace() does not replace all occurrences of a regular expression as the following example shows. The pattern &([^;][^;][^;][^;][^;]) is replaced with &\\1 in a string http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 The output is http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 which indicates for example that the last & was not replaced. Reproduce code: --- http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0";; $pattern="&([^;][^;][^;][^;][^;])"; $data=ereg_replace($pattern, "&\\1", $data); echo $data."\n"; // http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 $data=ereg_replace($pattern, "&\\1", $data); echo $data."\n"; // http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 ?> Expected result: http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 Actual result: -- http://search.networkworld.com/index.html?col=&qs=&qc=&si=0&ql=a&nh=25&lk=1&rf=0 -- Edit this bug report at http://bugs.php.net/?id=41235&edit=1
#41220 [Opn->Fbk]: serialize takes huge amount of memory
ID: 41220 Updated by: [EMAIL PROTECTED] Reported By: shshzi at gmail dot com -Status: Open +Status: Feedback -Bug Type: Reproducible crash +Bug Type: Unknown/Other Function Operating System: Linux i386 PHP Version: 5.2.1 New Comment: Not reproducible. Previous Comments: [2007-04-28 22:30:41] shshzi at gmail dot com Description: when using serialize on some large (not huge) array, it will take insane amounts of memory. look at the code , when using size array of 30,000 , the script works as expected, indicating that the output is a mere 2MB in length, while enlarging the array to 50,000 elements causes php to go crazy and hit the 600MB i've allowed him to consume... Reproduce code: --- -- Edit this bug report at http://bugs.php.net/?id=41220&edit=1
#41241 [Fbk->Opn]: Realization array type. SQL 2003, S091
ID: 41241
User updated by: develar at gmail dot com
Reported By: develar at gmail dot com
-Status: Feedback
+Status: Open
Bug Type: PDO related
Operating System: Windows XP SP2
PHP Version: 5.2.1
New Comment:
Pdo (pdo_pgsql) supports SQL 2003 T031 - BOOLEAN data type. If I select
value of a field, it will be returned with type BOOL.
I ask, that also was supported SQL 2003 S091 - ARRAY data type. If I
select proargnames with type text[], I expect, that PDO will return to
me array, instead of string.
Previous Comments:
[2007-04-30 13:08:34] [EMAIL PROTECTED]
Not enough information was provided for us to be able
to handle this bug. Please re-read the instructions at
http://bugs.php.net/how-to-report.php
If you can provide more information, feel free to add it
to this bug and change the status back to "Open".
Thank you for your interest in PHP.
[2007-04-30 13:06:53] develar at gmail dot com
"Actual result" and "Expected result" reverse, excuse me :).
[2007-04-30 13:01:28] develar at gmail dot com
Description:
SQL 2003, S091.
Reproduce code:
---
PostgreSQL 8.2
select proargnames from pg_catalog.pg_proc order by proargnames limit 1
Expected result:
array (
0 =>
array (
'proargnames' =>
'{"","","",id,uri,time_create,priority_base_time_create,title,time_modify,status,priority_base_title,type_name,leaf,version,author_name,author_surname,author_patronymic,privilege_edit,privilege_delete,privilege_grant}',
),
)
Actual result:
--
array (
0 =>
array (
'proargnames' =>
'{"","","",id,uri,time_create,priority_base_time_create,title,time_modify,array
(
0 =>
array (
'proargnames' =>
array (
0 => '',
1 => '',
2 => '',
3 => 'id',
4 => 'uri',
5 => 'time_create',
6 => 'priority_base_time_create',
7 => 'title',
8 => 'time_modify',
9 => 'status',
10 => 'priority_base_title',
11 => 'type_name',
12 => 'leaf',
13 => 'version',
14 => 'author_name',
15 => 'author_surname',
16 => 'author_patronymic',
17 => 'privilege_edit',
18 => 'privilege_delete',
19 => 'privilege_grant',
),
),
)}',
),
)
--
Edit this bug report at http://bugs.php.net/?id=41241&edit=1
#41225 [Csd->Bgs]: installation problem
ID: 41225 Updated by: [EMAIL PROTECTED] Reported By: t_ravilly at hotmail dot com -Status: Closed +Status: Bogus Bug Type: *General Issues Operating System: windows XP SP2 PHP Version: 5.2.1 Previous Comments: [2007-04-29 19:11:47] t_ravilly at hotmail dot com i sould have wrongly download the installer. i redownloaded it and now it's ok. sorry. and thanks you for all you do !! [2007-04-29 13:38:55] t_ravilly at hotmail dot com Description: during copying files : the Cabinet File "php.cab" requiered for this installation is corrupt and cannot be used. i tried an installation with apache, and tried without... (it was the msi file on your download page) sorry for my english Expected result: to have an installation pack who work !! ;) Actual result: -- no possible installation -- Edit this bug report at http://bugs.php.net/?id=41225&edit=1
#41230 [Csd->Bgs]: session written AFTER global variables have been destroyed
ID: 41230
Updated by: [EMAIL PROTECTED]
Reported By: amirlaher at yahoo dot co dot uk
-Status: Closed
+Status: Bogus
Bug Type: Session related
Operating System: Debian Sarge,Ubuntu
PHP Version: 5.2.1
Previous Comments:
[2007-04-29 21:07:01] amirlaher at yahoo dot co dot uk
I see this is now a 'feature' - see 'warning' in the
session_write_close() PHP manual page.
[2007-04-29 20:31:15] amirlaher at yahoo dot co dot uk
Description:
This issue affects custom session handlers (as defined with
session_set_save_handler()), in which the write function attempts to
access a *global object*.
When upgrading a custom session handler (calling a perl daemon), from
php5.1 to php5.2, the session was never written owing to a "Call to a
member function xxx() on a non-object" error.
This seemingly occurs because global objects get destructed *before*
session_write_close() is called, during php's shutdown sequence.
I worked around the issue using
register_shutdown_function("session_write_close"); near the top of the
script.
register_shutdown_function() does take effect when a script is
terminated using exit(); I would have to call "session_write_close();"
after having finished writing to the session.
This error also still springs up sometimes when the memory limit is
reached.
This behaviour-change was painful but could be an intentional change by
the php team (?) I am reporting it because it may be an unintentional
change, and I have seen that it has also affected other users including
Drupal users
(e.g. this Zend Forums thread:
http://www.zend.com/forums/index.php?t=msg&th=3464
and Drupal: http://drupal.org/node/92802)
Cheers
Amir
Reproduce code:
---
savePath;
}
function setSavePath($savePath) {
$this->savePath= $savePath;
}
function getFilePath() {
return $this->getSavePath().'/sess_'.$this->getId();
}
function getId() {
return $this->id;
}
function setId($id) {
$this->id= $id;
}
}
function open($savePath, $session_name)
{
global $sess;
$sess= new sess;
$sess->setSavePath('/tmp');
return(true);
}
function close()
{
return(true);
}
function read($id)
{
global $sess;
$sess->setId($id);
if(file_exists($sess->getFilePath())) {
return (string) file_get_contents($sess->getFilePath());
}
}
function write($id, $sess_data)
{
global $sess;
if ($fp = fopen($sess->getFilePath(), "w")) {
$return = fwrite($fp, $sess_data);
fclose($fp);
return $return;
} else {
return(false);
}
}
function destroy($id){
global $sess;
return(unlink($sess->getFilePath()));
}
function gc($maxlifetime)
{
global $sess;
foreach (glob($sess->getSavePath()."/sess_*") as $filename) {
if (filemtime($filename) + $maxlifetime < time()) {
unlink($filename);
}
}
return true;
}
session_set_save_handler("open", "close", "read", "write", "destroy",
"gc");
session_start();
print "session id : " . session_id() . "";
if(!$_SESSION['hi']) {
$_SESSION['hi']='ho';
} else {
$_SESSION['ts']=time();
}
print "session data: "; print_r($_SESSION);
print "\nclass data: \n "; print_r($sess);
Expected result:
session written ok
Actual result:
--
session id : 8b2596bea65174ce950bb47140edfe2a
session data:
Array
(
[hi] => ho
)
class data:
sess Object
(
[savePath] => /tmp
[id] => 8b2596bea65174ce950bb47140edfe2a
)
Fatal error: Call to a member function getFilePath() on a non-object
in /var/www/sesstest.php on line 52
--
Edit this bug report at http://bugs.php.net/?id=41230&edit=1
#41227 [Opn->Bgs]: mysql_real_escape_string
ID: 41227 Updated by: [EMAIL PROTECTED] Reported By: php_debugger at yahoo dot com -Status: Open +Status: Bogus Bug Type: MySQL related Operating System: Windows XP PHP Version: 5.2.1 New Comment: Sorry, but your problem does not imply a bug in PHP itself. For a list of more appropriate places to ask for help using PHP, please visit http://www.php.net/support.php as this bug system is not the appropriate forum for asking support questions. Due to the volume of reports we can not explain in detail here why your report is not a bug. The support channels will be able to provide an explanation for you. Thank you for your interest in PHP. Previous Comments: [2007-04-29 23:18:39] judas dot iscariote at gmail dot com You should use mysqli and the mysqli_set_charset() function. [2007-04-29 18:16:04] php_debugger at yahoo dot com Description: I have met this "bug" while I was creating a Russian site. I have a web form(ENCODING:UTF-8) and a simple query(My db uses UTF-8). When I submit the form, I get an error which says that the string is too long for the column but it is not the case(I just type the letters). After a while debugging I have found out what is the problem. When I use mysql_real_escape_string() I get the error else I do not. I have searched in the php site but I found nothing which has to do with encoding. Reproduce code: --- //I will give only the php code though it is very simple Expected result: No errors. Actual result: -- MySQL error which says that the string is too long. -- Edit this bug report at http://bugs.php.net/?id=41227&edit=1
#41232 [Opn->Fbk]: Eval failing on calls to class methods
ID: 41232 Updated by: [EMAIL PROTECTED] Reported By: Antoniocs at gmail dot com -Status: Open +Status: Feedback Bug Type: Output Control Operating System: Windows XP sp2 PHP Version: 5.2.1 New Comment: Thank you for this bug report. To properly diagnose the problem, we need a short but complete example script to be able to reproduce this bug ourselves. A proper reproducing script starts with , is max. 10-20 lines long and does not require any external resources such as databases, etc. If the script requires a database to demonstrate the issue, please make sure it creates all necessary tables, stored procedures etc. Please avoid embedding huge scripts into the report. Previous Comments: [2007-04-29 23:14:08] Antoniocs at gmail dot com Description: I am trying to use output control functions to replace text with processed pages. So I am using eval on a callback function Reproduce code: --- //page_parser.php file (small class to parse pages and replace with something set by me) URL: http://privatepaste.com/9818gOaFnI //teste_parser.php file (file that calls the class to use it to replace a string that I set with the contents of a php file) URL: http://privatepaste.com/32658Bx95B //teste_parser_2.php file (file that contains the php code to be processed by eval. Later this will replace the string set by me URL: http://privatepaste.com/0c1Ev1CTKA Expected result: The teste() method from the class bla to show it's echo (which is a simple Hello) Actual result: -- Nothing appears. The page is blank. All because I call a class method. -- Edit this bug report at http://bugs.php.net/?id=41232&edit=1
#41233 [Opn->Fbk]: "Per Directory Values" are not applied correctly
ID: 41233 Updated by: [EMAIL PROTECTED] Reported By: simon at bleif dot de -Status: Open +Status: Feedback Bug Type: IIS related Operating System: Windows Server 2003 x64 R2 SP2 PHP Version: 5.2.1 New Comment: Please try using this CVS snapshot: http://snaps.php.net/php5.2-latest.tar.gz For Windows: http://snaps.php.net/win32/php5.2-win32-latest.zip Previous Comments: [2007-04-30 02:50:33] simon at bleif dot de Description: According to http://www.php-center.de/en-html-manual/configuration.changes.html, you are able to change parts of the PHP-configuration on a per-directory-manner. After setting the desired configuration options there and restarting IIS, only the options for the first folder (if there are some "folders"/registry keys within the same "folder"/key each configured with individual options) are applied. A trace with Regmon shows the W3WP enumerating and (successfully) accessing all entries under the "Per Directory Values"-hive. Reproduce code: --- Setup at least 2 virtul hosts (located under C:\Inetpub\wwwroot\a and C:\Inetpub\wwwroot\b). Start Regedit, navigate to the key HKLM\Software\PHP (HKLM\Software\Wow6432Node\PHP if you are on x64), create the subkey "Per Directory Values". This key gets a new subkey named "c", that key gets one named "Inetpub", that one gets a subkey named "wwwroot". Now you create two keys (one named "a", the other one named "b") directly below the key "wwwroot". Now, create a new string value called for example "log_errors" in both of these keys and set that value to the opposite of the value defined in the php.ini (for example 0). Restart IIS, place a file named test.php in both physical virtual host directory containing the following: If you now launch your web browser, point it to virtual host A, the correct values (as they were set in the registry) are being displayed. If you do the same with host B, the default values (as set in the php.ini or a registry key above the virtual host B configuration-key) (or NULL) are beeing displayed, not the setting defined in the registry. If you now delete the key "a" in the registry and restart IIS, the correct value is displayed when you point your browser to virtual host B. If you create a third virtual host C (folder: c), the settings stored in the registry are applied only if there are no other hosts-settings defined before (in alphabetical order, starting with a) that host in the same registry-key. The same issue also appears on a second server (same OS, same versions, but different php.ini). The issue applies both in safe- and non-safe mode. Expected result: The settings are being applied correctly on a per folder-level. Actual result: -- The settings are only applied to the first (in alphabetical order) configured folder in a directory. Further details please see under "Reproduce code". -- Edit this bug report at http://bugs.php.net/?id=41233&edit=1
#41236 [Opn->Asn]: regression with SSL connections since fix for #39571
ID: 41236
Updated by: [EMAIL PROTECTED]
Reported By: judas dot iscariote at gmail dot com
-Status: Open
+Status: Assigned
Bug Type: Streams related
Operating System: linux 32/64bit, Win32
PHP Version: 5CVS-2007-04-30 (CVS)
-Assigned To:
+Assigned To: iliaa
Previous Comments:
[2007-04-30 07:20:11] judas dot iscariote at gmail dot com
Description:
Since PHP 5.2.1, our application that has jabber stopped working.all
previuos PHP version works fine, this regression is caused by the fix
for bug #39571,reverting xp_ssl.c to the version shipped with 5.2.0
solves the problem.
Reproduce code:
---
./configure --with-openssl
");
fwrite($j, "");
$data = '';
// Wait for a response until timeout is reached
$start = time();
$data = '';
$timeout = 10;
$wait = false;
do {
$read = trim(fread($j, 4096));
$data .= $read;
// ok.. this check aint part of the original code
// that breaks,but check if we timed out of reached eof..anyway..
$foo = stream_get_meta_data($j);
if(feof($j) || $foo['timed_out'] || $foo['eof']) break;
// yeah, we know aint nice, but works in **any** other PHP version.
} while (time() <= $start + $timeout && ($wait || $data == '' || $read
!= '' || (substr(rtrim($data), -1) != '>')));
fclose($j);
var_dump($data);
?>
Expected result:
response from the server var_dump'ed as in (all) previous versions
Actual result:
--
Fatal error: Maximum execution time of **30** seconds exceeded...an
xdebug session reveals that the script hang, in a **single** fread()
call
an strace shows something hangs permanently with EAGAIN signal..
(probably the loop in xp_ssl.c from line 398 to 410), note that checking
if has reached EOF or if the stream times out does not help either.
--
Edit this bug report at http://bugs.php.net/?id=41236&edit=1
#41241 [Opn->Fbk]: Realization array type. SQL 2003, S091
ID: 41241
Updated by: [EMAIL PROTECTED]
Reported By: develar at gmail dot com
-Status: Open
+Status: Feedback
Bug Type: PDO related
Operating System: Windows XP SP2
PHP Version: 5.2.1
New Comment:
Thank you for this bug report. To properly diagnose the problem, we
need a short but complete example script to be able to reproduce
this bug ourselves.
A proper reproducing script starts with ,
is max. 10-20 lines long and does not require any external
resources such as databases, etc. If the script requires a
database to demonstrate the issue, please make sure it creates
all necessary tables, stored procedures etc.
Please avoid embedding huge scripts into the report.
Try with a short reproduce code if you're unable to explain your
problem in English.
Previous Comments:
[2007-04-30 13:20:08] develar at gmail dot com
Pdo (pdo_pgsql) supports SQL 2003 T031 - BOOLEAN data type. If I select
value of a field, it will be returned with type BOOL.
I ask, that also was supported SQL 2003 S091 - ARRAY data type. If I
select proargnames with type text[], I expect, that PDO will return to
me array, instead of string.
[2007-04-30 13:08:34] [EMAIL PROTECTED]
Not enough information was provided for us to be able
to handle this bug. Please re-read the instructions at
http://bugs.php.net/how-to-report.php
If you can provide more information, feel free to add it
to this bug and change the status back to "Open".
Thank you for your interest in PHP.
[2007-04-30 13:06:53] develar at gmail dot com
"Actual result" and "Expected result" reverse, excuse me :).
[2007-04-30 13:01:28] develar at gmail dot com
Description:
SQL 2003, S091.
Reproduce code:
---
PostgreSQL 8.2
select proargnames from pg_catalog.pg_proc order by proargnames limit 1
Expected result:
array (
0 =>
array (
'proargnames' =>
'{"","","",id,uri,time_create,priority_base_time_create,title,time_modify,status,priority_base_title,type_name,leaf,version,author_name,author_surname,author_patronymic,privilege_edit,privilege_delete,privilege_grant}',
),
)
Actual result:
--
array (
0 =>
array (
'proargnames' =>
'{"","","",id,uri,time_create,priority_base_time_create,title,time_modify,array
(
0 =>
array (
'proargnames' =>
array (
0 => '',
1 => '',
2 => '',
3 => 'id',
4 => 'uri',
5 => 'time_create',
6 => 'priority_base_time_create',
7 => 'title',
8 => 'time_modify',
9 => 'status',
10 => 'priority_base_title',
11 => 'type_name',
12 => 'leaf',
13 => 'version',
14 => 'author_name',
15 => 'author_surname',
16 => 'author_patronymic',
17 => 'privilege_edit',
18 => 'privilege_delete',
19 => 'privilege_grant',
),
),
)}',
),
)
--
Edit this bug report at http://bugs.php.net/?id=41241&edit=1
#41234 [Bgs]: additional info
ID: 41234 User updated by: cpriest at warpmail dot net -Summary: Additional Notes Reported By: cpriest at warpmail dot net Status: Bogus Bug Type: MySQLi related Operating System: Windows Vista Business PHP Version: 5.2.1 New Comment: I've done a fresh install of both Apache and PHP twice. Default installation. I guess I'll try it again. No other reports of issues on Vista have shown up? Previous Comments: [2007-04-30 11:32:37] [EMAIL PROTECTED] You probably have another libmysql.dll or you have mixed up several PHP installs. The mysqli extension works fine. [2007-04-30 05:02:00] cpriest at warpmail dot net I have verified the following: - libmysql.dll is located in c:\php - c:\php is in the environment PATH variable (rebooted afterwards) - php_mysql.dll and php_mysqli.dll are both in .\ext and are being loaded (mysqli load fails) I have also tried the following: - Apache 2.2.x - Copying the libmysql.dll to c:\windows\system32 - Copying the libmysql.dll from my MySQL 4.1.x installation into c:\php All result in the same error "Procedure could not be found" while loading php_mysqli.dll I'm reasonably certain that this is related to my Windows Vista Business install, I have successfully installed this environment many times before on Windows XP and Linux installations with no problems. [2007-04-30 04:56:16] cpriest at warpmail dot net Description: Unable to get php_mysqli.dll to load properly on Windows Vista Business with Apache 2.0.59 or 2.2.x. Reproduce code: --- Install Apache 2.0.59 + PHP 5.2.1 Try loading the php_mysqli.dll Expected result: Able to load extension Actual result: -- Cannot load, get error "Procedure could not be found." -- Edit this bug report at http://bugs.php.net/?id=41234&edit=1
#41241 [Fbk->Opn]: Realization array type. SQL 2003, S091
ID: 41241
User updated by: develar at gmail dot com
Reported By: develar at gmail dot com
-Status: Feedback
+Status: Open
Bug Type: PDO related
Operating System: Windows XP SP2
PHP Version: 5.2.1
New Comment:
query('select proargnames from pg_catalog.pg_proc
order by proargnames limit 1')->fetchAll(PDO::FETCH_ASSOC));
?>
actual:
array (
0 =>
array (
'proargnames' => '{"",x,n}',
),
)
expected:
proargnames is array, but not string
array (
0 =>
array (
'proargnames' =>
array (
0 => '',
1 => 'x',
2 => 'n',
),
),
)
Previous Comments:
[2007-04-30 13:25:40] [EMAIL PROTECTED]
Thank you for this bug report. To properly diagnose the problem, we
need a short but complete example script to be able to reproduce
this bug ourselves.
A proper reproducing script starts with ,
is max. 10-20 lines long and does not require any external
resources such as databases, etc. If the script requires a
database to demonstrate the issue, please make sure it creates
all necessary tables, stored procedures etc.
Please avoid embedding huge scripts into the report.
Try with a short reproduce code if you're unable to explain your
problem in English.
[2007-04-30 13:20:08] develar at gmail dot com
Pdo (pdo_pgsql) supports SQL 2003 T031 - BOOLEAN data type. If I select
value of a field, it will be returned with type BOOL.
I ask, that also was supported SQL 2003 S091 - ARRAY data type. If I
select proargnames with type text[], I expect, that PDO will return to
me array, instead of string.
[2007-04-30 13:08:34] [EMAIL PROTECTED]
Not enough information was provided for us to be able
to handle this bug. Please re-read the instructions at
http://bugs.php.net/how-to-report.php
If you can provide more information, feel free to add it
to this bug and change the status back to "Open".
Thank you for your interest in PHP.
[2007-04-30 13:06:53] develar at gmail dot com
"Actual result" and "Expected result" reverse, excuse me :).
[2007-04-30 13:01:28] develar at gmail dot com
Description:
SQL 2003, S091.
Reproduce code:
---
PostgreSQL 8.2
select proargnames from pg_catalog.pg_proc order by proargnames limit 1
Expected result:
array (
0 =>
array (
'proargnames' =>
'{"","","",id,uri,time_create,priority_base_time_create,title,time_modify,status,priority_base_title,type_name,leaf,version,author_name,author_surname,author_patronymic,privilege_edit,privilege_delete,privilege_grant}',
),
)
Actual result:
--
array (
0 =>
array (
'proargnames' =>
'{"","","",id,uri,time_create,priority_base_time_create,title,time_modify,array
(
0 =>
array (
'proargnames' =>
array (
0 => '',
1 => '',
2 => '',
3 => 'id',
4 => 'uri',
5 => 'time_create',
6 => 'priority_base_time_create',
7 => 'title',
8 => 'time_modify',
9 => 'status',
10 => 'priority_base_title',
11 => 'type_name',
12 => 'leaf',
13 => 'version',
14 => 'author_name',
15 => 'author_surname',
16 => 'author_patronymic',
17 => 'privilege_edit',
18 => 'privilege_delete',
19 => 'privilege_grant',
),
),
)}',
),
)
--
Edit this bug report at http://bugs.php.net/?id=41241&edit=1
#41232 [Fbk->Opn]: Eval failing on calls to class methods
ID: 41232 User updated by: Antoniocs at gmail dot com Reported By: Antoniocs at gmail dot com -Status: Feedback +Status: Open Bug Type: Output Control Operating System: Windows XP sp2 PHP Version: 5.2.1 New Comment: I set all the code you would need in the paste site. Just run the teste_parser.php file and it will call the rest Previous Comments: [2007-04-30 13:22:18] [EMAIL PROTECTED] Thank you for this bug report. To properly diagnose the problem, we need a short but complete example script to be able to reproduce this bug ourselves. A proper reproducing script starts with , is max. 10-20 lines long and does not require any external resources such as databases, etc. If the script requires a database to demonstrate the issue, please make sure it creates all necessary tables, stored procedures etc. Please avoid embedding huge scripts into the report. [2007-04-29 23:14:08] Antoniocs at gmail dot com Description: I am trying to use output control functions to replace text with processed pages. So I am using eval on a callback function Reproduce code: --- //page_parser.php file (small class to parse pages and replace with something set by me) URL: http://privatepaste.com/9818gOaFnI //teste_parser.php file (file that calls the class to use it to replace a string that I set with the contents of a php file) URL: http://privatepaste.com/32658Bx95B //teste_parser_2.php file (file that contains the php code to be processed by eval. Later this will replace the string set by me URL: http://privatepaste.com/0c1Ev1CTKA Expected result: The teste() method from the class bla to show it's echo (which is a simple Hello) Actual result: -- Nothing appears. The page is blank. All because I call a class method. -- Edit this bug report at http://bugs.php.net/?id=41232&edit=1
#41190 [Opn]: Two proposed patches for sybase_ct module
ID: 41190
User updated by: nick at marden dot org
Reported By: nick at marden dot org
Status: Open
Bug Type: Feature/Change Request
Operating System: Linux
PHP Version: 5.2.1
Assigned To: thekid
New Comment:
The point of the second patch (specifically the {HOSTNAME} part) isn't
to be a substitute for a gethostname() function - although it would be
nice to have one.
The point is to have the sybct.hostname default to the local hostname
*without any user intervention*, something that is not currently the
case.
The reason I added the {HOSTNAME} support is that in our production
environment we often have multiple servers running PHP and it's nice to
know which one is connecting to Sybase in the case of long-running
queries, etc. So by putting
{HOSTNAME}-serverA
or
{HOSTNAME}-serverB
I give myself a shorthand to determine which process is causing trouble
on the Sybase database.
Previous Comments:
[2007-04-30 06:56:59] [EMAIL PROTECTED]
Assigning to myself, I like the first patch but am not sure about the
second - I'd rather like to see a gethostname() function in PHP with
which one could accomplish this in userland.
[2007-04-25 20:11:45] nick at marden dot org
Description:
I have two patches that I would like to propose for the PHP5 Sybase-CT
module.
The first adds the (very useful) functions sybase_output_params() and
sybase_return_status() functions. The patch is here:
http://www.marden.org/php-sybase-ct/php5-sybase_ct.return_status-and-output_params.patch
and the documentation is here:
http://www.marden.org/php-sybase-ct/return_status-and-output_params.txt
The second patch corrects (IMO) the behavior of sybct.hostname to
default to the hostname of the current system. It also supports
{HOSTNAME} as a substitution variable in the sybct.hostname parameter, a
feature that I have found useful in our server farm. The patch is here:
http://www.marden.org/php-sybase-ct/php-sybase_ct.hostname.patch
--
Edit this bug report at http://bugs.php.net/?id=41190&edit=1
#41241 [Opn]: Realization array type. SQL 2003, S091
ID: 41241
Updated by: [EMAIL PROTECTED]
Reported By: develar at gmail dot com
Status: Open
-Bug Type: PDO related
+Bug Type: Feature/Change Request
Operating System: Windows XP SP2
PHP Version: 5.2.1
New Comment:
Reclassified as feature request.
Previous Comments:
[2007-04-30 14:31:05] develar at gmail dot com
query('select proargnames from pg_catalog.pg_proc
order by proargnames limit 1')->fetchAll(PDO::FETCH_ASSOC));
?>
actual:
array (
0 =>
array (
'proargnames' => '{"",x,n}',
),
)
expected:
proargnames is array, but not string
array (
0 =>
array (
'proargnames' =>
array (
0 => '',
1 => 'x',
2 => 'n',
),
),
)
[2007-04-30 13:25:40] [EMAIL PROTECTED]
Thank you for this bug report. To properly diagnose the problem, we
need a short but complete example script to be able to reproduce
this bug ourselves.
A proper reproducing script starts with ,
is max. 10-20 lines long and does not require any external
resources such as databases, etc. If the script requires a
database to demonstrate the issue, please make sure it creates
all necessary tables, stored procedures etc.
Please avoid embedding huge scripts into the report.
Try with a short reproduce code if you're unable to explain your
problem in English.
[2007-04-30 13:20:08] develar at gmail dot com
Pdo (pdo_pgsql) supports SQL 2003 T031 - BOOLEAN data type. If I select
value of a field, it will be returned with type BOOL.
I ask, that also was supported SQL 2003 S091 - ARRAY data type. If I
select proargnames with type text[], I expect, that PDO will return to
me array, instead of string.
[2007-04-30 13:08:34] [EMAIL PROTECTED]
Not enough information was provided for us to be able
to handle this bug. Please re-read the instructions at
http://bugs.php.net/how-to-report.php
If you can provide more information, feel free to add it
to this bug and change the status back to "Open".
Thank you for your interest in PHP.
[2007-04-30 13:06:53] develar at gmail dot com
"Actual result" and "Expected result" reverse, excuse me :).
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/41241
--
Edit this bug report at http://bugs.php.net/?id=41241&edit=1
#41242 [NEW]: Array values destroyed after using as reference in function call
From: laurynas at by dot lt Operating system: Linux 2.6.17.7 PHP version: 5.2.1 PHP Bug Type: Arrays related Bug description: Array values destroyed after using as reference in function call Description: Array values are destroyed after using it as reference in function call defining in call dirrectly as array. If given array is constructed before function call in the same parameters line, PHP returns nothing, but really shoud return a value associated to array. Reproduce code: --- Expected result: Array ( [0] => some value ) Actual result: -- Array ( ) -- Edit bug report at http://bugs.php.net/?id=41242&edit=1 -- Try a CVS snapshot (PHP 4.4): http://bugs.php.net/fix.php?id=41242&r=trysnapshot44 Try a CVS snapshot (PHP 5.2): http://bugs.php.net/fix.php?id=41242&r=trysnapshot52 Try a CVS snapshot (PHP 6.0): http://bugs.php.net/fix.php?id=41242&r=trysnapshot60 Fixed in CVS: http://bugs.php.net/fix.php?id=41242&r=fixedcvs Fixed in release: http://bugs.php.net/fix.php?id=41242&r=alreadyfixed Need backtrace: http://bugs.php.net/fix.php?id=41242&r=needtrace Need Reproduce Script:http://bugs.php.net/fix.php?id=41242&r=needscript Try newer version:http://bugs.php.net/fix.php?id=41242&r=oldversion Not developer issue: http://bugs.php.net/fix.php?id=41242&r=support Expected behavior:http://bugs.php.net/fix.php?id=41242&r=notwrong Not enough info: http://bugs.php.net/fix.php?id=41242&r=notenoughinfo Submitted twice: http://bugs.php.net/fix.php?id=41242&r=submittedtwice register_globals: http://bugs.php.net/fix.php?id=41242&r=globals PHP 3 support discontinued: http://bugs.php.net/fix.php?id=41242&r=php3 Daylight Savings: http://bugs.php.net/fix.php?id=41242&r=dst IIS Stability:http://bugs.php.net/fix.php?id=41242&r=isapi Install GNU Sed: http://bugs.php.net/fix.php?id=41242&r=gnused Floating point limitations: http://bugs.php.net/fix.php?id=41242&r=float No Zend Extensions: http://bugs.php.net/fix.php?id=41242&r=nozend MySQL Configuration Error:http://bugs.php.net/fix.php?id=41242&r=mysqlcfg
#41232 [Opn->Fbk]: Eval failing on calls to class methods
ID: 41232 Updated by: [EMAIL PROTECTED] Reported By: Antoniocs at gmail dot com -Status: Open +Status: Feedback Bug Type: Output Control Operating System: Windows XP sp2 PHP Version: 5.2.1 New Comment: Please provide a SHORT but complete reproduce case in ONE short script. I believe HTML or several scripts are not required for this. Thank you. Previous Comments: [2007-04-30 14:44:01] Antoniocs at gmail dot com I set all the code you would need in the paste site. Just run the teste_parser.php file and it will call the rest [2007-04-30 13:22:18] [EMAIL PROTECTED] Thank you for this bug report. To properly diagnose the problem, we need a short but complete example script to be able to reproduce this bug ourselves. A proper reproducing script starts with , is max. 10-20 lines long and does not require any external resources such as databases, etc. If the script requires a database to demonstrate the issue, please make sure it creates all necessary tables, stored procedures etc. Please avoid embedding huge scripts into the report. [2007-04-29 23:14:08] Antoniocs at gmail dot com Description: I am trying to use output control functions to replace text with processed pages. So I am using eval on a callback function Reproduce code: --- //page_parser.php file (small class to parse pages and replace with something set by me) URL: http://privatepaste.com/9818gOaFnI //teste_parser.php file (file that calls the class to use it to replace a string that I set with the contents of a php file) URL: http://privatepaste.com/32658Bx95B //teste_parser_2.php file (file that contains the php code to be processed by eval. Later this will replace the string set by me URL: http://privatepaste.com/0c1Ev1CTKA Expected result: The teste() method from the class bla to show it's echo (which is a simple Hello) Actual result: -- Nothing appears. The page is blank. All because I call a class method. -- Edit this bug report at http://bugs.php.net/?id=41232&edit=1
#41242 [Opn->Bgs]: Array values destroyed after using as reference in function call
ID: 41242 Updated by: [EMAIL PROTECTED] Reported By: laurynas at by dot lt -Status: Open +Status: Bogus Bug Type: Arrays related Operating System: Linux 2.6.17.7 PHP Version: 5.2.1 New Comment: "$user = array()" is not a variable, but an expression. Previous Comments: [2007-04-30 16:04:03] laurynas at by dot lt Description: Array values are destroyed after using it as reference in function call defining in call dirrectly as array. If given array is constructed before function call in the same parameters line, PHP returns nothing, but really shoud return a value associated to array. Reproduce code: --- Expected result: Array ( [0] => some value ) Actual result: -- Array ( ) -- Edit this bug report at http://bugs.php.net/?id=41242&edit=1
#41232 [Fbk->Opn]: Eval failing on calls to class methods
ID: 41232 User updated by: Antoniocs at gmail dot com Reported By: Antoniocs at gmail dot com -Status: Feedback +Status: Open Bug Type: Output Control Operating System: Windows XP sp2 PHP Version: 5.2.1 New Comment: Here are the source files: http://www.antoniocs.org/eval_test.rar Unpack, and run teste_parser.php Previous Comments: [2007-04-30 16:05:05] [EMAIL PROTECTED] Please provide a SHORT but complete reproduce case in ONE short script. I believe HTML or several scripts are not required for this. Thank you. [2007-04-30 14:44:01] Antoniocs at gmail dot com I set all the code you would need in the paste site. Just run the teste_parser.php file and it will call the rest [2007-04-30 13:22:18] [EMAIL PROTECTED] Thank you for this bug report. To properly diagnose the problem, we need a short but complete example script to be able to reproduce this bug ourselves. A proper reproducing script starts with , is max. 10-20 lines long and does not require any external resources such as databases, etc. If the script requires a database to demonstrate the issue, please make sure it creates all necessary tables, stored procedures etc. Please avoid embedding huge scripts into the report. [2007-04-29 23:14:08] Antoniocs at gmail dot com Description: I am trying to use output control functions to replace text with processed pages. So I am using eval on a callback function Reproduce code: --- //page_parser.php file (small class to parse pages and replace with something set by me) URL: http://privatepaste.com/9818gOaFnI //teste_parser.php file (file that calls the class to use it to replace a string that I set with the contents of a php file) URL: http://privatepaste.com/32658Bx95B //teste_parser_2.php file (file that contains the php code to be processed by eval. Later this will replace the string set by me URL: http://privatepaste.com/0c1Ev1CTKA Expected result: The teste() method from the class bla to show it's echo (which is a simple Hello) Actual result: -- Nothing appears. The page is blank. All because I call a class method. -- Edit this bug report at http://bugs.php.net/?id=41232&edit=1
#39946 [Opn->Bgs]: Native Singleton Implementation
ID: 39946
Updated by: [EMAIL PROTECTED]
Reported By: scott dot mcnaught at synergy8 dot com
-Status: Open
+Status: Bogus
Bug Type: Feature/Change Request
Operating System: All
PHP Version: 6CVS-2006-12-25 (CVS)
New Comment:
We don't have any plans on implementing this in the nearest future.
Previous Comments:
[2007-03-06 05:51:30] toomuchphp-phpbugs at yahoo dot com
Perhaps some nicer solutions to your got-cha's:
- What happens when $members = new members() is called
If the class is defined as a singleton, then PHP should create the
instance as
normal (this includes calling __construct() ) and store it in the hash
table,
or return the cached instance if it's already created.
-- I say an exception being thrown and a suggestion to remove "new"
You should still be able to use 'new', with the expectation that if the
class
is a singleton, you might actually get a cached instance, rather than a
'new'
instance, but it's not very difficult to understand:
$a = new members;
$b = new members;
$c = new members;
// $a $b and $c are all the same instance because members is a
singleton
If you really think 'new' is too ambiguous, why not add a new token
such as
'only', which is used for instantiation of singletons?
$a = only members;
$b = only members;
$c = only members;
// it's a bit more obvious $a $b and $c are the same instance.
- Is it ok to go $members = members; ?
That's too ambiguous. In that example 'members' could be a constant,
singleton, or simply the string 'members'.
- Singleton constructors should not have parameters
Perhaps they could (and it might be useful), if it's done like this:
singleton class members { ... }
$foo = new members(1, 2, 3); // arguments are passed to __construct()
$bar = new members(3, 2, 1); // arguments are ignored, object has
already been constructed
$baz = new members(4, 5, 6); // arguments are ignored, object has
already been constructed
- Perhaps a new function singleton_getobjects needs to be added to
enable access to the internal hash table
Not a bad idea. Getting in 'through the back door' is really important.
Why not
also a 'singleton_is_loaded()', and an 'is_singleton()'? Obviously
this would
eventually need to be visible through the Reflection classes (etc,
etc).
- Never have singleton abstract classes?
I think that would be an unfair limitation.
You should just need to:
- add T_SINGLETON to the parser.
- add an is_singleton flag for class definitions, activated by
T_SINGLETON.
- overload the 'new' operator to retrieve from the hash table if the
class is
a singleton (otherwise call __construct() as normal).
- implement some of those helper functions such as
singleton_getobjects().
[2006-12-29 08:23:57] scott dot mcnaught at synergy8 dot com
But the term singleton and static class are two different things.
The problem with pure static classes is that you can't inherit because
an instance is never created.
With static classes, you arent invoking methods on an instance of a
class. You are simply calling static functions in a static namespace.
[2006-12-29 07:59:22] me at bashkim dot com dot au
I think the term "singleton" itself is a little confusing for newbies
(considering that it's who you want to target here). Perhaps what you're
looking for is a "static class" ...
[2006-12-25 17:45:13] scott dot mcnaught at synergy8 dot com
Description:
###
# Introduction
###
This document is an RFC for adding a small patch to the zend engine to
allow for native singleton classes. The problem
is that currently theres no clean way to implement singleton classes in
user-land over the entirety of a project.
Singleton classes are beneficial because:
- Removes the overhead of having multiple instances of the one object
when there is no need
- Allows you to keep the objects state rather than always starting from
an initial state.
- They provide namespaces with the benefits of polymorphism (eg -
singleton classes can override / inherit from each other)
Throughout this document, I will use an example of a singleton class
"members" which acts as an interface to a database table.
This class can save and load members from this database table simply by
calling the following methods in this class.
members::get($member_id) Loads a member from a member id and returns an
associative array with info about that member
members::save($member) Saves a member to the database from an array of
properties about that member
With the recent phase of tiered and service oriented architecture, the
need for Singleton has become more and more apparent.
###
# Singleton in php5
#41242 [Bgs]: Array values destroyed after using as reference in function call
ID: 41242 User updated by: laurynas at by dot lt Reported By: laurynas at by dot lt Status: Bogus Bug Type: Arrays related Operating System: Linux 2.6.17.7 PHP Version: 5.2.1 New Comment: Agree, this is expresion, but the expression is executed before function is called, right? So the variable should be defined *before* function takes variable as a parameter. Then why it works in PHP 4.x? Previous Comments: [2007-04-30 16:06:22] [EMAIL PROTECTED] "$user = array()" is not a variable, but an expression. [2007-04-30 16:04:03] laurynas at by dot lt Description: Array values are destroyed after using it as reference in function call defining in call dirrectly as array. If given array is constructed before function call in the same parameters line, PHP returns nothing, but really shoud return a value associated to array. Reproduce code: --- Expected result: Array ( [0] => some value ) Actual result: -- Array ( ) -- Edit this bug report at http://bugs.php.net/?id=41242&edit=1
#41232 [Opn->Fbk]: Eval failing on calls to class methods
ID: 41232 Updated by: [EMAIL PROTECTED] Reported By: Antoniocs at gmail dot com -Status: Open +Status: Feedback Bug Type: Output Control Operating System: Windows XP sp2 PHP Version: 5.2.1 New Comment: Hey!! Tudo bem?? OLAS Notice: Use of undefined constant TESTE - assumed 'TESTE' in /tmp/eval_test/page_parser.php(27) : eval()'d code on line 3 TESTE iojdsfoijdoidsd1177951395 Notice: Undefined variable: BLA in /tmp/eval_test/page_parser.php(27) : eval()'d code on line 12 Hello hello Previous Comments: [2007-04-30 16:06:43] Antoniocs at gmail dot com Here are the source files: http://www.antoniocs.org/eval_test.rar Unpack, and run teste_parser.php [2007-04-30 16:05:05] [EMAIL PROTECTED] Please provide a SHORT but complete reproduce case in ONE short script. I believe HTML or several scripts are not required for this. Thank you. [2007-04-30 14:44:01] Antoniocs at gmail dot com I set all the code you would need in the paste site. Just run the teste_parser.php file and it will call the rest [2007-04-30 13:22:18] [EMAIL PROTECTED] Thank you for this bug report. To properly diagnose the problem, we need a short but complete example script to be able to reproduce this bug ourselves. A proper reproducing script starts with , is max. 10-20 lines long and does not require any external resources such as databases, etc. If the script requires a database to demonstrate the issue, please make sure it creates all necessary tables, stored procedures etc. Please avoid embedding huge scripts into the report. [2007-04-29 23:14:08] Antoniocs at gmail dot com Description: I am trying to use output control functions to replace text with processed pages. So I am using eval on a callback function Reproduce code: --- //page_parser.php file (small class to parse pages and replace with something set by me) URL: http://privatepaste.com/9818gOaFnI //teste_parser.php file (file that calls the class to use it to replace a string that I set with the contents of a php file) URL: http://privatepaste.com/32658Bx95B //teste_parser_2.php file (file that contains the php code to be processed by eval. Later this will replace the string set by me URL: http://privatepaste.com/0c1Ev1CTKA Expected result: The teste() method from the class bla to show it's echo (which is a simple Hello) Actual result: -- Nothing appears. The page is blank. All because I call a class method. -- Edit this bug report at http://bugs.php.net/?id=41232&edit=1
#41242 [Bgs]: Array values destroyed after using as reference in function call
ID: 41242 Updated by: [EMAIL PROTECTED] Reported By: laurynas at by dot lt Status: Bogus Bug Type: Arrays related Operating System: Linux 2.6.17.7 PHP Version: 5.2.1 New Comment: No, the behavior is undefined (and currently executed after the function call - but that doesn't make it defined - might change anytime) Previous Comments: [2007-04-30 16:13:03] laurynas at by dot lt Agree, this is expresion, but the expression is executed before function is called, right? So the variable should be defined *before* function takes variable as a parameter. Then why it works in PHP 4.x? [2007-04-30 16:06:22] [EMAIL PROTECTED] "$user = array()" is not a variable, but an expression. [2007-04-30 16:04:03] laurynas at by dot lt Description: Array values are destroyed after using it as reference in function call defining in call dirrectly as array. If given array is constructed before function call in the same parameters line, PHP returns nothing, but really shoud return a value associated to array. Reproduce code: --- Expected result: Array ( [0] => some value ) Actual result: -- Array ( ) -- Edit this bug report at http://bugs.php.net/?id=41242&edit=1
#41242 [Bgs]: Array values destroyed after using as reference in function call
ID: 41242 User updated by: laurynas at by dot lt Reported By: laurynas at by dot lt Status: Bogus Bug Type: Arrays related Operating System: Linux 2.6.17.7 PHP Version: 5.2.1 New Comment: Logicaly, this should be fixed or defined anyway, because this is quite often situation by many developers. C/C# syntax executes these expressions *before* function call. Previous Comments: [2007-04-30 16:52:23] [EMAIL PROTECTED] No, the behavior is undefined (and currently executed after the function call - but that doesn't make it defined - might change anytime) [2007-04-30 16:13:03] laurynas at by dot lt Agree, this is expresion, but the expression is executed before function is called, right? So the variable should be defined *before* function takes variable as a parameter. Then why it works in PHP 4.x? [2007-04-30 16:06:22] [EMAIL PROTECTED] "$user = array()" is not a variable, but an expression. [2007-04-30 16:04:03] laurynas at by dot lt Description: Array values are destroyed after using it as reference in function call defining in call dirrectly as array. If given array is constructed before function call in the same parameters line, PHP returns nothing, but really shoud return a value associated to array. Reproduce code: --- Expected result: Array ( [0] => some value ) Actual result: -- Array ( ) -- Edit this bug report at http://bugs.php.net/?id=41242&edit=1
#41244 [NEW]: Multiple Queries in MSSQL Kills Connection and Prevents New Connections
From: mallinger at tc dot edu
Operating system: Windows 2003 Server
PHP version: 5.2.1
PHP Bug Type: MSSQL related
Bug description: Multiple Queries in MSSQL Kills Connection and Prevents New
Connections
Description:
When executing multiple queries from a single mssql_query call, the
connection is lost and subsequent connections cannot be created. There are
similar problems documented for PHP4, but it seems that (a) this is not
documented for PHP5; and (b) the inability to create subsequent connections
is not documented. In particular, it seems the mssql_select_db is what is
failing in the second connection.
Reproduce code:
---
$conn = mssql_connect('host', 'user', 'pass');
mssql_select_db("db", $conn);
mssql_query("QUERY 1", $conn);
mssql_query("QUERY 2; QUERY 3; QUERY 4;", $conn);
// this will fail:
mssql_query("QUERY 5", $conn);
// this is ok
mssql_close($conn);
// everything below will fail too:
$conn2 = mssql_connect('host', 'user', 'pass');
mssql_select_db("db", $conn2);
mssql_query("QUERY 6", $conn2);
Expected result:
The queries should execute with no problems.
Actual result:
--
Everything after the multiple query mssql_query call ("QUERY 2; QUERY 3;
QUERY 4;" will fail. New connections fail when trying to select a database
("mssql_select_db() [function.mssql-select-db]: Unable to select
database:...") and the existing connection cannot be used anymore
("mssql_query() [function.mssql-query]: Query failed in...") .
--
Edit bug report at http://bugs.php.net/?id=41244&edit=1
--
Try a CVS snapshot (PHP 4.4):
http://bugs.php.net/fix.php?id=41244&r=trysnapshot44
Try a CVS snapshot (PHP 5.2):
http://bugs.php.net/fix.php?id=41244&r=trysnapshot52
Try a CVS snapshot (PHP 6.0):
http://bugs.php.net/fix.php?id=41244&r=trysnapshot60
Fixed in CVS: http://bugs.php.net/fix.php?id=41244&r=fixedcvs
Fixed in release:
http://bugs.php.net/fix.php?id=41244&r=alreadyfixed
Need backtrace: http://bugs.php.net/fix.php?id=41244&r=needtrace
Need Reproduce Script:http://bugs.php.net/fix.php?id=41244&r=needscript
Try newer version:http://bugs.php.net/fix.php?id=41244&r=oldversion
Not developer issue: http://bugs.php.net/fix.php?id=41244&r=support
Expected behavior:http://bugs.php.net/fix.php?id=41244&r=notwrong
Not enough info:
http://bugs.php.net/fix.php?id=41244&r=notenoughinfo
Submitted twice:
http://bugs.php.net/fix.php?id=41244&r=submittedtwice
register_globals: http://bugs.php.net/fix.php?id=41244&r=globals
PHP 3 support discontinued: http://bugs.php.net/fix.php?id=41244&r=php3
Daylight Savings: http://bugs.php.net/fix.php?id=41244&r=dst
IIS Stability:http://bugs.php.net/fix.php?id=41244&r=isapi
Install GNU Sed: http://bugs.php.net/fix.php?id=41244&r=gnused
Floating point limitations: http://bugs.php.net/fix.php?id=41244&r=float
No Zend Extensions: http://bugs.php.net/fix.php?id=41244&r=nozend
MySQL Configuration Error:http://bugs.php.net/fix.php?id=41244&r=mysqlcfg
#41244 [Opn->Bgs]: Multiple Queries in MSSQL Kills Connection and Prevents New Connections
ID: 41244
Updated by: [EMAIL PROTECTED]
Reported By: mallinger at tc dot edu
-Status: Open
+Status: Bogus
Bug Type: MSSQL related
Operating System: Windows 2003 Server
PHP Version: 5.2.1
New Comment:
MS SQL server does not allow new queries on the same connection as long
as there are results pending the clients actions. Either fetch all data
or release the result before sending new queries.
Another solution is to have multiple connections. mssql_connect()
supports an optional parameter to force the creation of a new connection
even if host, user and password is the same as for an existing
connection.
Previous Comments:
[2007-04-30 17:58:34] mallinger at tc dot edu
Description:
When executing multiple queries from a single mssql_query call, the
connection is lost and subsequent connections cannot be created. There
are similar problems documented for PHP4, but it seems that (a) this is
not documented for PHP5; and (b) the inability to create subsequent
connections is not documented. In particular, it seems the
mssql_select_db is what is failing in the second connection.
Reproduce code:
---
$conn = mssql_connect('host', 'user', 'pass');
mssql_select_db("db", $conn);
mssql_query("QUERY 1", $conn);
mssql_query("QUERY 2; QUERY 3; QUERY 4;", $conn);
// this will fail:
mssql_query("QUERY 5", $conn);
// this is ok
mssql_close($conn);
// everything below will fail too:
$conn2 = mssql_connect('host', 'user', 'pass');
mssql_select_db("db", $conn2);
mssql_query("QUERY 6", $conn2);
Expected result:
The queries should execute with no problems.
Actual result:
--
Everything after the multiple query mssql_query call ("QUERY 2; QUERY
3; QUERY 4;" will fail. New connections fail when trying to select a
database ("mssql_select_db() [function.mssql-select-db]: Unable to
select database:...") and the existing connection cannot be used anymore
("mssql_query() [function.mssql-query]: Query failed in...") .
--
Edit this bug report at http://bugs.php.net/?id=41244&edit=1
#41233 [Fbk->Opn]: "Per Directory Values" are not applied correctly
ID: 41233 User updated by: simon at bleif dot de Reported By: simon at bleif dot de -Status: Feedback +Status: Open Bug Type: IIS related Operating System: Windows Server 2003 x64 R2 SP2 PHP Version: 5.2.1 New Comment: Thank you for your reply. Unfortunately, things are now broken completely in that snapshot - although all values are being read from the registry, none of them are being applied (not even in .php-files located in the first folder). Previous Comments: [2007-04-30 13:22:52] [EMAIL PROTECTED] Please try using this CVS snapshot: http://snaps.php.net/php5.2-latest.tar.gz For Windows: http://snaps.php.net/win32/php5.2-win32-latest.zip [2007-04-30 02:50:33] simon at bleif dot de Description: According to http://www.php-center.de/en-html-manual/configuration.changes.html, you are able to change parts of the PHP-configuration on a per-directory-manner. After setting the desired configuration options there and restarting IIS, only the options for the first folder (if there are some "folders"/registry keys within the same "folder"/key each configured with individual options) are applied. A trace with Regmon shows the W3WP enumerating and (successfully) accessing all entries under the "Per Directory Values"-hive. Reproduce code: --- Setup at least 2 virtul hosts (located under C:\Inetpub\wwwroot\a and C:\Inetpub\wwwroot\b). Start Regedit, navigate to the key HKLM\Software\PHP (HKLM\Software\Wow6432Node\PHP if you are on x64), create the subkey "Per Directory Values". This key gets a new subkey named "c", that key gets one named "Inetpub", that one gets a subkey named "wwwroot". Now you create two keys (one named "a", the other one named "b") directly below the key "wwwroot". Now, create a new string value called for example "log_errors" in both of these keys and set that value to the opposite of the value defined in the php.ini (for example 0). Restart IIS, place a file named test.php in both physical virtual host directory containing the following: If you now launch your web browser, point it to virtual host A, the correct values (as they were set in the registry) are being displayed. If you do the same with host B, the default values (as set in the php.ini or a registry key above the virtual host B configuration-key) (or NULL) are beeing displayed, not the setting defined in the registry. If you now delete the key "a" in the registry and restart IIS, the correct value is displayed when you point your browser to virtual host B. If you create a third virtual host C (folder: c), the settings stored in the registry are applied only if there are no other hosts-settings defined before (in alphabetical order, starting with a) that host in the same registry-key. The same issue also appears on a second server (same OS, same versions, but different php.ini). The issue applies both in safe- and non-safe mode. Expected result: The settings are being applied correctly on a per folder-level. Actual result: -- The settings are only applied to the first (in alphabetical order) configured folder in a directory. Further details please see under "Reproduce code". -- Edit this bug report at http://bugs.php.net/?id=41233&edit=1
#41233 [Opn->Bgs]: "Per Directory Values" are not applied correctly
ID: 41233 Updated by: [EMAIL PROTECTED] Reported By: simon at bleif dot de -Status: Open +Status: Bogus Bug Type: IIS related Operating System: Windows Server 2003 x64 R2 SP2 PHP Version: 5.2.1 New Comment: Duplicate of bug #41192. Previous Comments: [2007-04-30 21:38:23] simon at bleif dot de Thank you for your reply. Unfortunately, things are now broken completely in that snapshot - although all values are being read from the registry, none of them are being applied (not even in .php-files located in the first folder). [2007-04-30 13:22:52] [EMAIL PROTECTED] Please try using this CVS snapshot: http://snaps.php.net/php5.2-latest.tar.gz For Windows: http://snaps.php.net/win32/php5.2-win32-latest.zip [2007-04-30 02:50:33] simon at bleif dot de Description: According to http://www.php-center.de/en-html-manual/configuration.changes.html, you are able to change parts of the PHP-configuration on a per-directory-manner. After setting the desired configuration options there and restarting IIS, only the options for the first folder (if there are some "folders"/registry keys within the same "folder"/key each configured with individual options) are applied. A trace with Regmon shows the W3WP enumerating and (successfully) accessing all entries under the "Per Directory Values"-hive. Reproduce code: --- Setup at least 2 virtul hosts (located under C:\Inetpub\wwwroot\a and C:\Inetpub\wwwroot\b). Start Regedit, navigate to the key HKLM\Software\PHP (HKLM\Software\Wow6432Node\PHP if you are on x64), create the subkey "Per Directory Values". This key gets a new subkey named "c", that key gets one named "Inetpub", that one gets a subkey named "wwwroot". Now you create two keys (one named "a", the other one named "b") directly below the key "wwwroot". Now, create a new string value called for example "log_errors" in both of these keys and set that value to the opposite of the value defined in the php.ini (for example 0). Restart IIS, place a file named test.php in both physical virtual host directory containing the following: If you now launch your web browser, point it to virtual host A, the correct values (as they were set in the registry) are being displayed. If you do the same with host B, the default values (as set in the php.ini or a registry key above the virtual host B configuration-key) (or NULL) are beeing displayed, not the setting defined in the registry. If you now delete the key "a" in the registry and restart IIS, the correct value is displayed when you point your browser to virtual host B. If you create a third virtual host C (folder: c), the settings stored in the registry are applied only if there are no other hosts-settings defined before (in alphabetical order, starting with a) that host in the same registry-key. The same issue also appears on a second server (same OS, same versions, but different php.ini). The issue applies both in safe- and non-safe mode. Expected result: The settings are being applied correctly on a per folder-level. Actual result: -- The settings are only applied to the first (in alphabetical order) configured folder in a directory. Further details please see under "Reproduce code". -- Edit this bug report at http://bugs.php.net/?id=41233&edit=1
#41232 [Opn->Fbk]: Eval failing on calls to class methods
ID: 41232 Updated by: [EMAIL PROTECTED] Reported By: Antoniocs at gmail dot com -Status: Open +Status: Feedback Bug Type: Output Control Operating System: Windows XP sp2 PHP Version: 5.2.1 New Comment: It's commented out (and I get the output I pasted earlier). I'm asking you again: please provide a SHORT but compete reproduce script. Take these scripts, merge them into the one script and remove everything not necessary to reproduce the problem. That'll help you to find your bugs in the code. Also please try to execute it in console and look into the error log. The "I see an empty screen" thing is completely useless. Previous Comments: [2007-04-30 16:46:50] Antoniocs at gmail dot com I commented the $blas->teste(); If you remove the comments from that line you will get a blank page [2007-04-30 16:43:31] [EMAIL PROTECTED] Hey!! Tudo bem?? OLAS Notice: Use of undefined constant TESTE - assumed 'TESTE' in /tmp/eval_test/page_parser.php(27) : eval()'d code on line 3 TESTE iojdsfoijdoidsd1177951395 Notice: Undefined variable: BLA in /tmp/eval_test/page_parser.php(27) : eval()'d code on line 12 Hello hello [2007-04-30 16:06:43] Antoniocs at gmail dot com Here are the source files: http://www.antoniocs.org/eval_test.rar Unpack, and run teste_parser.php [2007-04-30 16:05:05] [EMAIL PROTECTED] Please provide a SHORT but complete reproduce case in ONE short script. I believe HTML or several scripts are not required for this. Thank you. [2007-04-30 14:44:01] Antoniocs at gmail dot com I set all the code you would need in the paste site. Just run the teste_parser.php file and it will call the rest The remainder of the comments for this report are too long. To view the rest of the comments, please view the bug report online at http://bugs.php.net/41232 -- Edit this bug report at http://bugs.php.net/?id=41232&edit=1
#41232 [Fbk->Opn]: Eval failing on calls to class methods
ID: 41232 User updated by: Antoniocs at gmail dot com Reported By: Antoniocs at gmail dot com -Status: Feedback +Status: Open Bug Type: Output Control Operating System: Windows XP sp2 PHP Version: 5.2.1 New Comment: Uncomment the line!! I can't put them all together because this is suppose to work like this. The page appears blank when I call the method Change //$blas->teste(); To $blas->teste(); Previous Comments: [2007-04-30 22:00:43] [EMAIL PROTECTED] It's commented out (and I get the output I pasted earlier). I'm asking you again: please provide a SHORT but compete reproduce script. Take these scripts, merge them into the one script and remove everything not necessary to reproduce the problem. That'll help you to find your bugs in the code. Also please try to execute it in console and look into the error log. The "I see an empty screen" thing is completely useless. [2007-04-30 16:46:50] Antoniocs at gmail dot com I commented the $blas->teste(); If you remove the comments from that line you will get a blank page [2007-04-30 16:43:31] [EMAIL PROTECTED] Hey!! Tudo bem?? OLAS Notice: Use of undefined constant TESTE - assumed 'TESTE' in /tmp/eval_test/page_parser.php(27) : eval()'d code on line 3 TESTE iojdsfoijdoidsd1177951395 Notice: Undefined variable: BLA in /tmp/eval_test/page_parser.php(27) : eval()'d code on line 12 Hello hello [2007-04-30 16:06:43] Antoniocs at gmail dot com Here are the source files: http://www.antoniocs.org/eval_test.rar Unpack, and run teste_parser.php [2007-04-30 16:05:05] [EMAIL PROTECTED] Please provide a SHORT but complete reproduce case in ONE short script. I believe HTML or several scripts are not required for this. Thank you. 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/41232 -- Edit this bug report at http://bugs.php.net/?id=41232&edit=1
#41192 [Com]: Per Directory Values only work for one key
ID: 41192 Comment by: simon at bleif dot de Reported By: martin at itmission dot com Status: Assigned Bug Type: PHP options/info functions Operating System: Windows Server 2003 PHP Version: 5.2.1 Assigned To: dmitry New Comment: Can be reproduced in Windows Server 2003 x64 R2. Filed a bugreport regarding the same issue (see http://bugs.php.net/bug.php?id=41233). I have to add that all entries are being read from the registry correctly (result: success in Regmon), but are not (as he already said) applied. Previous Comments: [2007-04-26 11:32:16] martin at itmission dot com I downloaded the newest Win32 Snapshot from snaps.php.net: php5.2 (zip) (9.5M) Built On: Apr 26, 2007 08:30 GMT In this version of PHP (5.2.2RC2-dev), Per Directory Values do not work at all. My phpinfo page shows display errors: off for both sites. When I roll back to PHP-5.2.1, it works for the first website in the registry, and with PHP-5.2.0, it works for both of them. [2007-04-26 11:05:39] [EMAIL PROTECTED] The bug should be already fixed. Could you check snapshot from http://snaps.php.net [2007-04-26 03:09:02] martin at itmission dot com I am assuming that it has something to do with the massive change to registry.c between 5.2.0 and 5.2.1 changelogged as "implemented registry cache that prevent registry lookup on each request. In case of modification of corresponding registry-tree PHP will reload it automatic". Perhaps something in the way it's being cached is not working properly. The simpler code of 5.2.0 works as expected (and it hits the registry with every request, but I can live with that). [2007-04-26 02:13:08] martin at itmission dot com Possibly noteworthy, the exact same configuration works fine in PHP 5.2.0. [2007-04-26 01:41:12] martin at itmission dot com I understand that I was wrong to try to make it work with auto_prepend_file. When I remove auto_prepend_file and specify only display_errors => on (PHP_INI_ALL), and change my scripts to , www1 shows Display Errors: On, and www2 shows Display Errors: off. Deleting www1 or renaming it to www3 causes www2 to show Display Errors: on. When I create a fake error, like trying to open a nonexistent file, www1 displays the error, while www2 does not. 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/41192 -- Edit this bug report at http://bugs.php.net/?id=41192&edit=1
#41232 [Opn->Bgs]: Eval failing on calls to class methods
ID: 41232 Updated by: [EMAIL PROTECTED] Reported By: Antoniocs at gmail dot com -Status: Open +Status: Bogus Bug Type: Output Control Operating System: Windows XP sp2 PHP Version: 5.2.1 New Comment: Debug your code. The code you're trying to eval() in the output handler is wrong. Previous Comments: [2007-04-30 22:04:22] Antoniocs at gmail dot com Uncomment the line!! I can't put them all together because this is suppose to work like this. The page appears blank when I call the method Change //$blas->teste(); To $blas->teste(); [2007-04-30 22:00:43] [EMAIL PROTECTED] It's commented out (and I get the output I pasted earlier). I'm asking you again: please provide a SHORT but compete reproduce script. Take these scripts, merge them into the one script and remove everything not necessary to reproduce the problem. That'll help you to find your bugs in the code. Also please try to execute it in console and look into the error log. The "I see an empty screen" thing is completely useless. [2007-04-30 16:46:50] Antoniocs at gmail dot com I commented the $blas->teste(); If you remove the comments from that line you will get a blank page [2007-04-30 16:43:31] [EMAIL PROTECTED] Hey!! Tudo bem?? OLAS Notice: Use of undefined constant TESTE - assumed 'TESTE' in /tmp/eval_test/page_parser.php(27) : eval()'d code on line 3 TESTE iojdsfoijdoidsd1177951395 Notice: Undefined variable: BLA in /tmp/eval_test/page_parser.php(27) : eval()'d code on line 12 Hello hello [2007-04-30 16:06:43] Antoniocs at gmail dot com Here are the source files: http://www.antoniocs.org/eval_test.rar Unpack, and run teste_parser.php The remainder of the comments for this report are too long. To view the rest of the comments, please view the bug report online at http://bugs.php.net/41232 -- Edit this bug report at http://bugs.php.net/?id=41232&edit=1
#41232 [Bgs->Opn]: Eval failing on calls to class methods
ID: 41232 User updated by: Antoniocs at gmail dot com Reported By: Antoniocs at gmail dot com -Status: Bogus +Status: Open Bug Type: Output Control Operating System: Windows XP sp2 PHP Version: 5.2.1 New Comment: The code is fine!! The way in which I call it is what is breaking up the code. But that shouldn't happen!! The $bla->teste() method is causing the script to somehow screw up, but there is nothing wrong in the class Previous Comments: [2007-04-30 22:12:18] [EMAIL PROTECTED] Debug your code. The code you're trying to eval() in the output handler is wrong. [2007-04-30 22:04:22] Antoniocs at gmail dot com Uncomment the line!! I can't put them all together because this is suppose to work like this. The page appears blank when I call the method Change //$blas->teste(); To $blas->teste(); [2007-04-30 22:00:43] [EMAIL PROTECTED] It's commented out (and I get the output I pasted earlier). I'm asking you again: please provide a SHORT but compete reproduce script. Take these scripts, merge them into the one script and remove everything not necessary to reproduce the problem. That'll help you to find your bugs in the code. Also please try to execute it in console and look into the error log. The "I see an empty screen" thing is completely useless. [2007-04-30 16:46:50] Antoniocs at gmail dot com I commented the $blas->teste(); If you remove the comments from that line you will get a blank page [2007-04-30 16:43:31] [EMAIL PROTECTED] Hey!! Tudo bem?? OLAS Notice: Use of undefined constant TESTE - assumed 'TESTE' in /tmp/eval_test/page_parser.php(27) : eval()'d code on line 3 TESTE iojdsfoijdoidsd1177951395 Notice: Undefined variable: BLA in /tmp/eval_test/page_parser.php(27) : eval()'d code on line 12 Hello hello 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/41232 -- Edit this bug report at http://bugs.php.net/?id=41232&edit=1
#41232 [Opn->Bgs]: Eval failing on calls to class methods
ID: 41232 Updated by: [EMAIL PROTECTED] Reported By: Antoniocs at gmail dot com -Status: Open +Status: Bogus Bug Type: Output Control Operating System: Windows XP sp2 PHP Version: 5.2.1 New Comment: As long as you're unable to provide a short reproduce code, it's obvious that the problem is in your code. Please keep this report marked as bogus, thank you. Previous Comments: [2007-04-30 22:28:11] Antoniocs at gmail dot com The code is fine!! The way in which I call it is what is breaking up the code. But that shouldn't happen!! The $bla->teste() method is causing the script to somehow screw up, but there is nothing wrong in the class [2007-04-30 22:12:18] [EMAIL PROTECTED] Debug your code. The code you're trying to eval() in the output handler is wrong. [2007-04-30 22:04:22] Antoniocs at gmail dot com Uncomment the line!! I can't put them all together because this is suppose to work like this. The page appears blank when I call the method Change //$blas->teste(); To $blas->teste(); [2007-04-30 22:00:43] [EMAIL PROTECTED] It's commented out (and I get the output I pasted earlier). I'm asking you again: please provide a SHORT but compete reproduce script. Take these scripts, merge them into the one script and remove everything not necessary to reproduce the problem. That'll help you to find your bugs in the code. Also please try to execute it in console and look into the error log. The "I see an empty screen" thing is completely useless. [2007-04-30 16:46:50] Antoniocs at gmail dot com I commented the $blas->teste(); If you remove the comments from that line you will get a blank page 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/41232 -- Edit this bug report at http://bugs.php.net/?id=41232&edit=1
#41232 [Bgs]: Eval failing on calls to class methods
ID: 41232 User updated by: Antoniocs at gmail dot com Reported By: Antoniocs at gmail dot com Status: Bogus Bug Type: Output Control Operating System: Windows XP sp2 PHP Version: 5.2.1 New Comment: This is not bogus!! Go here and see the script running http://www.antoniocs.org/eval_test/ http://www.antoniocs.org/eval_test/eval_test.rar <-- download the code here When you run the code you will only get the tag which is before ob_start. Eval screws up everything Previous Comments: [2007-04-30 22:32:34] [EMAIL PROTECTED] As long as you're unable to provide a short reproduce code, it's obvious that the problem is in your code. Please keep this report marked as bogus, thank you. [2007-04-30 22:28:11] Antoniocs at gmail dot com The code is fine!! The way in which I call it is what is breaking up the code. But that shouldn't happen!! The $bla->teste() method is causing the script to somehow screw up, but there is nothing wrong in the class [2007-04-30 22:12:18] [EMAIL PROTECTED] Debug your code. The code you're trying to eval() in the output handler is wrong. [2007-04-30 22:04:22] Antoniocs at gmail dot com Uncomment the line!! I can't put them all together because this is suppose to work like this. The page appears blank when I call the method Change //$blas->teste(); To $blas->teste(); [2007-04-30 22:00:43] [EMAIL PROTECTED] It's commented out (and I get the output I pasted earlier). I'm asking you again: please provide a SHORT but compete reproduce script. Take these scripts, merge them into the one script and remove everything not necessary to reproduce the problem. That'll help you to find your bugs in the code. Also please try to execute it in console and look into the error log. The "I see an empty screen" thing is completely useless. 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/41232 -- Edit this bug report at http://bugs.php.net/?id=41232&edit=1
#41246 [NEW]: Setting locales to german and applying utf8_encode to a float messes it up.
From: phil at goli dot at Operating system: Linux Gentoo PHP version: 5.2.1 PHP Bug Type: ICONV related Bug description: Setting locales to german and applying utf8_encode to a float messes it up. Description: Applies to PHP versions 4.4.1, 4.4.4, 5.2.1, etc. 1. setlocales to german: setlocale(LC_ALL, '[EMAIL PROTECTED]', 'de_DE', 'deu_deu'); 2. assign a float to a variable by defining a string: $myTest = '10.01'; 3. add a float to that string: $myTest = $myTest + 0.01; 4. convert that float to utf8: $myTest = utf8_encode($myTest); 5. add another float to that utf8 string: $myTest = $myTest + 0.01; 6. if the german locales are installed on the system then the output will be wrong. utf8_encode chops off the positions after decimal point. Reproduce code: --- $loc_de = setlocale(LC_ALL, '[EMAIL PROTECTED]', 'de_DE', 'deu_deu'); echo ''; echo 'Preferred locale for german on this system is "' . $loc_de . '"' . "\n"; echo 'Example only works if locales for german are installed!' . "\n\n"; echo 'Normal:' . "\n"; $myTest = '10.01'; var_dump($myTest); $myTest = $myTest + 0.01; var_dump($myTest); // no utf8_encode $myTest = $myTest + 0.01; echo 'Result should be 10.03 and it is: '; var_dump($myTest); echo "\n\n" . 'With utf8_encode:' . "\n"; $myTest = '10.01'; var_dump($myTest); $myTest = $myTest + 0.01; var_dump($myTest); $myTest = utf8_encode($myTest); var_dump($myTest); $myTest = $myTest + 0.01; echo 'Result should be 10.03 BUT it is: '; var_dump($myTest); echo ''; exit; Expected result: Preferred locale for german on this system is "[EMAIL PROTECTED]" Example only works if locales for german are installed! Normal: string(5) "10.01" float(10,02) Result should be 10.03 and it is: float(10,03) With utf8_encode: string(5) "10.01" float(10,02) string(5) "10,02" Result should be 10.03 ... and it is: float(10,03) Actual result: -- Preferred locale for german on this system is "[EMAIL PROTECTED]" Example only works if locales for german are installed! Normal: string(5) "10.01" float(10,02) Result should be 10.03 and it is: float(10,03) With utf8_encode: string(5) "10.01" float(10,02) string(5) "10,02" Result should be 10.03 ... and it is: float(10,01) -- Edit bug report at http://bugs.php.net/?id=41246&edit=1 -- Try a CVS snapshot (PHP 4.4): http://bugs.php.net/fix.php?id=41246&r=trysnapshot44 Try a CVS snapshot (PHP 5.2): http://bugs.php.net/fix.php?id=41246&r=trysnapshot52 Try a CVS snapshot (PHP 6.0): http://bugs.php.net/fix.php?id=41246&r=trysnapshot60 Fixed in CVS: http://bugs.php.net/fix.php?id=41246&r=fixedcvs Fixed in release: http://bugs.php.net/fix.php?id=41246&r=alreadyfixed Need backtrace: http://bugs.php.net/fix.php?id=41246&r=needtrace Need Reproduce Script:http://bugs.php.net/fix.php?id=41246&r=needscript Try newer version:http://bugs.php.net/fix.php?id=41246&r=oldversion Not developer issue: http://bugs.php.net/fix.php?id=41246&r=support Expected behavior:http://bugs.php.net/fix.php?id=41246&r=notwrong Not enough info: http://bugs.php.net/fix.php?id=41246&r=notenoughinfo Submitted twice: http://bugs.php.net/fix.php?id=41246&r=submittedtwice register_globals: http://bugs.php.net/fix.php?id=41246&r=globals PHP 3 support discontinued: http://bugs.php.net/fix.php?id=41246&r=php3 Daylight Savings: http://bugs.php.net/fix.php?id=41246&r=dst IIS Stability:http://bugs.php.net/fix.php?id=41246&r=isapi Install GNU Sed: http://bugs.php.net/fix.php?id=41246&r=gnused Floating point limitations: http://bugs.php.net/fix.php?id=41246&r=float No Zend Extensions: http://bugs.php.net/fix.php?id=41246&r=nozend MySQL Configuration Error:http://bugs.php.net/fix.php?id=41246&r=mysqlcfg
#41195 [Opn]: Indexing non-array should raise notice
ID: 41195 User updated by: lbayuk at pobox dot com Reported By: lbayuk at pobox dot com Status: Open -Bug Type: Feature/Change Request +Bug Type: Arrays related Operating System: Linux PHP Version: 5.2.1 New Comment: Changed category... I thought I put it in as "Arrays related", but I guess I missed and it ended up as "Feature/Change Request". To me, this is either a bug, or a documentation improvement request, not a feature request. Previous Comments: [2007-04-26 01:38:31] lbayuk at pobox dot com Description: If I index a scalar or NULL variable as if it were an array, I get NULL back with no warning or notice. This is true for any index value, including 0. I expected an Undefined offset or incorrect type message. The manual says an integer variable = n will convert to array(0=>$n); that works on casting to array but not on indexing. Reproduce code: --- 1); $notarry = NULL; $intvar = 123; echo "1. arry[1] = " . $arry[1] . "\n"; echo "2. arry[2] = " . $arry[2] . "\n"; echo "3. intvar[0] = " . $intvar[0] . "\n"; echo "4. intvar[1] = " . $intvar[1] . "\n"; echo "5. notarry[1] = " . $notarry[1] . "\n"; Expected result: "Notice: Undefined offset" (or incorrect type) for (2), (3), (4), and (5). Or maybe (3) should echo "123" via implicit cast. But only (2) raises a Notice; the others just echo NULL. Actual result: -- 1. arry[1] = 1 Notice: Undefined offset: 2 in /a5/home/ljb/lang/php/p.php on line 7 2. arry[2] = 3. intvar[0] = 4. intvar[1] = 5. notarry[1] = -- Edit this bug report at http://bugs.php.net/?id=41195&edit=1
#41232 [Fbk->Opn]: Eval failing on calls to class methods
ID: 41232 User updated by: Antoniocs at gmail dot com Reported By: Antoniocs at gmail dot com -Status: Feedback +Status: Open Bug Type: Output Control Operating System: Windows XP sp2 PHP Version: 5.2.1 New Comment: I commented the $blas->teste(); If you remove the comments from that line you will get a blank page Previous Comments: [2007-04-30 16:43:31] [EMAIL PROTECTED] Hey!! Tudo bem?? OLAS Notice: Use of undefined constant TESTE - assumed 'TESTE' in /tmp/eval_test/page_parser.php(27) : eval()'d code on line 3 TESTE iojdsfoijdoidsd1177951395 Notice: Undefined variable: BLA in /tmp/eval_test/page_parser.php(27) : eval()'d code on line 12 Hello hello [2007-04-30 16:06:43] Antoniocs at gmail dot com Here are the source files: http://www.antoniocs.org/eval_test.rar Unpack, and run teste_parser.php [2007-04-30 16:05:05] [EMAIL PROTECTED] Please provide a SHORT but complete reproduce case in ONE short script. I believe HTML or several scripts are not required for this. Thank you. [2007-04-30 14:44:01] Antoniocs at gmail dot com I set all the code you would need in the paste site. Just run the teste_parser.php file and it will call the rest [2007-04-30 13:22:18] [EMAIL PROTECTED] Thank you for this bug report. To properly diagnose the problem, we need a short but complete example script to be able to reproduce this bug ourselves. A proper reproducing script starts with , is max. 10-20 lines long and does not require any external resources such as databases, etc. If the script requires a database to demonstrate the issue, please make sure it creates all necessary tables, stored procedures etc. Please avoid embedding huge scripts into the report. The remainder of the comments for this report are too long. To view the rest of the comments, please view the bug report online at http://bugs.php.net/41232 -- Edit this bug report at http://bugs.php.net/?id=41232&edit=1
#41248 [NEW]: DOMDocument / DOMNode parameters
From: ifland at gmail dot com
Operating system: All
PHP version: 5.2.1
PHP Bug Type: XSLT related
Bug description: DOMDocument / DOMNode parameters
Description:
As far as I can tell, the setParameter() function
(http://php.net/xsl_xsltprocessor_set_parameter) doesn't support passing
DOMDocument objects or DOMNode objects as parameters.
I was hoping to get it to work like it does in the .NET 2.0 framework.
(See
http://msdn2.microsoft.com/en-us/library/system.xml.xsl.xsltargumentlist.addparam.aspx)
But after looking at the PHP 5.2.1 source code,
/php-5.2.1/ext/xsl/xsltprocessor.c line 604-650, it appears that it's not
possible to do so in PHP even though it appears that libxslt supports it
(see
http://xmlsoft.org/XSLT/html/libxslt-variables.html#xsltParseGlobalParam)
Reproduce code:
---
...
$xslt = new XSLTProcessor();
//load stylesheet
...
$param = new DOMDocument();
$param->loadXML('');
$xslt->setParameter('', 'PARAMS', $param);
...
//use the value like this in the XSL
...
Expected result:
The function should take the whole XML blob that's passed in and make it
available to the XSL as a global parameter variable.
Actual result:
--
Just a misleading warning ("Wrong parameter count")
--
Edit bug report at http://bugs.php.net/?id=41248&edit=1
--
Try a CVS snapshot (PHP 4.4):
http://bugs.php.net/fix.php?id=41248&r=trysnapshot44
Try a CVS snapshot (PHP 5.2):
http://bugs.php.net/fix.php?id=41248&r=trysnapshot52
Try a CVS snapshot (PHP 6.0):
http://bugs.php.net/fix.php?id=41248&r=trysnapshot60
Fixed in CVS: http://bugs.php.net/fix.php?id=41248&r=fixedcvs
Fixed in release:
http://bugs.php.net/fix.php?id=41248&r=alreadyfixed
Need backtrace: http://bugs.php.net/fix.php?id=41248&r=needtrace
Need Reproduce Script:http://bugs.php.net/fix.php?id=41248&r=needscript
Try newer version:http://bugs.php.net/fix.php?id=41248&r=oldversion
Not developer issue: http://bugs.php.net/fix.php?id=41248&r=support
Expected behavior:http://bugs.php.net/fix.php?id=41248&r=notwrong
Not enough info:
http://bugs.php.net/fix.php?id=41248&r=notenoughinfo
Submitted twice:
http://bugs.php.net/fix.php?id=41248&r=submittedtwice
register_globals: http://bugs.php.net/fix.php?id=41248&r=globals
PHP 3 support discontinued: http://bugs.php.net/fix.php?id=41248&r=php3
Daylight Savings: http://bugs.php.net/fix.php?id=41248&r=dst
IIS Stability:http://bugs.php.net/fix.php?id=41248&r=isapi
Install GNU Sed: http://bugs.php.net/fix.php?id=41248&r=gnused
Floating point limitations: http://bugs.php.net/fix.php?id=41248&r=float
No Zend Extensions: http://bugs.php.net/fix.php?id=41248&r=nozend
MySQL Configuration Error:http://bugs.php.net/fix.php?id=41248&r=mysqlcfg
#41249 [NEW]: [raw]urldecode memory consumption
From: 3172continuum at gmail dot com Operating system: w2k PHP version: 5.2.1 PHP Bug Type: Performance problem Bug description: [raw]urldecode memory consumption Description: every call of [raw]urldecode consumes (iirc) like 200 extra bytes of my memory! if i just delete parts of the array like below, i have a workaround for the next few weeks... this may though not be useful to other users... i traced it down to this single line which starts with $IMPORT_filepath Reproduce code: --- //the next line use several extra bytes on each call, why that? #$IMPORT_filepath = rawurldecode($res[$i]->path); $IMPORT_filepath = urldecode($res[$i]->path); //with this extra line, memory usage gets lowered: $res[$i] = ''; -- Edit bug report at http://bugs.php.net/?id=41249&edit=1 -- Try a CVS snapshot (PHP 4.4): http://bugs.php.net/fix.php?id=41249&r=trysnapshot44 Try a CVS snapshot (PHP 5.2): http://bugs.php.net/fix.php?id=41249&r=trysnapshot52 Try a CVS snapshot (PHP 6.0): http://bugs.php.net/fix.php?id=41249&r=trysnapshot60 Fixed in CVS: http://bugs.php.net/fix.php?id=41249&r=fixedcvs Fixed in release: http://bugs.php.net/fix.php?id=41249&r=alreadyfixed Need backtrace: http://bugs.php.net/fix.php?id=41249&r=needtrace Need Reproduce Script:http://bugs.php.net/fix.php?id=41249&r=needscript Try newer version:http://bugs.php.net/fix.php?id=41249&r=oldversion Not developer issue: http://bugs.php.net/fix.php?id=41249&r=support Expected behavior:http://bugs.php.net/fix.php?id=41249&r=notwrong Not enough info: http://bugs.php.net/fix.php?id=41249&r=notenoughinfo Submitted twice: http://bugs.php.net/fix.php?id=41249&r=submittedtwice register_globals: http://bugs.php.net/fix.php?id=41249&r=globals PHP 3 support discontinued: http://bugs.php.net/fix.php?id=41249&r=php3 Daylight Savings: http://bugs.php.net/fix.php?id=41249&r=dst IIS Stability:http://bugs.php.net/fix.php?id=41249&r=isapi Install GNU Sed: http://bugs.php.net/fix.php?id=41249&r=gnused Floating point limitations: http://bugs.php.net/fix.php?id=41249&r=float No Zend Extensions: http://bugs.php.net/fix.php?id=41249&r=nozend MySQL Configuration Error:http://bugs.php.net/fix.php?id=41249&r=mysqlcfg
