#45726 [Fbk->Opn]: PHP_Archive / Archive.php missing
ID: 45726
User updated by: Bjorn dot Wiberg at its dot uu dot se
Reported By: Bjorn dot Wiberg at its dot uu dot se
-Status: Feedback
+Status: Open
Bug Type: PHAR related
Operating System: IBM AIX 5.3 5300-08-01-0819
-PHP Version: 5.3.0alpha1
+PHP Version: 5.3-200808060630
New Comment:
Sorry, same error with 5.3-dev (5.3-200808060630):
---8<--- excerpt ---8<---
t/power_510_32/usr/lib
-L/gestconf/project/GNOME_ACL/GNOME/build/acc_gnome2_14_0f1/export/power_510_32/usr/lib
-L/gestconf/project/GNOME_ACL/GNOME/build/GNOME_2_14_0/export/power_510_32/usr/lib
-L/users/project/PDP/PDP_51_vac_6_14/usr/ccs/lib
-L/users/project/PDP/PDP_51_vac_6_14/usr/lib -lz -lm -lz -ljpeg -lssl
-lcrypto -lgdbm -lbz2 -lz -lssl -lcrypto -lm -lz -liconv -lm -lcurl
-lssl -lcrypto -lz -lssl -lcrypto -lz -lz -liconv -lm -lz -liconv -lm
-lz -liconv -lm -lz -liconv -lm -lz -liconv -lm -lz -liconv -lm -lz
-liconv -lm -lz -liconv -lm -lxslt -lz -liconv -lm -lxml2 -lpthread -lz
-liconv -lm -lz -liconv -lm
-Wl,-blibpath:/usr/local/lib:/opt/freeware/lib:/usr/X11R6/lib:/opt/freeware/lib/gcc/powerpc-ibm-aix5.3.0.0/4.0.0:/opt/freeware/lib/gcc/powerpc-ibm-aix5.3.0.0/4.0.0/../../..:/usr/lib:/lib
ld: 0711-224 WARNING: Duplicate symbol: php_optidx
ld: 0711-345 Use the -bloadmap or -bnoquiet option to obtain more
information.
ld: 0711-783 WARNING: TOC overflow. TOC size: 72664 Maximum size:
65536
Extra instructions are being generated for each reference to a
TOC
symbol if the symbol is in the TOC overflow area.
Generating phar.php
Generating phar.phar
Pear package PHP_Archive or Archive.php class file not found.
Fatal error: Uncaught exception 'PharException' with message 'illegal
stub for phar
"/home/bwiberg/rpm/BUILD/php5.3-200808060630/ext/phar/phar.phar"' in
/home/bwiberg/rpm/BUILD/php5.3-200808060630/ext/phar/phar.php:994
Stack trace:
#0 /home/bwiberg/rpm/BUILD/php5.3-200808060630/ext/phar/phar.php(994):
Phar->setStub('#!/apache/php/b...')
#1 /home/bwiberg/rpm/BUILD/php5.3-200808060630/ext/phar/phar.php(1054):
PharCommand->phar_set_stub_begin(Object(Phar), '/home/bwiberg/r...',
'/home/bwiberg/r...', '/apache/php/bin...')
#2 [internal function]: PharCommand->cli_cmd_run_pack(Array)
#3 /home/bwiberg/rpm/BUILD/php5.3-200808060630/ext/phar/phar.php(225):
call_user_func(Array, Array)
#4 /home/bwiberg/rpm/BUILD/php5.3-200808060630/ext/phar/phar.php(2073):
CLICommand->__construct(19, Array)
#5 {main}
thrown in
/home/bwiberg/rpm/BUILD/php5.3-200808060630/ext/phar/phar.php on line
994
make: *** [ext/phar/phar.phar] Error 255
Bad exit status from /var/opt/freeware/tmp/rpm-tmp.10312 (%build)
[EMAIL PROTECTED]:~/rpm/SPECS$ cd ..
[EMAIL PROTECTED]:~/rpm$ cd BUILD/php5.3-200808060630/
[EMAIL PROTECTED]:~/rpm/BUILD/php5.3-200808060630$
/opt/freeware/bin/find -iname Archive.php
[EMAIL PROTECTED]:~/rpm/BUILD/php5.3-200808060630$
--->8--- end excerpt --->8---
(No Archive.php anywhere...)
Previous Comments:
[2008-08-05 15:13:28] [EMAIL PROTECTED]
Please try using this CVS snapshot:
http://snaps.php.net/php5.3-latest.tar.gz
For Windows (zip):
http://snaps.php.net/win32/php5.3-win32-latest.zip
For Windows (installer):
http://snaps.php.net/win32/php5.3-win32-installer-latest.msi
[2008-08-05 13:53:40] Bjorn dot Wiberg at its dot uu dot se
Description:
PHP complains about Archive.php / PHP_Archive not being found during
compilation.
Reproduce code:
---
#! /bin/sh
#
# Created by configure
LDFLAGS='-Wl,-bbigtoc' \
CC='gcc' \
'./configure' \
'--disable-fileinfo' \
'--enable-bcmath' \
'--enable-calendar' \
'--enable-cli' \
'--enable-dba' \
'--enable-dbase' \
'--enable-debug' \
'--enable-exif' \
'--enable-flatfile' \
'--enable-ftp' \
'--enable-gd-jis-conv' \
'--enable-gd-native-ttf' \
'--enable-inifile' \
'--enable-mbstring' \
'--enable-pcntl' \
'--enable-shmop' \
'--enable-soap' \
'--enable-sockets' \
'--enable-sqlite-utf8' \
'--enable-sysvmsg' \
'--enable-sysvsem' \
'--enable-sysvshm' \
'--enable-wddx' \
'--enable-zip' \
'--enable-zend-multibyte' \
'--prefix=/apache/php' \
'--with-apxs2=/apache/bin/apxs' \
'--with-bz2' \
'--with-cdb' \
'--with-curl' \
'--with-freetype-dir' \
'--with-gd' \
'--with-gdbm' \
'--with-gettext' \
'--with-jpeg-dir' \
'--with-ldap' \
'--with-libxml-dir=/usr/local' \
'--with-mime-magic' \
'--with-mysql=mysqlnd' \
'--with-mysqli=mysqlnd' \
'--with-openssl=/opt/freeware' \
'--with-pdo-mysql=mysqlnd' \
'--with-png-dir' \
'--with-xmlrpc' \
'--with-xpm-dir' \
'--with-xsl' \
'--with-zlib' \
'--with-zlib-dir' \
"$@"
Followed by '/opt/freeware/bin/make' (GNU make).
Expected result:
No error message.
Actual result:
--
gcc -I/usr/include -g -fvisibility=hidden -O0 -Wall -Wl,-bbigtoc
-Wl,-brtl -Wl,-bE:php.sym ext/ereg/.libs/ereg.o
ext/ereg/regex/.l
#45706 [Opn]: Serializing of ArrayIterator extended Objects
ID: 45706
User updated by: bobka at bigfoot dot com
Reported By: bobka at bigfoot dot com
Status: Open
-Bug Type: Documentation problem
+Bug Type: Session related
Operating System: Linux
PHP Version: 5.2.6
New Comment:
The bug is somewhere in "php" or "php_bin" serialize_handler that is
used to serialize _SESSION
Previous Comments:
[2008-08-06 08:38:06] bobka at bigfoot dot com
To define the class before session_start is the usual case. But if two
applications use the same session data, where one part of session data
is shared between applications and other are used exclusively, you don't
have to include *all* class definitions in each application. In this
case the part of session data, that is not defined, should be preserved
and not corrupted, shouldn't it?
[2008-08-04 12:51:07] [EMAIL PROTECTED]
The class has to be defined before session_start() call.
Documentation for session seems to lack this little note..
[2008-08-04 12:20:45] bobka at bigfoot dot com
Description:
This is a very special case.
If an object of class that extends ArrayIterator is saved in session
data and the class definition is not available in a later call, then the
session data gets corrupted.
Call the example code three times to see the bug.
Reproduce code:
---
".print_r($_SESSION,true)."";
// already initialized?
if (!isset($_SESSION['first']))
{
// bug occures only if the class is not defined later
// and only if it extends ArrayIterator
class my_array extends ArrayIterator
{
}
$_SESSION['first'] = "The first entry is ok";
$t = new my_array();
// at least one member needed
$t[] = "this gets lost";
$_SESSION['killer'] = $t;
$_SESSION['this_is_going_to_be_messed_up'] = "The value of the entry
is ok";
$_SESSION['last'] = "The last entry is ok";
}
Expected result:
At first call:
Array
(
)
At second call:
Array
(
[first] => The first entry is ok
[killer] => __PHP_Incomplete_Class Object
(
[__PHP_Incomplete_Class_Name] => my_array
[0] => this gets lost
)
[this_is_going_to_be_messed_up] => The value of the entry is ok
[last] => The last entry is ok
)
At third call
Array
(
[first] => The first entry is ok
[killer] => __PHP_Incomplete_Class Object
(
[__PHP_Incomplete_Class_Name] => my_array
[0] => this gets lost
)
[this_is_going_to_be_messed_up] => The value of the entry is ok
[last] => The last entry is ok
)
Actual result:
--
At third call:
Notice: session_start() [function.session-start]: Unexpected end of
serialized data in test.php on line 3
Array
(
[first] => The first entry is ok
[killer] =>
[}this_is_going_to_be_messed_up] => The value of the entry is ok
[last] => The last entry is ok
)
--
Edit this bug report at http://bugs.php.net/?id=45706&edit=1
#45729 [NEW]: PHP crashes when fetching mails with long to-address field via pop3
From: kirsch at mediafinanz dot de
Operating system: Windows
PHP version: 5.2.6
PHP Bug Type: Reproducible crash
Bug description: PHP crashes when fetching mails with long to-address field
via pop3
Description:
When trying to receive a mail with a really long to-address field (e.g.
Spam with 276 receivers) over pop3, PHP crashes. But it seems that it
depends of the way I open the connection to the mailbox.
PHP crashes on long to-address mail if I use $mailbox =
imap_open('{mail.myserver.net:110/pop3/novalidate-cert}INBOX',
'[EMAIL PROTECTED]', 'mydamnsecretpassword');, but everything works fine if
I use $mailbox = imap_open('{mail.myserver.net:143}INBOX',
'[EMAIL PROTECTED]', 'mydamnsecretpassword');
Reproduce code:
---
//you'll need a mail with a lot of receivers for this test
$mailbox = imap_open('{mail.myserver.net:110/pop3/novalidate-cert}INBOX',
'[EMAIL PROTECTED]', 'mydamnsecretpassword');
$check = imap_check($mailbox);
for ($i=1; $i <= $check->Nmsgs; $i++)
{
$uid = imap_uid($mailbox, $i);
echo 'UID: '.$uid;
$messageNumber = imap_msgno($mailbox, $uid);
echo 'MessageNo: '.$messageNumber;
$headerinfo = imap_headerinfo($mailbox, $messageNumber);
[...]
}
Expected result:
$headerinfo contains an stdObject with headerinformation, like it does if
it is a 'normal' mail
Actual result:
--
Crash with message 'This application has requested the Runtime to
terminiate it in an unusual way. Please contact the application's support
team for more information'
--
Edit bug report at http://bugs.php.net/?id=45729&edit=1
--
Try a CVS snapshot (PHP 5.2):
http://bugs.php.net/fix.php?id=45729&r=trysnapshot52
Try a CVS snapshot (PHP 5.3):
http://bugs.php.net/fix.php?id=45729&r=trysnapshot53
Try a CVS snapshot (PHP 6.0):
http://bugs.php.net/fix.php?id=45729&r=trysnapshot60
Fixed in CVS: http://bugs.php.net/fix.php?id=45729&r=fixedcvs
Fixed in release:
http://bugs.php.net/fix.php?id=45729&r=alreadyfixed
Need backtrace: http://bugs.php.net/fix.php?id=45729&r=needtrace
Need Reproduce Script:http://bugs.php.net/fix.php?id=45729&r=needscript
Try newer version:http://bugs.php.net/fix.php?id=45729&r=oldversion
Not developer issue: http://bugs.php.net/fix.php?id=45729&r=support
Expected behavior:http://bugs.php.net/fix.php?id=45729&r=notwrong
Not enough info:
http://bugs.php.net/fix.php?id=45729&r=notenoughinfo
Submitted twice:
http://bugs.php.net/fix.php?id=45729&r=submittedtwice
register_globals: http://bugs.php.net/fix.php?id=45729&r=globals
PHP 4 support discontinued: http://bugs.php.net/fix.php?id=45729&r=php4
Daylight Savings: http://bugs.php.net/fix.php?id=45729&r=dst
IIS Stability:http://bugs.php.net/fix.php?id=45729&r=isapi
Install GNU Sed: http://bugs.php.net/fix.php?id=45729&r=gnused
Floating point limitations: http://bugs.php.net/fix.php?id=45729&r=float
No Zend Extensions: http://bugs.php.net/fix.php?id=45729&r=nozend
MySQL Configuration Error:http://bugs.php.net/fix.php?id=45729&r=mysqlcfg
#45730 [NEW]: missing dll files
From: jwebb at obeliskinternational dot com Operating system: windows xp PHP version: 5.2.6 PHP Bug Type: *General Issues Bug description: missing dll files Description: i just downloaded php-5.2.6.tar.gz and have been through the manual setup for iis5 on my xp machine But there are no dll files in my download So how can i finish setting up IIS to use isapi ?? -- Edit bug report at http://bugs.php.net/?id=45730&edit=1 -- Try a CVS snapshot (PHP 5.2): http://bugs.php.net/fix.php?id=45730&r=trysnapshot52 Try a CVS snapshot (PHP 5.3): http://bugs.php.net/fix.php?id=45730&r=trysnapshot53 Try a CVS snapshot (PHP 6.0): http://bugs.php.net/fix.php?id=45730&r=trysnapshot60 Fixed in CVS: http://bugs.php.net/fix.php?id=45730&r=fixedcvs Fixed in release: http://bugs.php.net/fix.php?id=45730&r=alreadyfixed Need backtrace: http://bugs.php.net/fix.php?id=45730&r=needtrace Need Reproduce Script:http://bugs.php.net/fix.php?id=45730&r=needscript Try newer version:http://bugs.php.net/fix.php?id=45730&r=oldversion Not developer issue: http://bugs.php.net/fix.php?id=45730&r=support Expected behavior:http://bugs.php.net/fix.php?id=45730&r=notwrong Not enough info: http://bugs.php.net/fix.php?id=45730&r=notenoughinfo Submitted twice: http://bugs.php.net/fix.php?id=45730&r=submittedtwice register_globals: http://bugs.php.net/fix.php?id=45730&r=globals PHP 4 support discontinued: http://bugs.php.net/fix.php?id=45730&r=php4 Daylight Savings: http://bugs.php.net/fix.php?id=45730&r=dst IIS Stability:http://bugs.php.net/fix.php?id=45730&r=isapi Install GNU Sed: http://bugs.php.net/fix.php?id=45730&r=gnused Floating point limitations: http://bugs.php.net/fix.php?id=45730&r=float No Zend Extensions: http://bugs.php.net/fix.php?id=45730&r=nozend MySQL Configuration Error:http://bugs.php.net/fix.php?id=45730&r=mysqlcfg
#45731 [NEW]: child pid exit Segmentation fault with php5.2.6 in putenv
From: dharmeshmaniar at rediffmail dot com Operating system: CentOS 3.8 PHP version: 5.2.6 PHP Bug Type: Reproducible crash Bug description: child pid exit Segmentation fault with php5.2.6 in putenv Description: When we use putenv in php5.2.6 it generates segmentation fault. Program received signal SIGSEGV, Segmentation fault. [Switching to Thread -1390175328 (LWP 13275)] 0x009ec303 in strchr () from /lib/tls/libc.so.6 This do not happen every time. The behaviour is random. We removed put_env from our code and everything is working properly. No more Segmentation fault after removing putenv. Note: We are not using any of the 3rd party apps:APC, eAccelerator, Turck MMCache, ionCube loader, Xcache, Xdebug Actual result: -- Below is the dump for gdb trace: [New Thread -1537033312 (LWP 13289)] [Thread -1211847776 (LWP 13258) exited] Program received signal SIGSEGV, Segmentation fault. [Switching to Thread -1390175328 (LWP 13275)] 0x009ec303 in strchr () from /lib/tls/libc.so.6 (gdb) where #0 0x009ec303 in strchr () from /lib/tls/libc.so.6 #1 0x009ace17 in putenv () from /lib/tls/libc.so.6 #2 0x003e2440 in php_putenv_destructor (pe=0xa325bb9c) at /usr/local/src/php-5.2.5/ext/standard/basic_functions.c:3861 #3 0x0049953c in zend_hash_destroy (ht=0xa35336cc) at /usr/local/src/php-5.2.5/Zend/zend_hash.c:526 #4 0x003e2e0a in zm_deactivate_basic (type=1, module_number=9, tsrm_ls=0xa39d1d28) at /usr/local/src/php-5.2.5/ext/standard/basic_functions.c:4179 #5 0x004940ba in module_registry_cleanup (module=0x9392988, tsrm_ls=0xa39d1d28) at /usr/local/src/php-5.2.5/Zend/zend_API.c:1968 #6 0x00499814 in zend_hash_apply (ht=0x601f80, apply_func=0x494098 , tsrm_ls=0xa39d1d28) at /usr/local/src/php-5.2.5/Zend/zend_hash.c:673 #7 0x0048f753 in zend_deactivate_modules (tsrm_ls=0xa39d1d28) at /usr/local/src/php-5.2.5/Zend/zend.c:838 #8 0x00451594 in php_request_shutdown (dummy=0x0) at /usr/local/src/php-5.2.5/main/main.c:1459 #9 0x004ff862 in php_handler (r=0x95c86c0) at /usr/local/src/php-5.2.5/sapi/apache2handler/sapi_apache2.c:471 #10 0x08085f92 in ap_run_handler () #11 0x0808635d in ap_invoke_handler () #12 0x0806feb5 in ap_process_request () #13 0x0806b96d in _start () #14 0x095c86c0 in ?? () #15 0x0004 in ?? () #16 0x095c86c0 in ?? () #17 0x095c86c0 in ?? () #18 0x0954d150 in ?? () #19 0x0954d060 in ?? () #20 0x in ?? () -- Edit bug report at http://bugs.php.net/?id=45731&edit=1 -- Try a CVS snapshot (PHP 5.2): http://bugs.php.net/fix.php?id=45731&r=trysnapshot52 Try a CVS snapshot (PHP 5.3): http://bugs.php.net/fix.php?id=45731&r=trysnapshot53 Try a CVS snapshot (PHP 6.0): http://bugs.php.net/fix.php?id=45731&r=trysnapshot60 Fixed in CVS: http://bugs.php.net/fix.php?id=45731&r=fixedcvs Fixed in release: http://bugs.php.net/fix.php?id=45731&r=alreadyfixed Need backtrace: http://bugs.php.net/fix.php?id=45731&r=needtrace Need Reproduce Script:http://bugs.php.net/fix.php?id=45731&r=needscript Try newer version:http://bugs.php.net/fix.php?id=45731&r=oldversion Not developer issue: http://bugs.php.net/fix.php?id=45731&r=support Expected behavior:http://bugs.php.net/fix.php?id=45731&r=notwrong Not enough info: http://bugs.php.net/fix.php?id=45731&r=notenoughinfo Submitted twice: http://bugs.php.net/fix.php?id=45731&r=submittedtwice register_globals: http://bugs.php.net/fix.php?id=45731&r=globals PHP 4 support discontinued: http://bugs.php.net/fix.php?id=45731&r=php4 Daylight Savings: http://bugs.php.net/fix.php?id=45731&r=dst IIS Stability:http://bugs.php.net/fix.php?id=45731&r=isapi Install GNU Sed: http://bugs.php.net/fix.php?id=45731&r=gnused Floating point limitations: http://bugs.php.net/fix.php?id=45731&r=float No Zend Extensions: http://bugs.php.net/fix.php?id=45731&r=nozend MySQL Configuration Error:http://bugs.php.net/fix.php?id=45731&r=mysqlcfg
#45731 [Opn->Ver]: child pid exit Segmentation fault with php5.2.6 in putenv
ID: 45731 Updated by: [EMAIL PROTECTED] Reported By: dharmeshmaniar at rediffmail dot com -Status: Open +Status: Verified Bug Type: Reproducible crash Operating System: CentOS 3.8 PHP Version: 5.2.6 Previous Comments: [2008-08-06 10:51:26] dharmeshmaniar at rediffmail dot com Description: When we use putenv in php5.2.6 it generates segmentation fault. Program received signal SIGSEGV, Segmentation fault. [Switching to Thread -1390175328 (LWP 13275)] 0x009ec303 in strchr () from /lib/tls/libc.so.6 This do not happen every time. The behaviour is random. We removed put_env from our code and everything is working properly. No more Segmentation fault after removing putenv. Note: We are not using any of the 3rd party apps:APC, eAccelerator, Turck MMCache, ionCube loader, Xcache, Xdebug Actual result: -- Below is the dump for gdb trace: [New Thread -1537033312 (LWP 13289)] [Thread -1211847776 (LWP 13258) exited] Program received signal SIGSEGV, Segmentation fault. [Switching to Thread -1390175328 (LWP 13275)] 0x009ec303 in strchr () from /lib/tls/libc.so.6 (gdb) where #0 0x009ec303 in strchr () from /lib/tls/libc.so.6 #1 0x009ace17 in putenv () from /lib/tls/libc.so.6 #2 0x003e2440 in php_putenv_destructor (pe=0xa325bb9c) at /usr/local/src/php-5.2.5/ext/standard/basic_functions.c:3861 #3 0x0049953c in zend_hash_destroy (ht=0xa35336cc) at /usr/local/src/php-5.2.5/Zend/zend_hash.c:526 #4 0x003e2e0a in zm_deactivate_basic (type=1, module_number=9, tsrm_ls=0xa39d1d28) at /usr/local/src/php-5.2.5/ext/standard/basic_functions.c:4179 #5 0x004940ba in module_registry_cleanup (module=0x9392988, tsrm_ls=0xa39d1d28) at /usr/local/src/php-5.2.5/Zend/zend_API.c:1968 #6 0x00499814 in zend_hash_apply (ht=0x601f80, apply_func=0x494098 , tsrm_ls=0xa39d1d28) at /usr/local/src/php-5.2.5/Zend/zend_hash.c:673 #7 0x0048f753 in zend_deactivate_modules (tsrm_ls=0xa39d1d28) at /usr/local/src/php-5.2.5/Zend/zend.c:838 #8 0x00451594 in php_request_shutdown (dummy=0x0) at /usr/local/src/php-5.2.5/main/main.c:1459 #9 0x004ff862 in php_handler (r=0x95c86c0) at /usr/local/src/php-5.2.5/sapi/apache2handler/sapi_apache2.c:471 #10 0x08085f92 in ap_run_handler () #11 0x0808635d in ap_invoke_handler () #12 0x0806feb5 in ap_process_request () #13 0x0806b96d in _start () #14 0x095c86c0 in ?? () #15 0x0004 in ?? () #16 0x095c86c0 in ?? () #17 0x095c86c0 in ?? () #18 0x0954d150 in ?? () #19 0x0954d060 in ?? () #20 0x in ?? () -- Edit this bug report at http://bugs.php.net/?id=45731&edit=1
#45732 [NEW]: odbc.defaultlrl = 0 != Passthru
From: nektarin at me dot com Operating system: Windows PHP version: 5.2.6 PHP Bug Type: *Configuration Issues Bug description: odbc.defaultlrl = 0 != Passthru Description: The information given in php.ini is wrong. Setting "odbc.defaultlrl = 0" is not the same as Passthru (as described in php.ini). To use passthru you have to comment the line containing odbc.defaultlrl. Reproduce code: --- odbc.defaultlrl = 0 Expected result: odbc.defaultlrl = PASSTHRU Actual result: -- odbc.defaultlrl = 0 (0 bytes is received for every SQL query) -- Edit bug report at http://bugs.php.net/?id=45732&edit=1 -- Try a CVS snapshot (PHP 5.2): http://bugs.php.net/fix.php?id=45732&r=trysnapshot52 Try a CVS snapshot (PHP 5.3): http://bugs.php.net/fix.php?id=45732&r=trysnapshot53 Try a CVS snapshot (PHP 6.0): http://bugs.php.net/fix.php?id=45732&r=trysnapshot60 Fixed in CVS: http://bugs.php.net/fix.php?id=45732&r=fixedcvs Fixed in release: http://bugs.php.net/fix.php?id=45732&r=alreadyfixed Need backtrace: http://bugs.php.net/fix.php?id=45732&r=needtrace Need Reproduce Script:http://bugs.php.net/fix.php?id=45732&r=needscript Try newer version:http://bugs.php.net/fix.php?id=45732&r=oldversion Not developer issue: http://bugs.php.net/fix.php?id=45732&r=support Expected behavior:http://bugs.php.net/fix.php?id=45732&r=notwrong Not enough info: http://bugs.php.net/fix.php?id=45732&r=notenoughinfo Submitted twice: http://bugs.php.net/fix.php?id=45732&r=submittedtwice register_globals: http://bugs.php.net/fix.php?id=45732&r=globals PHP 4 support discontinued: http://bugs.php.net/fix.php?id=45732&r=php4 Daylight Savings: http://bugs.php.net/fix.php?id=45732&r=dst IIS Stability:http://bugs.php.net/fix.php?id=45732&r=isapi Install GNU Sed: http://bugs.php.net/fix.php?id=45732&r=gnused Floating point limitations: http://bugs.php.net/fix.php?id=45732&r=float No Zend Extensions: http://bugs.php.net/fix.php?id=45732&r=nozend MySQL Configuration Error:http://bugs.php.net/fix.php?id=45732&r=mysqlcfg
#31419 [Com]: error_log not able to be overridden...
ID: 31419 Comment by: alexey dot portnov at gmail dot com Reported By: mfoxx at hotmail dot com Status: No Feedback Bug Type: PHP options/info functions Operating System: Debian 3.0 PHP Version: 5.0.4 New Comment: Description: FreeBSD 6.2-RELEASE apache 2.2.9, php-5.2.6, (from ports) php.ini config: /usr/local/etc/php.ini error_log = /home/sites/php_logs/php_errors.log PHP will not even write the errors to the original error_log value set in the php.ini. It will instead write them to apache's error log for that particular . The error does occur, and it DOES get logged, just not to the right file, it now gets logged to the APACHE error log file, not even the original php.ini error_log setting, which I find very strange. Reproduce code: --- Expected result: When I force a PHP error, calling a "nonexistent()" function, I should get the error logged into my /home/sites/php_logs/php_errors.log file as specified in /usr/local/etc/php.ini Actual result: -- the error (call to undefined function) DOES get logged, but to the wrong file... it gets logged to the Apache error file for that particular . It doesn't even get logged to the original php.ini file's setting for error_log. PHP writes errors in $error_log only when I run scripts from CLI: php /path-to/script.php Previous Comments: [2005-07-26 01:00:04] 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". [2005-07-18 19:26:03] [EMAIL PROTECTED] Using latest CVS snapshot (5.1-dev) this works just fine, provided the path passed to error_log is such that the webserver can write into it. And I don't get any PHP errors in the apache logs either. Check the permissions. And if that wasn't the cause, check what php.ini file is used (shown in phpinfo() output) and do a 'diff -u' between that and php.ini-dist from the latest CVS snapshot you're using. [2005-07-18 18:52:09] mfoxx at hotmail dot com Also, here is the VirutalHost directive from httpd.conf: ServerAdmin [EMAIL PROTECTED] DocumentRoot /www/xxx ServerName www.xxx.com ErrorLog /www/logs/xxx_com-error.log CustomLog /www/logs/xxx_com-access.log common The apache error log I keep referring to, that is getting the PHP errors logged to it, is the one specified above at "ErrorLog", xxx_com-error.log. [2005-07-18 18:46:41] mfoxx at hotmail dot com I just upgraded to the latest Apache 2.0.54, and PHP 5.0.4, and again, I am still having the same problem. I go into my php.ini file, and I comment out the setting for the php error_log. I then create a script like this: When I examine the output of phpinfo(), i see the value set in the ini_set() call. However, the error that occurs when I call the nonexistent() function does NOT get logged to that file (nor does the file even get created). It however does get logged to the virtual-host specific apache error log (error.log) that is set to be in the same directory. I even tried creating an empty file (thinking maybe it was the creation of the file that wasn't working) in the logs directory, and then restarting apache. Reran the script, and still, the error gets logged to the apache log for that virtual-host, not to the value set by ini_set(). [2005-02-28 23:24:39] mfoxx at hotmail dot com I just installed the new CVS snapshot of this, and I am now seeing a different behavior, but still not the correct one. Now, no matter whether I specify an error_log in php.ini or not, ALL php errors are going to the each virtual-host specific apache error log. I can't override it with a valid error_log setting in php.ini, nor can I override it in code, as my original bug report submitted that I had tried. 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/31419 -- Edit this bug report at http://bugs.php.net/?id=31419&edit=1
#45733 [NEW]: __get, __set error when undefined class member
From: twilek at twilek dot hu
Operating system: Ubuntu Linux
PHP version: 5.2.6
PHP Bug Type: Reproducible crash
Bug description: __get, __set error when undefined class member
Description:
When i call the __get or __set with an invalid member variable, the php
interpreter is crash.
If i check the name with isset() then no crash!
PHP Version 5.2.4-2ubuntu5.3 (newest version in ubuntu)
Reproduce code:
---
class Test
{
private $mVar = "var";
public function __get( $name )
{
$name = "m" . $name;
// if ( !isset( $this->$name ) ) return ""; //NO CRASH!!
return $this->$name;
}
public function __set( $name, $value )
{
$name = "m" . $name;
// if ( !isset( $this->$name ) ) return; //NO CRASH!!
$this->$name = $value;
echo( "Set: \$this->" . $name . " = " . $value );
}
}
$t = new Test;
echo "Good: " . $t->Var;
echo "Bad: " . $t->Variable;
$t->Variable = "test";
Expected result:
Good: "var"
Bad: ""
Set: $this->Variable = "test"
Actual result:
--
exit signal Segmentation fault (11)
--
Edit bug report at http://bugs.php.net/?id=45733&edit=1
--
Try a CVS snapshot (PHP 5.2):
http://bugs.php.net/fix.php?id=45733&r=trysnapshot52
Try a CVS snapshot (PHP 5.3):
http://bugs.php.net/fix.php?id=45733&r=trysnapshot53
Try a CVS snapshot (PHP 6.0):
http://bugs.php.net/fix.php?id=45733&r=trysnapshot60
Fixed in CVS: http://bugs.php.net/fix.php?id=45733&r=fixedcvs
Fixed in release:
http://bugs.php.net/fix.php?id=45733&r=alreadyfixed
Need backtrace: http://bugs.php.net/fix.php?id=45733&r=needtrace
Need Reproduce Script:http://bugs.php.net/fix.php?id=45733&r=needscript
Try newer version:http://bugs.php.net/fix.php?id=45733&r=oldversion
Not developer issue: http://bugs.php.net/fix.php?id=45733&r=support
Expected behavior:http://bugs.php.net/fix.php?id=45733&r=notwrong
Not enough info:
http://bugs.php.net/fix.php?id=45733&r=notenoughinfo
Submitted twice:
http://bugs.php.net/fix.php?id=45733&r=submittedtwice
register_globals: http://bugs.php.net/fix.php?id=45733&r=globals
PHP 4 support discontinued: http://bugs.php.net/fix.php?id=45733&r=php4
Daylight Savings: http://bugs.php.net/fix.php?id=45733&r=dst
IIS Stability:http://bugs.php.net/fix.php?id=45733&r=isapi
Install GNU Sed: http://bugs.php.net/fix.php?id=45733&r=gnused
Floating point limitations: http://bugs.php.net/fix.php?id=45733&r=float
No Zend Extensions: http://bugs.php.net/fix.php?id=45733&r=nozend
MySQL Configuration Error:http://bugs.php.net/fix.php?id=45733&r=mysqlcfg
#45711 [Fbk->Opn]: Undefined References
ID: 45711 User updated by: dmass83 at hotmail dot com Reported By: dmass83 at hotmail dot com -Status: Feedback +Status: Open Bug Type: Compile Failure Operating System: Solaris 10 PHP Version: 5.2.6 New Comment: Greetings, I tried the suggested method of removing everything from the configure except openssl and that worked. Additionally, I went through and removed any options that are no longer used with php5. Here is the new configure: ./configure \ --prefix=$APACHE_PREFIX \ --sysconfdir=$APACHE_PREFIX/conf \ --with-config-file-path=$APACHE_PREFIX/conf \ --with-pic \ --with-curl=$CURL_PREFIX \ --disable-rpath \ --enable-inline-optimization \ --with-png-dir=/usr \ --enable-gd-native-ttf \ --with-ttf \ --with-expat-dir=/usr/local \ --with-gettext \ --with-jpeg-dir=/usr/local \ --with-openssl=$OPENSSL_PREFIX \ --with-regex=system \ --with-libxml-dir \ --with-zlib \ --with-layout=GNU \ --enable-bcmath \ --enable-exif \ --enable-ftp \ --enable-magic-quotes \ --enable-safe-mode \ --enable-sockets \--enable-sysvsem \ --enable-sysvshm \ --enable-discard-path \ --enable-wddx \ --with-pear \ --with-kerberos=/usr/local/krb5 \ --with-ldap \ --with-mysql \ --with-mcrypt=$MCRYPT_PREFIX \ --with-oci8=$ORACLE_HOME \ --enable-memory-limit \ --enable-shmop \ --enable-versioning \ --enable-calendar \ --enable-dbx \ --enable-dio \ --with-xpm-dir= \ --with-freetype-dir= \ --with-apxs2=$APACHE_PREFIX/bin/apxs \ I still get the same error, therefore it is not related to any options that don't exist anymore. Thanks in advance. Previous Comments: [2008-08-05 00:58:37] [EMAIL PROTECTED] Start by reducing your configure line to something that is actually supported by PHP. You have dozens of configure options there which don't even exist anymore. Try this first: # rm config.cache && ./configure --disable-all --with-openssl [2008-08-05 00:55:14] dmass83 at hotmail dot com Description: I have encountered an issue installing php-5.2.6 on Solaris 10. -openssl-0.9.8h I have encountered the following error: Undefined first referenced symbol in file libiconv_set_relocation_prefix /usr/local/lib/libintl.so EVP_CIPHER_CTX_block_size ext/openssl/.libs/openssl.o ld: fatal: Symbol referencing errors. No output written to sapi/cli/php collect2: ld returned 1 exit status make: *** [sapi/cli/php] Error 1 I am new to the php installation and not real sure where to go from here. I am using the following for "configure": ./configure \ --prefix=$APACHE_PREFIX \ --sysconfdir=$APACHE_PREFIX/conf \ --with-config-file-path=$APACHE_PREFIX/conf \ --disable-debug \ --with-pic \ --with-curl=$CURL_PREFIX \ --disable-rpath \ --enable-inline-optimization \ --with-png-dir=/usr \ --with-gd \ --enable-gd-native-ttf \ --with-ttf \ --with-dom=/usr/local \ --with-expat-dir=/usr/local \ --with-gettext \ --with-jpeg-dir=/usr/local \ --with-openssl=$OPENSSL_PREFIX \ --with-png \ --with-regex=system \ --with-xml \ --with-zlib \ --with-layout=GNU \ --enable-bcmath \ --enable-exif \ --enable-ftp \ --enable-magic-quotes \ --enable-safe-mode \ --enable-sockets \ --enable-sysvsem \ --enable-sysvshm \ --enable-discard-path \ --enable-wddx \ --with-pear \ --with-kerberos=/usr/kerberos \ --with-ldap \ --with-mcrypt=$MCRYPT_PREFIX \ --with-oci8=$ORACLE_HOME \ --enable-memory-limit \ --enable-shmop \ --enable-versioning \ --enable-calendar \ --enable-dbx \ --enable-dio \ --enable-mcal \ --with-apxs2=$APACHE_PREFIX/bin/apxs \ I have searched many forums but have found no resolution to this issue. -- Edit this bug report at http://bugs.php.net/?id=45711&edit=1
#45717 [Asn->Csd]: Fileinfo/libmagic build fails, missing err.h and getopt.h
ID: 45717 Updated by: [EMAIL PROTECTED] Reported By: Bjorn dot Wiberg at its dot uu dot se -Status: Assigned +Status: Closed Bug Type: Compile Failure Operating System: IBM AIX 5.3 5300-08-01-0819 PHP Version: 5.3.0alpha1 Assigned To: derick New Comment: This bug has been fixed in CVS. Snapshots of the sources are packaged every three hours; this change will be in the next snapshot. You can grab the snapshot at http://snaps.php.net/. Thank you for the report, and for helping us make PHP better. Previous Comments: [2008-08-05 09:01:42] Bjorn dot Wiberg at its dot uu dot se Description: The bundled Fileinfo extension comes with libmagic sources, which look for err.h and getopt.h, which appear not to exist on IBM AIX systems. Is there perhaps a possibility of pointing to an existing libmagic installation? I have libmagic installed in /usr/local but cannot find a way to make the bundled Fileinfo extension use that one. Currently (5.2.6) we're using the Fileinfo 1.0.4 extension with the existing libmagic (file 4.21) installation, and that works fine... but would like to be able to use the Fileinfo bundled with PHP 5.3 instead. Reproduce code: --- #! /bin/sh # # Created by configure CC='gcc' \ './configure' \ '--enable-bcmath' \ '--enable-calendar' \ '--enable-cli' \ '--enable-dba' \ '--enable-dbase' \ '--enable-debug' \ '--enable-exif' \ '--enable-flatfile' \ '--enable-ftp' \ '--enable-gd-jis-conv' \ '--enable-gd-native-ttf' \ '--enable-inifile' \ '--enable-mbstring' \ '--enable-pcntl' \ '--enable-shmop' \ '--enable-soap' \ '--enable-sockets' \ '--enable-sqlite-utf8' \ '--enable-sysvmsg' \ '--enable-sysvsem' \ '--enable-sysvshm' \ '--enable-wddx' \ '--enable-zip' \ '--enable-zend-multibyte' \ '--prefix=/apache/php' \ '--with-apxs2=/apache/bin/apxs' \ '--with-bz2' \ '--with-cdb' \ '--with-curl' \ '--with-freetype-dir' \ '--with-gd' \ '--with-gdbm' \ '--with-gettext' \ '--with-jpeg-dir' \ '--with-ldap' \ '--with-libxml-dir=/usr/local' \ '--with-mime-magic' \ '--with-mysql=mysqlnd' \ '--with-mysqli=mysqlnd' \ '--with-openssl=/opt/freeware' \ '--with-pdo-mysql=mysqlnd' \ '--with-png-dir' \ '--with-ttf' \ '--with-xmlrpc' \ '--with-xpm-dir' \ '--with-xsl' \ '--with-zlib' \ '--with-zlib-dir' \ "$@" Expected result: Successful compilation. Actual result: -- ---8<--- begin excerpt ---8<--- gcc -I/home/bwiberg/rpm/BUILD/php-5.3.0alpha1/ext/fileinfo/libmagic -Iext/fileinfo/ -I/home/bwiberg/rpm/BUILD/php-5.3.0alpha1/ext/fileinfo/ -DPHP_ATOM_INC -I/home/bwiberg/rpm/BUILD/php-5.3.0alpha1/include -I/home/bwiberg/rpm/BUILD/php-5.3.0alpha1/main -I/home/bwiberg/rpm/BUILD/php-5.3.0alpha1 -I/home/bwiberg/rpm/BUILD/php-5.3.0alpha1/ext/ereg/regex -I/usr/local/include/libxml2 -I/opt/freeware/include -I/usr/local/include -I/home/bwiberg/rpm/BUILD/php-5.3.0alpha1/ext/date/lib -I/usr/X11R6/include -I/usr/include/freetype2 -I/home/bwiberg/rpm/BUILD/php-5.3.0alpha1/ext/mbstring/oniguruma -I/home/bwiberg/rpm/BUILD/php-5.3.0alpha1/ext/mbstring/libmbfl -I/home/bwiberg/rpm/BUILD/php-5.3.0alpha1/ext/mbstring/libmbfl/mbfl -I/home/bwiberg/rpm/BUILD/php-5.3.0alpha1/ext/sqlite3/libsqlite -I/home/bwiberg/rpm/BUILD/php-5.3.0alpha1/TSRM -I/home/bwiberg/rpm/BUILD/php-5.3.0alpha1/Zend -I/usr/include -g -fvisibility=hidden -O0 -Wall -c /home/bwiberg/rpm/BUILD/php-5.3.0alpha1/ext/fileinfo/libmagic/getopt_long.c -DPIC -o ext/fileinfo/libmagic/.libs/getopt_long.o /home/bwiberg/rpm/BUILD/php-5.3.0alpha1/ext/fileinfo/libmagic/getopt_long.c:33:17: error: err.h: No such file or directory /home/bwiberg/rpm/BUILD/php-5.3.0alpha1/ext/fileinfo/libmagic/getopt_long.c:35:20: error: getopt.h: No such file or directory /home/bwiberg/rpm/BUILD/php-5.3.0alpha1/ext/fileinfo/libmagic/getopt_long.c:47: warning: visibility attribute not supported in this configuration; ignored /home/bwiberg/rpm/BUILD/php-5.3.0alpha1/ext/fileinfo/libmagic/getopt_long.c:48: warning: visibility attribute not supported in this configuration; ignored /home/bwiberg/rpm/BUILD/php-5.3.0alpha1/ext/fileinfo/libmagic/getopt_long.c:49: warning: visibility attribute not supported in this configuration; ignored /home/bwiberg/rpm/BUILD/php-5.3.0alpha1/ext/fileinfo/libmagic/getopt_long.c:76: error: parse error before '__P' /home/bwiberg/rpm/BUILD/php-5.3.0alpha1/ext/fileinfo/libmagic/getopt_long.c:77: error: parse error before '__P' /home/bwiberg/rpm/BUILD/php-5.3.0alpha1/ext/fileinfo/libmagic/getopt_long.c:78: error: parse error before '__P' /home/bwiberg/rpm/BUILD/php-5.3.0alpha1/ext/fileinfo/libmagic/getopt_long.c: In function 'getopt_internal': /home/bwiberg/rpm/BUILD/php-5.3.0alpha1/ext/fileinfo/libmagic/getopt_long.c:250: warning: implicit declaration of function 'warnx' /home/bwiberg/rpm/BUILD/php-5.3.0alpha1/ext/fileinfo/libmagic/getopt_long.c: In function 'getopt_long': /home/bwiberg/
#45733 [Opn->Bgs]: __get, __set error when undefined class member
ID: 45733
Updated by: [EMAIL PROTECTED]
Reported By: twilek at twilek dot hu
-Status: Open
+Status: Bogus
Bug Type: Reproducible crash
Operating System: Ubuntu Linux
PHP Version: 5.2.6
New Comment:
Thank you for taking the time to write to us, but this is not
a bug. Please double-check the documentation available at
http://www.php.net/manual/ and the instructions on how to report
a bug at http://bugs.php.net/how-to-report.php
Using $this->$name in __get() creates an infinite recursion (it calls
__get(), which calls __get(), etc), this is expected.
Previous Comments:
[2008-08-06 14:28:02] twilek at twilek dot hu
Description:
When i call the __get or __set with an invalid member variable, the php
interpreter is crash.
If i check the name with isset() then no crash!
PHP Version 5.2.4-2ubuntu5.3 (newest version in ubuntu)
Reproduce code:
---
class Test
{
private $mVar = "var";
public function __get( $name )
{
$name = "m" . $name;
// if ( !isset( $this->$name ) ) return ""; //NO CRASH!!
return $this->$name;
}
public function __set( $name, $value )
{
$name = "m" . $name;
// if ( !isset( $this->$name ) ) return; //NO CRASH!!
$this->$name = $value;
echo( "Set: \$this->" . $name . " = " . $value );
}
}
$t = new Test;
echo "Good: " . $t->Var;
echo "Bad: " . $t->Variable;
$t->Variable = "test";
Expected result:
Good: "var"
Bad: ""
Set: $this->Variable = "test"
Actual result:
--
exit signal Segmentation fault (11)
--
Edit this bug report at http://bugs.php.net/?id=45733&edit=1
#45636 [Asn->Fbk]: fileinfo ext duplicate strndup
ID: 45636 Updated by: [EMAIL PROTECTED] Reported By: Fedora at FamilleCollet dot com -Status: Assigned +Status: Feedback Bug Type: Compile Failure Operating System: Linux (Fedora 9) PHP Version: 5.3CVS-2008-07-27 (snap) Assigned To: derick New Comment: Committed in CVS, could you please verify that it works for you too now? Previous Comments: [2008-07-29 21:03:13] [EMAIL PROTECTED] Assigned to the maintainer. [2008-07-27 15:59:37] Fedora at FamilleCollet dot com ok, thanks for the information about system lib. So, error occur with bundled lib. [2008-07-27 15:30:26] [EMAIL PROTECTED] FYI: This extension only compiles with the bundled library. There is no way to use external libmagic anymore. [2008-07-27 15:26:11] Fedora at FamilleCollet dot com First try with --enable-fileinfo=shared Second try with --enable-fileinfo=shared,/usr In fact, "file-devel" is installed, which provides libmagic.so and was used by PECL extension Fileinfo 1.0.4. The bundled extension doesn't seem tu use the system library. Regards [2008-07-27 09:13:46] [EMAIL PROTECTED] Compiles fine for me in Linux. What was the full configure line you used? 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/45636 -- Edit this bug report at http://bugs.php.net/?id=45636&edit=1
#45734 [NEW]: PHP segfaults in date_timezone_set for a timezone without ID
From: alex at kiesel dot name Operating system: Linux PHP version: 5.2.6 PHP Bug Type: Date/time related Bug description: PHP segfaults in date_timezone_set for a timezone without ID Description: PHP segfaults after issueing a WARNING. Reproduce code: --- Expected result: Warning: date_timezone_set(): Can only do this for zones with ID for now in /home/ak/timezone-segfault.php on line 4 Actual result: -- Warning: date_timezone_set(): Can only do this for zones with ID for now in /home/ak/timezone-segfault.php on line 4 Segmentation fault -- Edit bug report at http://bugs.php.net/?id=45734&edit=1 -- Try a CVS snapshot (PHP 5.2): http://bugs.php.net/fix.php?id=45734&r=trysnapshot52 Try a CVS snapshot (PHP 5.3): http://bugs.php.net/fix.php?id=45734&r=trysnapshot53 Try a CVS snapshot (PHP 6.0): http://bugs.php.net/fix.php?id=45734&r=trysnapshot60 Fixed in CVS: http://bugs.php.net/fix.php?id=45734&r=fixedcvs Fixed in release: http://bugs.php.net/fix.php?id=45734&r=alreadyfixed Need backtrace: http://bugs.php.net/fix.php?id=45734&r=needtrace Need Reproduce Script:http://bugs.php.net/fix.php?id=45734&r=needscript Try newer version:http://bugs.php.net/fix.php?id=45734&r=oldversion Not developer issue: http://bugs.php.net/fix.php?id=45734&r=support Expected behavior:http://bugs.php.net/fix.php?id=45734&r=notwrong Not enough info: http://bugs.php.net/fix.php?id=45734&r=notenoughinfo Submitted twice: http://bugs.php.net/fix.php?id=45734&r=submittedtwice register_globals: http://bugs.php.net/fix.php?id=45734&r=globals PHP 4 support discontinued: http://bugs.php.net/fix.php?id=45734&r=php4 Daylight Savings: http://bugs.php.net/fix.php?id=45734&r=dst IIS Stability:http://bugs.php.net/fix.php?id=45734&r=isapi Install GNU Sed: http://bugs.php.net/fix.php?id=45734&r=gnused Floating point limitations: http://bugs.php.net/fix.php?id=45734&r=float No Zend Extensions: http://bugs.php.net/fix.php?id=45734&r=nozend MySQL Configuration Error:http://bugs.php.net/fix.php?id=45734&r=mysqlcfg
#45734 [Com]: PHP segfaults in date_timezone_set for a timezone without ID
ID: 45734 Comment by: alex at kiesel dot name Reported By: alex at kiesel dot name Status: Open Bug Type: Date/time related Operating System: Linux PHP Version: 5.2.6 New Comment: Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 0x2b9db3cf24f0 (LWP 16473)] 0x0048d5ba in timelib_tzinfo_clone () (gdb) bt #0 0x0048d5ba in timelib_tzinfo_clone () #1 0x004702bb in zif_date_timezone_set () #2 0x005f0747 in ?? () #3 0x005e05c3 in execute () #4 0x005c130d in zend_execute_scripts () #5 0x0057f8f8 in php_execute_script () #6 0x006443f5 in main () Previous Comments: [2008-08-06 14:44:55] alex at kiesel dot name Description: PHP segfaults after issueing a WARNING. Reproduce code: --- Expected result: Warning: date_timezone_set(): Can only do this for zones with ID for now in /home/ak/timezone-segfault.php on line 4 Actual result: -- Warning: date_timezone_set(): Can only do this for zones with ID for now in /home/ak/timezone-segfault.php on line 4 Segmentation fault -- Edit this bug report at http://bugs.php.net/?id=45734&edit=1
#31419 [NoF->Opn]: error_log not able to be overridden...
ID: 31419 User updated by: php at mfoxx dot myspamkiller dot com -Reported By: mfoxx at hotmail dot com +Reported By: php at mfoxx dot myspamkiller dot com -Status: No Feedback +Status: Open Bug Type: PHP options/info functions -Operating System: Debian 3.0 +Operating System: Debian 4.0 -PHP Version: 5.0.4 +PHP Version: 5.2.0-8+etch10 New Comment: I *still* have this same problem, 3+ years later, in that if I run PHP (now 5.2.0 on Debian Etch, originally v4.x and then 5.0.4 on Debian Woody) through Apache with VirtualHosts that set apache "ErrorLog" settings in the VirtualHost, then PHP will not log it's errors to a separate ini specified (either in php.ini or with ini_set()) php error file, but will instead put the errors in the apache error log. This same problem has occured whether I use a custom build of PHP or whether I use the debian package builds (as I do now). Previous Comments: [2008-08-06 13:58:37] alexey dot portnov at gmail dot com Description: FreeBSD 6.2-RELEASE apache 2.2.9, php-5.2.6, (from ports) php.ini config: /usr/local/etc/php.ini error_log = /home/sites/php_logs/php_errors.log PHP will not even write the errors to the original error_log value set in the php.ini. It will instead write them to apache's error log for that particular . The error does occur, and it DOES get logged, just not to the right file, it now gets logged to the APACHE error log file, not even the original php.ini error_log setting, which I find very strange. Reproduce code: --- Expected result: When I force a PHP error, calling a "nonexistent()" function, I should get the error logged into my /home/sites/php_logs/php_errors.log file as specified in /usr/local/etc/php.ini Actual result: -- the error (call to undefined function) DOES get logged, but to the wrong file... it gets logged to the Apache error file for that particular . It doesn't even get logged to the original php.ini file's setting for error_log. PHP writes errors in $error_log only when I run scripts from CLI: php /path-to/script.php [2005-07-26 01:00:04] 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". [2005-07-18 19:26:03] [EMAIL PROTECTED] Using latest CVS snapshot (5.1-dev) this works just fine, provided the path passed to error_log is such that the webserver can write into it. And I don't get any PHP errors in the apache logs either. Check the permissions. And if that wasn't the cause, check what php.ini file is used (shown in phpinfo() output) and do a 'diff -u' between that and php.ini-dist from the latest CVS snapshot you're using. [2005-07-18 18:52:09] mfoxx at hotmail dot com Also, here is the VirutalHost directive from httpd.conf: ServerAdmin [EMAIL PROTECTED] DocumentRoot /www/xxx ServerName www.xxx.com ErrorLog /www/logs/xxx_com-error.log CustomLog /www/logs/xxx_com-access.log common The apache error log I keep referring to, that is getting the PHP errors logged to it, is the one specified above at "ErrorLog", xxx_com-error.log. [2005-07-18 18:46:41] mfoxx at hotmail dot com I just upgraded to the latest Apache 2.0.54, and PHP 5.0.4, and again, I am still having the same problem. I go into my php.ini file, and I comment out the setting for the php error_log. I then create a script like this: When I examine the output of phpinfo(), i see the value set in the ini_set() call. However, the error that occurs when I call the nonexistent() function does NOT get logged to that file (nor does the file even get created). It however does get logged to the virtual-host specific apache error log (error.log) that is set to be in the same directory. I even tried creating an empty file (thinking maybe it was the creation of the file that wasn't working) in the logs directory, and then restarting apache. Reran the script, and still, the error gets logged to the apache log for that virtual-host, not to the value set by ini_set(). 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/31419 -- Edit this bug report at http://bugs.php.net/?id=31419&edit=1
#45734 [Opn->Fbk]: PHP segfaults in date_timezone_set for a timezone without ID
ID: 45734 Updated by: [EMAIL PROTECTED] Reported By: alex at kiesel dot name -Status: Open +Status: Feedback Bug Type: Date/time related Operating System: Linux PHP Version: 5.2.6 New Comment: Please try using this CVS snapshot: http://snaps.php.net/php5.3-latest.tar.gz For Windows (zip): http://snaps.php.net/win32/php5.3-win32-latest.zip For Windows (installer): http://snaps.php.net/win32/php5.3-win32-installer-latest.msi Previous Comments: [2008-08-06 14:48:55] alex at kiesel dot name Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 0x2b9db3cf24f0 (LWP 16473)] 0x0048d5ba in timelib_tzinfo_clone () (gdb) bt #0 0x0048d5ba in timelib_tzinfo_clone () #1 0x004702bb in zif_date_timezone_set () #2 0x005f0747 in ?? () #3 0x005e05c3 in execute () #4 0x005c130d in zend_execute_scripts () #5 0x0057f8f8 in php_execute_script () #6 0x006443f5 in main () [2008-08-06 14:44:55] alex at kiesel dot name Description: PHP segfaults after issueing a WARNING. Reproduce code: --- Expected result: Warning: date_timezone_set(): Can only do this for zones with ID for now in /home/ak/timezone-segfault.php on line 4 Actual result: -- Warning: date_timezone_set(): Can only do this for zones with ID for now in /home/ak/timezone-segfault.php on line 4 Segmentation fault -- Edit this bug report at http://bugs.php.net/?id=45734&edit=1
#44645 [Com]: php5ts.dll Crash (x86_64 base machines)
ID: 44645 Comment by: sylvaint at sylvaint dot ca Reported By: gary dot wilson at coull dot biz Status: No Feedback Bug Type: Reproducible crash Operating System: Windows Server 2003 PHP Version: 5.2.5 New Comment: This bug is not strictly unique to X64 based machines. The Same bug appeared while using an X32 based machine and the same solution worked. Thanks Previous Comments: [2008-08-02 22:05:01] lotusxxl at gmail dot com Event viewer shows: Event Type: Error Event Source: Application Error Event Category: (100) Event ID: 1000 Date: 02/08/2008 Time: 20:17:41 User: N/A Computer: SERVERX64 Description: Faulting application httpd.exe, version 2.2.9.0, faulting module php5ts.dll, version 5.2.5.5, fault address 0xe8e4. For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp. Data: : 41 70 70 6c 69 63 61 74 Applicat 0008: 69 6f 6e 20 46 61 69 6c ion Fail 0010: 75 72 65 20 20 68 74 74 ure htt 0018: 70 64 2e 65 78 65 20 32 pd.exe 2 0020: 2e 32 2e 39 2e 30 20 69 .2.9.0 i 0028: 6e 20 70 68 70 35 74 73 n php5ts 0030: 2e 64 6c 6c 20 35 2e 32 .dll 5.2 0038: 2e 35 2e 35 20 61 74 20 .5.5 at 0040: 6f 66 66 73 65 74 20 30 offset 0 0048: 30 30 30 30 30 30 30 30 0050: 30 30 30 65 38 65 34 000e8e4 [2008-08-02 19:53:36] [EMAIL PROTECTED] "I fix this bug renaming the libmysql.dll in C:\Program Files\MySQL\MySQL Server 5.1\bin to libmysql.dll_ Now php uses libmysql.dll in c:\php directory and phpinfo shows Client Api version 5.0.19. Before showed 5.1.25rc" Ah that is definitively a good thing to do, using 5.1 DLLs with PHP binaries is wrong. About the backtrace, it seems that you are missing the debug pack, I can't see any symbols ( Error WARNING - DebugDiag was not able to locate debug symbols for php5ts.dll, so the information below may be incomplete.). [2008-08-02 18:50:58] lotusxxl at gmailc dot om I fix this bug renaming the libmysql.dll in C:\Program Files\MySQL\MySQL Server 5.1\bin to libmysql.dll_ Now php uses libmysql.dll in c:\php directory and phpinfo shows Client Api version 5.0.19 Before showed 5.1.25rc It works fine with Apache 2.2.9 x64 and PHP 5.2.5 x64 version in Windows 2003 Server Standard x64 [2008-07-28 12:08:40] neil dot smith at coull dot com [5 Jul 12:38pm UTC] [EMAIL PROTECTED] "Thank you for this bug report. To properly diagnose the problem, we need a backtrace to see what is happening behind the scenes" [3 Jun 4:11pm UTC] neil dot smith at coull dot com Debug backtrace was already provided on 3rd June as required - was that not the information you were requesting ? Please consider reactivating this bug and reviewing the backtrace provided. We can retry the repro case and provide additional information if needed, but that appears to be what was requested. [2008-07-18 21:30:55] madhav_2k at yahoo dot com i am getting the same error as well. "Httpd.exe - Application error The instruction at xxx referenced memory at xxx. The memory could not be read" This has completely halted my development and has pretty much rendered PHP on my laptop unusable. I use a Dell Latitude with a Intel Centrino Duo Core processor running Win2k Sp2. Software versions are MySQL 5.1 , Apache webserver 2.2 and Php 5.2.3 This has become a major showstopper and a quick resolution is highly desirable!! 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/44645 -- Edit this bug report at http://bugs.php.net/?id=44645&edit=1
#45735 [NEW]: preg_match fails with Segmentation Fault on capturing subpattern
From: johnston dot joshua at gmail dot com
Operating system: OS X and OpenSuSE
PHP version: 5.2.6
PHP Bug Type: PCRE related
Bug description: preg_match fails with Segmentation Fault on capturing
subpattern
Description:
When running a preg_match with a capturing subpattern against large input,
php crashes with a Segmentation Fault
I tested this on OS X and OpenSuSE, same result
OS X:~$ php -v
PHP 5.2.6 (cli) (built: Jul 15 2008 12:18:21)
OpenSuSE:~# php -v
PHP 4.4.7 (cgi) (built: May 12 2008 10:19:51)
Reproduce code:
---
http://google.ru";, 2000));
?>
Expected result:
PHP should handle the error or something other than letting PCRE crash php
Actual result:
--
jjohnston:~$ php -r 'preg_match("/http:\/\/(.)+\.ru/i",
str_repeat("http://google.ru";, 2000));'
Segmentation fault
--
Edit bug report at http://bugs.php.net/?id=45735&edit=1
--
Try a CVS snapshot (PHP 5.2):
http://bugs.php.net/fix.php?id=45735&r=trysnapshot52
Try a CVS snapshot (PHP 5.3):
http://bugs.php.net/fix.php?id=45735&r=trysnapshot53
Try a CVS snapshot (PHP 6.0):
http://bugs.php.net/fix.php?id=45735&r=trysnapshot60
Fixed in CVS: http://bugs.php.net/fix.php?id=45735&r=fixedcvs
Fixed in release:
http://bugs.php.net/fix.php?id=45735&r=alreadyfixed
Need backtrace: http://bugs.php.net/fix.php?id=45735&r=needtrace
Need Reproduce Script:http://bugs.php.net/fix.php?id=45735&r=needscript
Try newer version:http://bugs.php.net/fix.php?id=45735&r=oldversion
Not developer issue: http://bugs.php.net/fix.php?id=45735&r=support
Expected behavior:http://bugs.php.net/fix.php?id=45735&r=notwrong
Not enough info:
http://bugs.php.net/fix.php?id=45735&r=notenoughinfo
Submitted twice:
http://bugs.php.net/fix.php?id=45735&r=submittedtwice
register_globals: http://bugs.php.net/fix.php?id=45735&r=globals
PHP 4 support discontinued: http://bugs.php.net/fix.php?id=45735&r=php4
Daylight Savings: http://bugs.php.net/fix.php?id=45735&r=dst
IIS Stability:http://bugs.php.net/fix.php?id=45735&r=isapi
Install GNU Sed: http://bugs.php.net/fix.php?id=45735&r=gnused
Floating point limitations: http://bugs.php.net/fix.php?id=45735&r=float
No Zend Extensions: http://bugs.php.net/fix.php?id=45735&r=nozend
MySQL Configuration Error:http://bugs.php.net/fix.php?id=45735&r=mysqlcfg
#45044 [Ver]: relative paths not resolved correctly
ID: 45044 Updated by: [EMAIL PROTECTED] Reported By: felipensp at gmail dot com Status: Verified Bug Type: Scripting Engine problem Operating System: FreeBSD 6.3 PHP Version: 5.3CVS-2008-05-20 (CVS) New Comment: I can verify this on OS X 10.4 with this reproducable script http://trash.chregu.tv/include-bug.php.txt Previous Comments: [2008-07-24 19:27:21] [EMAIL PROTECTED] virtual_file_ex() assume that if realpath() does not return NULL the file exists. The problem is that FreeBSD's realpath() will not return NULL if all components *but the last* exist. So php_resolve_path()/tsrm_realpath() will pass /usr/home/felipe/foo.php to virtual_file_ex() and assume it exists. [2008-05-30 20:19:00] [EMAIL PROTECTED] status->verified [2008-05-20 02:44:43] [EMAIL PROTECTED] [EMAIL PROTECTED]:/home$ cd felipe [EMAIL PROTECTED]:/home/felipe$ mkdir test [EMAIL PROTECTED]:/home/felipe$ cd test [EMAIL PROTECTED]:/home/felipe/test$ echo "" >> test.php [EMAIL PROTECTED]:/home/felipe/test$ echo "foobar" >> foo.php [EMAIL PROTECTED]:/home/felipe/test$ cd .. [EMAIL PROTECTED]:/home/felipe$ ../cellog/workspace/php5/sapi/cli/php test/test.php foobar I am on linux. I suspect zend_resolve_path is broken on FreeBSD and this is part of the segfault you described in phar on IRC. Can we find another FreeBSDer to confirm this? [2008-05-20 01:27:00] felipensp at gmail dot com Description: I've created: ~/test ~/test/test.php: ~/test/foo.php: "foobar" Due this issue all tests (.phpt) that uses include/require fails. PS: This works fine on 5_2. Reproduce code: --- [EMAIL PROTECTED] ~]$ php5/sapi/cli/php test/test.php Expected result: foobar Actual result: -- Warning: include_once(/usr/home/felipe/foo.php): failed to open stream: No such file or directory in /usr/home/felipe/test/test.php on line 6 Warning: include_once(): Failed opening 'foo.php' for inclusion (include_path='.:/usr/local/lib/php') in /usr/home/felipe/test/test.php on line 6 -- Edit this bug report at http://bugs.php.net/?id=45044&edit=1
#45044 [Ver->Ctl]: relative paths not resolved correctly
ID: 45044 Updated by: [EMAIL PROTECTED] Reported By: felipensp at gmail dot com -Status: Verified +Status: Critical Bug Type: Scripting Engine problem Operating System: FreeBSD 6.3 PHP Version: 5.3CVS-2008-05-20 (CVS) -Assigned To: +Assigned To: dmitry New Comment: Dmitry, could you take a look at this, plz? Previous Comments: [2008-08-06 15:33:28] [EMAIL PROTECTED] I can verify this on OS X 10.4 with this reproducable script http://trash.chregu.tv/include-bug.php.txt [2008-07-24 19:27:21] [EMAIL PROTECTED] virtual_file_ex() assume that if realpath() does not return NULL the file exists. The problem is that FreeBSD's realpath() will not return NULL if all components *but the last* exist. So php_resolve_path()/tsrm_realpath() will pass /usr/home/felipe/foo.php to virtual_file_ex() and assume it exists. [2008-05-30 20:19:00] [EMAIL PROTECTED] status->verified [2008-05-20 02:44:43] [EMAIL PROTECTED] [EMAIL PROTECTED]:/home$ cd felipe [EMAIL PROTECTED]:/home/felipe$ mkdir test [EMAIL PROTECTED]:/home/felipe$ cd test [EMAIL PROTECTED]:/home/felipe/test$ echo "" >> test.php [EMAIL PROTECTED]:/home/felipe/test$ echo "foobar" >> foo.php [EMAIL PROTECTED]:/home/felipe/test$ cd .. [EMAIL PROTECTED]:/home/felipe$ ../cellog/workspace/php5/sapi/cli/php test/test.php foobar I am on linux. I suspect zend_resolve_path is broken on FreeBSD and this is part of the segfault you described in phar on IRC. Can we find another FreeBSDer to confirm this? [2008-05-20 01:27:00] felipensp at gmail dot com Description: I've created: ~/test ~/test/test.php: ~/test/foo.php: "foobar" Due this issue all tests (.phpt) that uses include/require fails. PS: This works fine on 5_2. Reproduce code: --- [EMAIL PROTECTED] ~]$ php5/sapi/cli/php test/test.php Expected result: foobar Actual result: -- Warning: include_once(/usr/home/felipe/foo.php): failed to open stream: No such file or directory in /usr/home/felipe/test/test.php on line 6 Warning: include_once(): Failed opening 'foo.php' for inclusion (include_path='.:/usr/local/lib/php') in /usr/home/felipe/test/test.php on line 6 -- Edit this bug report at http://bugs.php.net/?id=45044&edit=1
#45736 [NEW]: Include parameter name in debug_backtrace
From: RQuadling at GMail dot com Operating system: n/a PHP version: 5.3.0alpha1 PHP Bug Type: Feature/Change Request Bug description: Include parameter name in debug_backtrace Description: The debug_backtrace() provides very useful information when debugging. Could it be possible to extend the args element within the trace so that rather than just the index of the argument being shown, show the actual text that was used? Either as a key to the args array or as a new array which could be array_combined in userland. Whilst the file and line values help in getting the code that made the call, if the call is over many lines ... fn( var1, // comments var2, // comments var3 // comments ); sort of thing, then the line number is the end of the call. The output below shows line 15 which is ... ); Not a lot of use and using regex with all the potential of unmatched ) etc. is a pain. Thanks for looking if you do. Richard Quadling. Reproduce code: --- 'www.php.net', 'Docs' => 'docs.php.net', 'Wiki' => 'wiki.php.net', ); fn ( $a_Sites ); Expected result: Array ( [0] => Array ( [file] => C:\dbt.php [line] => 13 [function] => fn [args] => Array ( [$a_Sites] => Array ( [Web] => www.php.net [Docs] => docs.php.net [Wiki] => wiki.php.net ) ) ) ) or Array ( [0] => Array ( [file] => C:\dbt.php [line] => 15 [function] => fn [args] => Array ( [0] => Array ( [Web] => www.php.net [Docs] => docs.php.net [Wiki] => wiki.php.net ) ) [params] => Array ( [0] => $a_Sites ) ) ) Actual result: -- Array ( [0] => Array ( [file] => C:\dbt.php [line] => 15 [function] => fn [args] => Array ( [0] => Array ( [Web] => www.php.net [Docs] => docs.php.net [Wiki] => wiki.php.net ) ) ) ) -- Edit bug report at http://bugs.php.net/?id=45736&edit=1 -- Try a CVS snapshot (PHP 5.2): http://bugs.php.net/fix.php?id=45736&r=trysnapshot52 Try a CVS snapshot (PHP 5.3): http://bugs.php.net/fix.php?id=45736&r=trysnapshot53 Try a CVS snapshot (PHP 6.0): http://bugs.php.net/fix.php?id=45736&r=trysnapshot60 Fixed in CVS: http://bugs.php.net/fix.php?id=45736&r=fixedcvs Fixed in release: http://bugs.php.net/fix.php?id=45736&r=alreadyfixed Need backtrace: http://bugs.php.net/fix.php?id=45736&r=needtrace Need Reproduce Script:http://bugs.php.net/fix.php?id=45736&r=needscript Try newer version:http://bugs.php.net/fix.php?id=45736&r=oldversion Not developer issue: http://bugs.php.net/fix.php?id=45736&r=support Expected behavior:http://bugs.php.net/fix.php?id=45736&r=notwrong Not enough info: http://bugs.php.net/fix.php?id=45736&r=notenoughinfo Submitted twice: http://bugs.php.net/fix.php?id=45736&r=submittedtwice register_globals: http://bugs.php.net/fix.php?id=45736&r=globals PHP 4 support discontinued: http://bugs.php.net/fix.php?id=45736&r=php4 Daylight Savings: http://bugs.php.net/fix.php?id=45736&r=dst IIS Stability:http://bugs.php.net/fix.php?id=45736&r=isapi Install GNU Sed: http://bugs.php.net/fix.php?id=45736&r=gnused Floating point limitations: http://bugs.php.net/fix.php?id=45736&r=float No Zend Extensions: http://bugs.php.net/fix.php?id=45736&r=nozend MySQL Configuration Error:http://bugs.php.net/fix.php?id=45736&r=mysqlcfg
#45599 [Ver->Bgs]: strip_tags() truncates rest of string with invalid attribute
ID: 45599
Updated by: [EMAIL PROTECTED]
Reported By: david at grudl dot com
-Status: Verified
+Status: Bogus
Bug Type: Strings related
Operating System: *
PHP Version: 5.2CVS, 5.3CVS, 6CVS (2008-07-22)
New Comment:
Thank you for taking the time to write to us, but this is not
a bug. Please double-check the documentation available at
http://www.php.net/manual/ and the instructions on how to report
a bug at http://bugs.php.net/how-to-report.php
The parser continues until it founds the end of the tag, which can not
be in an attribute value (XML allows all characters except [%&'] in
attribute values).
In the given examples the attribute value never terminates and the end
of the tag is never found, which causes the rest of the string to be
truncated.
This change as been made to fix the following bug:
http://bugs.php.net/bug.php?id=40432
Previous Comments:
[2008-07-30 04:42:40] jet at synth-tec dot com
I am having the same problem. If an attribute has an extra quote in
it, will cut off all the text afterwards.
Example Input:
strip_tags('
text before link
http://google.com"";>google.com
text after link
test 1
test 2
')
Expected Output:
-
text before link
text after link
test 1
test 2
Actual Output:
--
text before link
Note, I do not have this problem in PHP 5.0.4 or previous versions
[2008-07-22 23:37:57] david at grudl dot com
Description:
Problematic backslash in HTML attribute (bug exists since PHP 5.2.2)
Reproduce code:
---
1)
echo strip_tags('Hello World');
2) this case is not HTML valid, but who cares...
echo strip_tags('Hello World');
Expected result:
Hello World
(in both cases)
Actual result:
--
Hello
(in both cases)
--
Edit this bug report at http://bugs.php.net/?id=45599&edit=1
#45735 [Opn->Ver]: preg_match fails with Segmentation Fault on capturing subpattern
ID: 45735
Updated by: [EMAIL PROTECTED]
Reported By: johnston dot joshua at gmail dot com
-Status: Open
+Status: Verified
Bug Type: PCRE related
Operating System: *
PHP Version: 5.2CVS, 5.3CVS, 6CVS (2008-08-06)
Previous Comments:
[2008-08-06 15:22:20] johnston dot joshua at gmail dot com
Description:
When running a preg_match with a capturing subpattern against large
input, php crashes with a Segmentation Fault
I tested this on OS X and OpenSuSE, same result
OS X:~$ php -v
PHP 5.2.6 (cli) (built: Jul 15 2008 12:18:21)
OpenSuSE:~# php -v
PHP 4.4.7 (cgi) (built: May 12 2008 10:19:51)
Reproduce code:
---
http://google.ru";,
2000));
?>
Expected result:
PHP should handle the error or something other than letting PCRE crash
php
Actual result:
--
jjohnston:~$ php -r 'preg_match("/http:\/\/(.)+\.ru/i",
str_repeat("http://google.ru";, 2000));'
Segmentation fault
--
Edit this bug report at http://bugs.php.net/?id=45735&edit=1
#45735 [Ver]: preg_match fails with Segmentation Fault on capturing subpattern
ID: 45735
Updated by: [EMAIL PROTECTED]
Reported By: johnston dot joshua at gmail dot com
Status: Verified
Bug Type: PCRE related
Operating System: *
PHP Version: 5.2CVS, 5.3CVS, 6CVS (2008-08-06)
New Comment:
==18470== Stack overflow in thread 1: can't grow stack to 0x7FE601FA8
==18470==
==18470== Process terminating with default action of signal 11
(SIGSEGV)
==18470== Access not within mapped region at address 0x7FE601FA8
==18470==at 0x4358C0: match (pcre_exec.c:403)
==18470== Stack overflow in thread 1: can't grow stack to 0x7FE601EC8
Previous Comments:
[2008-08-06 15:22:20] johnston dot joshua at gmail dot com
Description:
When running a preg_match with a capturing subpattern against large
input, php crashes with a Segmentation Fault
I tested this on OS X and OpenSuSE, same result
OS X:~$ php -v
PHP 5.2.6 (cli) (built: Jul 15 2008 12:18:21)
OpenSuSE:~# php -v
PHP 4.4.7 (cgi) (built: May 12 2008 10:19:51)
Reproduce code:
---
http://google.ru";,
2000));
?>
Expected result:
PHP should handle the error or something other than letting PCRE crash
php
Actual result:
--
jjohnston:~$ php -r 'preg_match("/http:\/\/(.)+\.ru/i",
str_repeat("http://google.ru";, 2000));'
Segmentation fault
--
Edit this bug report at http://bugs.php.net/?id=45735&edit=1
#45735 [Ver->Fbk]: preg_match fails with Segmentation Fault on capturing subpattern
ID: 45735
Updated by: [EMAIL PROTECTED]
Reported By: johnston dot joshua at gmail dot com
-Status: Verified
+Status: Feedback
Bug Type: PCRE related
Operating System: *
PHP Version: 5.2CVS, 5.3CVS, 6CVS (2008-08-06)
New Comment:
Setting pcre.backtrack_limit = 1 prevents the crash for me.
Previous Comments:
[2008-08-06 16:36:27] [EMAIL PROTECTED]
==18470== Stack overflow in thread 1: can't grow stack to 0x7FE601FA8
==18470==
==18470== Process terminating with default action of signal 11
(SIGSEGV)
==18470== Access not within mapped region at address 0x7FE601FA8
==18470==at 0x4358C0: match (pcre_exec.c:403)
==18470== Stack overflow in thread 1: can't grow stack to 0x7FE601EC8
[2008-08-06 15:22:20] johnston dot joshua at gmail dot com
Description:
When running a preg_match with a capturing subpattern against large
input, php crashes with a Segmentation Fault
I tested this on OS X and OpenSuSE, same result
OS X:~$ php -v
PHP 5.2.6 (cli) (built: Jul 15 2008 12:18:21)
OpenSuSE:~# php -v
PHP 4.4.7 (cgi) (built: May 12 2008 10:19:51)
Reproduce code:
---
http://google.ru";,
2000));
?>
Expected result:
PHP should handle the error or something other than letting PCRE crash
php
Actual result:
--
jjohnston:~$ php -r 'preg_match("/http:\/\/(.)+\.ru/i",
str_repeat("http://google.ru";, 2000));'
Segmentation fault
--
Edit this bug report at http://bugs.php.net/?id=45735&edit=1
#45731 [Ver->Fbk]: child pid exit Segmentation fault in putenv
ID: 45731 Updated by: [EMAIL PROTECTED] Reported By: dharmeshmaniar at rediffmail dot com -Status: Verified +Status: Feedback Bug Type: Reproducible crash Operating System: CentOS 3.8 PHP Version: 5.2.6 New Comment: Please try using this CVS snapshot: http://snaps.php.net/php5.2-latest.tar.gz For Windows (zip): http://snaps.php.net/win32/php5.2-win32-latest.zip For Windows (installer): http://snaps.php.net/win32/php5.2-win32-installer-latest.msi Previous Comments: [2008-08-06 10:51:26] dharmeshmaniar at rediffmail dot com Description: When we use putenv in php5.2.6 it generates segmentation fault. Program received signal SIGSEGV, Segmentation fault. [Switching to Thread -1390175328 (LWP 13275)] 0x009ec303 in strchr () from /lib/tls/libc.so.6 This do not happen every time. The behaviour is random. We removed put_env from our code and everything is working properly. No more Segmentation fault after removing putenv. Note: We are not using any of the 3rd party apps:APC, eAccelerator, Turck MMCache, ionCube loader, Xcache, Xdebug Actual result: -- Below is the dump for gdb trace: [New Thread -1537033312 (LWP 13289)] [Thread -1211847776 (LWP 13258) exited] Program received signal SIGSEGV, Segmentation fault. [Switching to Thread -1390175328 (LWP 13275)] 0x009ec303 in strchr () from /lib/tls/libc.so.6 (gdb) where #0 0x009ec303 in strchr () from /lib/tls/libc.so.6 #1 0x009ace17 in putenv () from /lib/tls/libc.so.6 #2 0x003e2440 in php_putenv_destructor (pe=0xa325bb9c) at /usr/local/src/php-5.2.5/ext/standard/basic_functions.c:3861 #3 0x0049953c in zend_hash_destroy (ht=0xa35336cc) at /usr/local/src/php-5.2.5/Zend/zend_hash.c:526 #4 0x003e2e0a in zm_deactivate_basic (type=1, module_number=9, tsrm_ls=0xa39d1d28) at /usr/local/src/php-5.2.5/ext/standard/basic_functions.c:4179 #5 0x004940ba in module_registry_cleanup (module=0x9392988, tsrm_ls=0xa39d1d28) at /usr/local/src/php-5.2.5/Zend/zend_API.c:1968 #6 0x00499814 in zend_hash_apply (ht=0x601f80, apply_func=0x494098 , tsrm_ls=0xa39d1d28) at /usr/local/src/php-5.2.5/Zend/zend_hash.c:673 #7 0x0048f753 in zend_deactivate_modules (tsrm_ls=0xa39d1d28) at /usr/local/src/php-5.2.5/Zend/zend.c:838 #8 0x00451594 in php_request_shutdown (dummy=0x0) at /usr/local/src/php-5.2.5/main/main.c:1459 #9 0x004ff862 in php_handler (r=0x95c86c0) at /usr/local/src/php-5.2.5/sapi/apache2handler/sapi_apache2.c:471 #10 0x08085f92 in ap_run_handler () #11 0x0808635d in ap_invoke_handler () #12 0x0806feb5 in ap_process_request () #13 0x0806b96d in _start () #14 0x095c86c0 in ?? () #15 0x0004 in ?? () #16 0x095c86c0 in ?? () #17 0x095c86c0 in ?? () #18 0x0954d150 in ?? () #19 0x0954d060 in ?? () #20 0x in ?? () -- Edit this bug report at http://bugs.php.net/?id=45731&edit=1
#45711 [Opn->Fbk]: Undefined References
ID: 45711 Updated by: [EMAIL PROTECTED] Reported By: dmass83 at hotmail dot com -Status: Open +Status: Feedback Bug Type: Compile Failure Operating System: Solaris 10 PHP Version: 5.2.6 New Comment: Get the latest snapshot here: http://snaps.php.net/php5.2-latest.tar.gz And provide the shortest possible configure line which still causes the problem. Note: You STILL have non-existing options there. There is a note in the end of the configure run which points to the unknown options.. Previous Comments: [2008-08-06 14:28:16] dmass83 at hotmail dot com Greetings, I tried the suggested method of removing everything from the configure except openssl and that worked. Additionally, I went through and removed any options that are no longer used with php5. Here is the new configure: ./configure \ --prefix=$APACHE_PREFIX \ --sysconfdir=$APACHE_PREFIX/conf \ --with-config-file-path=$APACHE_PREFIX/conf \ --with-pic \ --with-curl=$CURL_PREFIX \ --disable-rpath \ --enable-inline-optimization \ --with-png-dir=/usr \ --enable-gd-native-ttf \ --with-ttf \ --with-expat-dir=/usr/local \ --with-gettext \ --with-jpeg-dir=/usr/local \ --with-openssl=$OPENSSL_PREFIX \ --with-regex=system \ --with-libxml-dir \ --with-zlib \ --with-layout=GNU \ --enable-bcmath \ --enable-exif \ --enable-ftp \ --enable-magic-quotes \ --enable-safe-mode \ --enable-sockets \--enable-sysvsem \ --enable-sysvshm \ --enable-discard-path \ --enable-wddx \ --with-pear \ --with-kerberos=/usr/local/krb5 \ --with-ldap \ --with-mysql \ --with-mcrypt=$MCRYPT_PREFIX \ --with-oci8=$ORACLE_HOME \ --enable-memory-limit \ --enable-shmop \ --enable-versioning \ --enable-calendar \ --enable-dbx \ --enable-dio \ --with-xpm-dir= \ --with-freetype-dir= \ --with-apxs2=$APACHE_PREFIX/bin/apxs \ I still get the same error, therefore it is not related to any options that don't exist anymore. Thanks in advance. [2008-08-05 00:58:37] [EMAIL PROTECTED] Start by reducing your configure line to something that is actually supported by PHP. You have dozens of configure options there which don't even exist anymore. Try this first: # rm config.cache && ./configure --disable-all --with-openssl [2008-08-05 00:55:14] dmass83 at hotmail dot com Description: I have encountered an issue installing php-5.2.6 on Solaris 10. -openssl-0.9.8h I have encountered the following error: Undefined first referenced symbol in file libiconv_set_relocation_prefix /usr/local/lib/libintl.so EVP_CIPHER_CTX_block_size ext/openssl/.libs/openssl.o ld: fatal: Symbol referencing errors. No output written to sapi/cli/php collect2: ld returned 1 exit status make: *** [sapi/cli/php] Error 1 I am new to the php installation and not real sure where to go from here. I am using the following for "configure": ./configure \ --prefix=$APACHE_PREFIX \ --sysconfdir=$APACHE_PREFIX/conf \ --with-config-file-path=$APACHE_PREFIX/conf \ --disable-debug \ --with-pic \ --with-curl=$CURL_PREFIX \ --disable-rpath \ --enable-inline-optimization \ --with-png-dir=/usr \ --with-gd \ --enable-gd-native-ttf \ --with-ttf \ --with-dom=/usr/local \ --with-expat-dir=/usr/local \ --with-gettext \ --with-jpeg-dir=/usr/local \ --with-openssl=$OPENSSL_PREFIX \ --with-png \ --with-regex=system \ --with-xml \ --with-zlib \ --with-layout=GNU \ --enable-bcmath \ --enable-exif \ --enable-ftp \ --enable-magic-quotes \ --enable-safe-mode \ --enable-sockets \ --enable-sysvsem \ --enable-sysvshm \ --enable-discard-path \ --enable-wddx \ --with-pear \ --with-kerberos=/usr/kerberos \ --with-ldap \ --with-mcrypt=$MCRYPT_PREFIX \ --with-oci8=$ORACLE_HOME \ --enable-memory-limit \ --enable-shmop \ --enable-versioning \ --enable-calendar \ --enable-dbx \ --enable-dio \ --enable-mcal \ --with-apxs2=$APACHE_PREFIX/bin/apxs \ I have searched many forums but have found no resolution to this issue. -- Edit this bug report at http://bugs.php.net/?id=45711&edit=1
#45706 [Opn->Fbk]: Serializing of ArrayIterator extended Objects
ID: 45706
Updated by: [EMAIL PROTECTED]
Reported By: bobka at bigfoot dot com
-Status: Open
+Status: Feedback
Bug Type: Session related
Operating System: Linux
PHP Version: 5.2.6
New Comment:
Does this happen only with the ArrayIterator?
Previous Comments:
[2008-08-06 08:45:17] bobka at bigfoot dot com
The bug is somewhere in "php" or "php_bin" serialize_handler that is
used to serialize _SESSION
[2008-08-06 08:38:06] bobka at bigfoot dot com
To define the class before session_start is the usual case. But if two
applications use the same session data, where one part of session data
is shared between applications and other are used exclusively, you don't
have to include *all* class definitions in each application. In this
case the part of session data, that is not defined, should be preserved
and not corrupted, shouldn't it?
[2008-08-04 12:51:07] [EMAIL PROTECTED]
The class has to be defined before session_start() call.
Documentation for session seems to lack this little note..
[2008-08-04 12:20:45] bobka at bigfoot dot com
Description:
This is a very special case.
If an object of class that extends ArrayIterator is saved in session
data and the class definition is not available in a later call, then the
session data gets corrupted.
Call the example code three times to see the bug.
Reproduce code:
---
".print_r($_SESSION,true)."";
// already initialized?
if (!isset($_SESSION['first']))
{
// bug occures only if the class is not defined later
// and only if it extends ArrayIterator
class my_array extends ArrayIterator
{
}
$_SESSION['first'] = "The first entry is ok";
$t = new my_array();
// at least one member needed
$t[] = "this gets lost";
$_SESSION['killer'] = $t;
$_SESSION['this_is_going_to_be_messed_up'] = "The value of the entry
is ok";
$_SESSION['last'] = "The last entry is ok";
}
Expected result:
At first call:
Array
(
)
At second call:
Array
(
[first] => The first entry is ok
[killer] => __PHP_Incomplete_Class Object
(
[__PHP_Incomplete_Class_Name] => my_array
[0] => this gets lost
)
[this_is_going_to_be_messed_up] => The value of the entry is ok
[last] => The last entry is ok
)
At third call
Array
(
[first] => The first entry is ok
[killer] => __PHP_Incomplete_Class Object
(
[__PHP_Incomplete_Class_Name] => my_array
[0] => this gets lost
)
[this_is_going_to_be_messed_up] => The value of the entry is ok
[last] => The last entry is ok
)
Actual result:
--
At third call:
Notice: session_start() [function.session-start]: Unexpected end of
serialized data in test.php on line 3
Array
(
[first] => The first entry is ok
[killer] =>
[}this_is_going_to_be_messed_up] => The value of the entry is ok
[last] => The last entry is ok
)
--
Edit this bug report at http://bugs.php.net/?id=45706&edit=1
#45706 [Com]: Serializing of ArrayIterator extended Objects
ID: 45706
Comment by: bobka at bigfoot dot com
Reported By: bobka at bigfoot dot com
Status: Feedback
Bug Type: Session related
Operating System: Linux
PHP Version: 5.2.6
New Comment:
Yes, it happens only with classes that extend ArrayIterator
Previous Comments:
[2008-08-06 16:46:37] [EMAIL PROTECTED]
Does this happen only with the ArrayIterator?
[2008-08-06 08:45:17] bobka at bigfoot dot com
The bug is somewhere in "php" or "php_bin" serialize_handler that is
used to serialize _SESSION
[2008-08-06 08:38:06] bobka at bigfoot dot com
To define the class before session_start is the usual case. But if two
applications use the same session data, where one part of session data
is shared between applications and other are used exclusively, you don't
have to include *all* class definitions in each application. In this
case the part of session data, that is not defined, should be preserved
and not corrupted, shouldn't it?
[2008-08-04 12:51:07] [EMAIL PROTECTED]
The class has to be defined before session_start() call.
Documentation for session seems to lack this little note..
[2008-08-04 12:20:45] bobka at bigfoot dot com
Description:
This is a very special case.
If an object of class that extends ArrayIterator is saved in session
data and the class definition is not available in a later call, then the
session data gets corrupted.
Call the example code three times to see the bug.
Reproduce code:
---
".print_r($_SESSION,true)."";
// already initialized?
if (!isset($_SESSION['first']))
{
// bug occures only if the class is not defined later
// and only if it extends ArrayIterator
class my_array extends ArrayIterator
{
}
$_SESSION['first'] = "The first entry is ok";
$t = new my_array();
// at least one member needed
$t[] = "this gets lost";
$_SESSION['killer'] = $t;
$_SESSION['this_is_going_to_be_messed_up'] = "The value of the entry
is ok";
$_SESSION['last'] = "The last entry is ok";
}
Expected result:
At first call:
Array
(
)
At second call:
Array
(
[first] => The first entry is ok
[killer] => __PHP_Incomplete_Class Object
(
[__PHP_Incomplete_Class_Name] => my_array
[0] => this gets lost
)
[this_is_going_to_be_messed_up] => The value of the entry is ok
[last] => The last entry is ok
)
At third call
Array
(
[first] => The first entry is ok
[killer] => __PHP_Incomplete_Class Object
(
[__PHP_Incomplete_Class_Name] => my_array
[0] => this gets lost
)
[this_is_going_to_be_messed_up] => The value of the entry is ok
[last] => The last entry is ok
)
Actual result:
--
At third call:
Notice: session_start() [function.session-start]: Unexpected end of
serialized data in test.php on line 3
Array
(
[first] => The first entry is ok
[killer] =>
[}this_is_going_to_be_messed_up] => The value of the entry is ok
[last] => The last entry is ok
)
--
Edit this bug report at http://bugs.php.net/?id=45706&edit=1
#45731 [Fbk]: child pid exit Segmentation fault in putenv
ID: 45731 Updated by: [EMAIL PROTECTED] Reported By: dharmeshmaniar at rediffmail dot com Status: Feedback Bug Type: Reproducible crash Operating System: CentOS 3.8 PHP Version: 5.2.6 New Comment: This is reproducible with ZTS enabled. Reproduce code: $ cat /var/www/putenv.php $ ab -n 1 -c 100 http://localhost/putenv.php Previous Comments: [2008-08-06 16:41:34] [EMAIL PROTECTED] Please try using this CVS snapshot: http://snaps.php.net/php5.2-latest.tar.gz For Windows (zip): http://snaps.php.net/win32/php5.2-win32-latest.zip For Windows (installer): http://snaps.php.net/win32/php5.2-win32-installer-latest.msi [2008-08-06 10:51:26] dharmeshmaniar at rediffmail dot com Description: When we use putenv in php5.2.6 it generates segmentation fault. Program received signal SIGSEGV, Segmentation fault. [Switching to Thread -1390175328 (LWP 13275)] 0x009ec303 in strchr () from /lib/tls/libc.so.6 This do not happen every time. The behaviour is random. We removed put_env from our code and everything is working properly. No more Segmentation fault after removing putenv. Note: We are not using any of the 3rd party apps:APC, eAccelerator, Turck MMCache, ionCube loader, Xcache, Xdebug Actual result: -- Below is the dump for gdb trace: [New Thread -1537033312 (LWP 13289)] [Thread -1211847776 (LWP 13258) exited] Program received signal SIGSEGV, Segmentation fault. [Switching to Thread -1390175328 (LWP 13275)] 0x009ec303 in strchr () from /lib/tls/libc.so.6 (gdb) where #0 0x009ec303 in strchr () from /lib/tls/libc.so.6 #1 0x009ace17 in putenv () from /lib/tls/libc.so.6 #2 0x003e2440 in php_putenv_destructor (pe=0xa325bb9c) at /usr/local/src/php-5.2.5/ext/standard/basic_functions.c:3861 #3 0x0049953c in zend_hash_destroy (ht=0xa35336cc) at /usr/local/src/php-5.2.5/Zend/zend_hash.c:526 #4 0x003e2e0a in zm_deactivate_basic (type=1, module_number=9, tsrm_ls=0xa39d1d28) at /usr/local/src/php-5.2.5/ext/standard/basic_functions.c:4179 #5 0x004940ba in module_registry_cleanup (module=0x9392988, tsrm_ls=0xa39d1d28) at /usr/local/src/php-5.2.5/Zend/zend_API.c:1968 #6 0x00499814 in zend_hash_apply (ht=0x601f80, apply_func=0x494098 , tsrm_ls=0xa39d1d28) at /usr/local/src/php-5.2.5/Zend/zend_hash.c:673 #7 0x0048f753 in zend_deactivate_modules (tsrm_ls=0xa39d1d28) at /usr/local/src/php-5.2.5/Zend/zend.c:838 #8 0x00451594 in php_request_shutdown (dummy=0x0) at /usr/local/src/php-5.2.5/main/main.c:1459 #9 0x004ff862 in php_handler (r=0x95c86c0) at /usr/local/src/php-5.2.5/sapi/apache2handler/sapi_apache2.c:471 #10 0x08085f92 in ap_run_handler () #11 0x0808635d in ap_invoke_handler () #12 0x0806feb5 in ap_process_request () #13 0x0806b96d in _start () #14 0x095c86c0 in ?? () #15 0x0004 in ?? () #16 0x095c86c0 in ?? () #17 0x095c86c0 in ?? () #18 0x0954d150 in ?? () #19 0x0954d060 in ?? () #20 0x in ?? () -- Edit this bug report at http://bugs.php.net/?id=45731&edit=1
#45599 [Bgs]: strip_tags() truncates rest of string with invalid attribute
ID: 45599
User updated by: david at grudl dot com
Reported By: david at grudl dot com
Status: Bogus
Bug Type: Strings related
Operating System: *
PHP Version: 5.2CVS, 5.3CVS, 6CVS (2008-07-22)
New Comment:
Character \ is allowed in tag attribute, so strip_tags('Hello World') leading to "Hello" (without "World") is bug.
Previous Comments:
[2008-08-06 16:30:17] [EMAIL PROTECTED]
Thank you for taking the time to write to us, but this is not
a bug. Please double-check the documentation available at
http://www.php.net/manual/ and the instructions on how to report
a bug at http://bugs.php.net/how-to-report.php
The parser continues until it founds the end of the tag, which can not
be in an attribute value (XML allows all characters except [%&'] in
attribute values).
In the given examples the attribute value never terminates and the end
of the tag is never found, which causes the rest of the string to be
truncated.
This change as been made to fix the following bug:
http://bugs.php.net/bug.php?id=40432
[2008-07-30 04:42:40] jet at synth-tec dot com
I am having the same problem. If an attribute has an extra quote in
it, will cut off all the text afterwards.
Example Input:
strip_tags('
text before link
http://google.com"";>google.com
text after link
test 1
test 2
')
Expected Output:
-
text before link
text after link
test 1
test 2
Actual Output:
--
text before link
Note, I do not have this problem in PHP 5.0.4 or previous versions
[2008-07-22 23:37:57] david at grudl dot com
Description:
Problematic backslash in HTML attribute (bug exists since PHP 5.2.2)
Reproduce code:
---
1)
echo strip_tags('Hello World');
2) this case is not HTML valid, but who cares...
echo strip_tags('Hello World');
Expected result:
Hello World
(in both cases)
Actual result:
--
Hello
(in both cases)
--
Edit this bug report at http://bugs.php.net/?id=45599&edit=1
#45636 [Fbk->Opn]: fileinfo ext duplicate strndup
ID: 45636 User updated by: Fedora at FamilleCollet dot com Reported By: Fedora at FamilleCollet dot com -Status: Feedback +Status: Open Bug Type: Compile Failure Operating System: Linux (Fedora 9) PHP Version: 5.3CVS-2008-07-27 (snap) New Comment: Check with lastest snapshot (200808061630) Build ok. Thanks for the fix. Regards Previous Comments: [2008-08-06 14:43:21] [EMAIL PROTECTED] Committed in CVS, could you please verify that it works for you too now? [2008-07-29 21:03:13] [EMAIL PROTECTED] Assigned to the maintainer. [2008-07-27 15:59:37] Fedora at FamilleCollet dot com ok, thanks for the information about system lib. So, error occur with bundled lib. [2008-07-27 15:30:26] [EMAIL PROTECTED] FYI: This extension only compiles with the bundled library. There is no way to use external libmagic anymore. [2008-07-27 15:26:11] Fedora at FamilleCollet dot com First try with --enable-fileinfo=shared Second try with --enable-fileinfo=shared,/usr In fact, "file-devel" is installed, which provides libmagic.so and was used by PECL extension Fileinfo 1.0.4. The bundled extension doesn't seem tu use the system library. Regards 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/45636 -- Edit this bug report at http://bugs.php.net/?id=45636&edit=1
#45735 [Fbk->Opn]: preg_match fails with Segmentation Fault on capturing subpattern
ID: 45735
User updated by: johnston dot joshua at gmail dot com
Reported By: johnston dot joshua at gmail dot com
-Status: Feedback
+Status: Open
Bug Type: PCRE related
Operating System: *
PHP Version: 5.2CVS, 5.3CVS, 6CVS (2008-08-06)
New Comment:
In my case the parens were unnecessary so I removed them which stopped
the Seg Fault.
Is there a way to handle this error from inside php?
If I set pcre.backtrack_limit to any other value will it always seg
fault if it is too low?
Previous Comments:
[2008-08-06 16:38:30] [EMAIL PROTECTED]
Setting pcre.backtrack_limit = 1 prevents the crash for me.
[2008-08-06 16:36:27] [EMAIL PROTECTED]
==18470== Stack overflow in thread 1: can't grow stack to 0x7FE601FA8
==18470==
==18470== Process terminating with default action of signal 11
(SIGSEGV)
==18470== Access not within mapped region at address 0x7FE601FA8
==18470==at 0x4358C0: match (pcre_exec.c:403)
==18470== Stack overflow in thread 1: can't grow stack to 0x7FE601EC8
[2008-08-06 15:22:20] johnston dot joshua at gmail dot com
Description:
When running a preg_match with a capturing subpattern against large
input, php crashes with a Segmentation Fault
I tested this on OS X and OpenSuSE, same result
OS X:~$ php -v
PHP 5.2.6 (cli) (built: Jul 15 2008 12:18:21)
OpenSuSE:~# php -v
PHP 4.4.7 (cgi) (built: May 12 2008 10:19:51)
Reproduce code:
---
http://google.ru";,
2000));
?>
Expected result:
PHP should handle the error or something other than letting PCRE crash
php
Actual result:
--
jjohnston:~$ php -r 'preg_match("/http:\/\/(.)+\.ru/i",
str_repeat("http://google.ru";, 2000));'
Segmentation fault
--
Edit this bug report at http://bugs.php.net/?id=45735&edit=1
#45599 [Bgs->Opn]: strip_tags() truncates rest of string with invalid attribute
ID: 45599
Updated by: [EMAIL PROTECTED]
Reported By: david at grudl dot com
-Status: Bogus
+Status: Open
Bug Type: Strings related
Operating System: *
PHP Version: 5.2CVS, 5.3CVS, 6CVS (2008-07-22)
Previous Comments:
[2008-08-06 16:52:29] david at grudl dot com
Character \ is allowed in tag attribute, so strip_tags('Hello World') leading to "Hello" (without "World") is bug.
[2008-08-06 16:30:17] [EMAIL PROTECTED]
Thank you for taking the time to write to us, but this is not
a bug. Please double-check the documentation available at
http://www.php.net/manual/ and the instructions on how to report
a bug at http://bugs.php.net/how-to-report.php
The parser continues until it founds the end of the tag, which can not
be in an attribute value (XML allows all characters except [%&'] in
attribute values).
In the given examples the attribute value never terminates and the end
of the tag is never found, which causes the rest of the string to be
truncated.
This change as been made to fix the following bug:
http://bugs.php.net/bug.php?id=40432
[2008-07-30 04:42:40] jet at synth-tec dot com
I am having the same problem. If an attribute has an extra quote in
it, will cut off all the text afterwards.
Example Input:
strip_tags('
text before link
http://google.com"";>google.com
text after link
test 1
test 2
')
Expected Output:
-
text before link
text after link
test 1
test 2
Actual Output:
--
text before link
Note, I do not have this problem in PHP 5.0.4 or previous versions
[2008-07-22 23:37:57] david at grudl dot com
Description:
Problematic backslash in HTML attribute (bug exists since PHP 5.2.2)
Reproduce code:
---
1)
echo strip_tags('Hello World');
2) this case is not HTML valid, but who cares...
echo strip_tags('Hello World');
Expected result:
Hello World
(in both cases)
Actual result:
--
Hello
(in both cases)
--
Edit this bug report at http://bugs.php.net/?id=45599&edit=1
#42266 [Com]: BLOB functions do not work on 64bit systems
ID: 42266 Comment by: hasul at etoscomp dot eu Reported By: karasek at ceskyserver dot cz Status: Assigned Bug Type: InterBase related Operating System: Linux 64-bit PHP Version: 5.2.4 Assigned To: abies New Comment: Hi, is this bug still open? Because BLOB is working on my system (Centos 5.2 64bit). I used php 5.2.6 from Jason Litka repo (http://www.jasonlitka.com/yum-repository/) and FirebirdSS-2.1.0.17798-0.amd64.rpm from official firebird site. Previous Comments: [2008-05-16 07:19:43] [EMAIL PROTECTED] "You're maybe right, but the bug appeared in 5.2.4 while 5.2.1. works fine, something has changed in PHP, not in firebird." I meant support as in 2nd level support not for the quality of the driver :) About the handle question, it is how they do it in firebird itself. A quick search in codesearch: http://google.com/codesearch?q=lang%3Ac+typedef+FB_API_HANDLE&hl=en&btnG=Search+Code So the problem is somewhere else, but I can't help as I don't use firebird. Let wait a bit more for the maintainers answer. [2008-05-16 06:39:45] ale dot pas at tiscali dot it You're maybe right, but the bug appeared in 5.2.4 while 5.2.1. works fine, something has changed in PHP, not in firebird. [2008-05-15 09:42:35] [EMAIL PROTECTED] Let get it assigned. It will certainly help to bring some love here. > It's a pity we choose PHP for almost all of our projects, do not make > this error, choose Java, Python or Ruby you will find a better support > there. I doubt you will find better ibase support anywhere but InterBase directly. That's the main problem here, not PHP. [2008-05-15 09:35:39] ale dot pas at tiscali dot it Ok, after 10 months this bug is sill open without a comment from a developer. This clearly proves that PHP is definitely not ready for enterprise solutions. It's a pity we choose PHP for almost all of our projects, do not make this error, choose Java, Python or Ruby you will find a better support there. [2008-02-12 05:32:12] info at olemera dot com Is the a new patch for the (64-bit version) with Firebird 2.0 ? I get an integer Error. Regards, http://www.olemera.com/loans/home-loans/integer-home-loans/ [url=http://www.olemera.com/loans/home-loans/integer-home-loans/]Olemera[/url] 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/42266 -- Edit this bug report at http://bugs.php.net/?id=42266&edit=1
#45735 [Opn]: preg_match fails with Segmentation Fault on capturing subpattern
ID: 45735
Updated by: [EMAIL PROTECTED]
Reported By: johnston dot joshua at gmail dot com
Status: Open
Bug Type: PCRE related
Operating System: *
PHP Version: 5.2CVS, 5.3CVS, 6CVS (2008-08-06)
New Comment:
That is known PCRE behavior.
See: http://manpages.courier-mta.org/htmlman3/pcrestack.3.html
Previous Comments:
[2008-08-06 17:29:06] johnston dot joshua at gmail dot com
In my case the parens were unnecessary so I removed them which stopped
the Seg Fault.
Is there a way to handle this error from inside php?
If I set pcre.backtrack_limit to any other value will it always seg
fault if it is too low?
[2008-08-06 16:38:30] [EMAIL PROTECTED]
Setting pcre.backtrack_limit = 1 prevents the crash for me.
[2008-08-06 16:36:27] [EMAIL PROTECTED]
==18470== Stack overflow in thread 1: can't grow stack to 0x7FE601FA8
==18470==
==18470== Process terminating with default action of signal 11
(SIGSEGV)
==18470== Access not within mapped region at address 0x7FE601FA8
==18470==at 0x4358C0: match (pcre_exec.c:403)
==18470== Stack overflow in thread 1: can't grow stack to 0x7FE601EC8
[2008-08-06 15:22:20] johnston dot joshua at gmail dot com
Description:
When running a preg_match with a capturing subpattern against large
input, php crashes with a Segmentation Fault
I tested this on OS X and OpenSuSE, same result
OS X:~$ php -v
PHP 5.2.6 (cli) (built: Jul 15 2008 12:18:21)
OpenSuSE:~# php -v
PHP 4.4.7 (cgi) (built: May 12 2008 10:19:51)
Reproduce code:
---
http://google.ru";,
2000));
?>
Expected result:
PHP should handle the error or something other than letting PCRE crash
php
Actual result:
--
jjohnston:~$ php -r 'preg_match("/http:\/\/(.)+\.ru/i",
str_repeat("http://google.ru";, 2000));'
Segmentation fault
--
Edit this bug report at http://bugs.php.net/?id=45735&edit=1
#45735 [Opn->Bgs]: preg_match fails with Segmentation Fault on capturing subpattern
ID: 45735
Updated by: [EMAIL PROTECTED]
Reported By: johnston dot joshua at gmail dot com
-Status: Open
+Status: Bogus
Bug Type: PCRE related
Operating System: *
PHP Version: 5.2CVS, 5.3CVS, 6CVS (2008-08-06)
New Comment:
.
Previous Comments:
[2008-08-06 17:40:08] [EMAIL PROTECTED]
That is known PCRE behavior.
See: http://manpages.courier-mta.org/htmlman3/pcrestack.3.html
[2008-08-06 17:29:06] johnston dot joshua at gmail dot com
In my case the parens were unnecessary so I removed them which stopped
the Seg Fault.
Is there a way to handle this error from inside php?
If I set pcre.backtrack_limit to any other value will it always seg
fault if it is too low?
[2008-08-06 16:38:30] [EMAIL PROTECTED]
Setting pcre.backtrack_limit = 1 prevents the crash for me.
[2008-08-06 16:36:27] [EMAIL PROTECTED]
==18470== Stack overflow in thread 1: can't grow stack to 0x7FE601FA8
==18470==
==18470== Process terminating with default action of signal 11
(SIGSEGV)
==18470== Access not within mapped region at address 0x7FE601FA8
==18470==at 0x4358C0: match (pcre_exec.c:403)
==18470== Stack overflow in thread 1: can't grow stack to 0x7FE601EC8
[2008-08-06 15:22:20] johnston dot joshua at gmail dot com
Description:
When running a preg_match with a capturing subpattern against large
input, php crashes with a Segmentation Fault
I tested this on OS X and OpenSuSE, same result
OS X:~$ php -v
PHP 5.2.6 (cli) (built: Jul 15 2008 12:18:21)
OpenSuSE:~# php -v
PHP 4.4.7 (cgi) (built: May 12 2008 10:19:51)
Reproduce code:
---
http://google.ru";,
2000));
?>
Expected result:
PHP should handle the error or something other than letting PCRE crash
php
Actual result:
--
jjohnston:~$ php -r 'preg_match("/http:\/\/(.)+\.ru/i",
str_repeat("http://google.ru";, 2000));'
Segmentation fault
--
Edit this bug report at http://bugs.php.net/?id=45735&edit=1
#45735 [Bgs]: preg_match fails with Segmentation Fault on capturing subpattern
ID: 45735 User updated by: johnston dot joshua at gmail dot com Reported By: johnston dot joshua at gmail dot com Status: Bogus Bug Type: PCRE related Operating System: * PHP Version: 5.2CVS, 5.3CVS, 6CVS (2008-08-06) New Comment: I understand it is a known pcre behavior but does that mean it is ok to let pcre trigger a segmentation fault? Is there no possible way to handle this failure gracefully? Previous Comments: [2008-08-06 17:40:50] [EMAIL PROTECTED] . [2008-08-06 17:40:08] [EMAIL PROTECTED] That is known PCRE behavior. See: http://manpages.courier-mta.org/htmlman3/pcrestack.3.html [2008-08-06 17:29:06] johnston dot joshua at gmail dot com In my case the parens were unnecessary so I removed them which stopped the Seg Fault. Is there a way to handle this error from inside php? If I set pcre.backtrack_limit to any other value will it always seg fault if it is too low? [2008-08-06 16:38:30] [EMAIL PROTECTED] Setting pcre.backtrack_limit = 1 prevents the crash for me. [2008-08-06 16:36:27] [EMAIL PROTECTED] ==18470== Stack overflow in thread 1: can't grow stack to 0x7FE601FA8 ==18470== ==18470== Process terminating with default action of signal 11 (SIGSEGV) ==18470== Access not within mapped region at address 0x7FE601FA8 ==18470==at 0x4358C0: match (pcre_exec.c:403) ==18470== Stack overflow in thread 1: can't grow stack to 0x7FE601EC8 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/45735 -- Edit this bug report at http://bugs.php.net/?id=45735&edit=1
#45636 [Opn->Csd]: fileinfo ext duplicate strndup
ID: 45636 Updated by: [EMAIL PROTECTED] Reported By: Fedora at FamilleCollet dot com -Status: Open +Status: Closed Bug Type: Compile Failure Operating System: Linux (Fedora 9) PHP Version: 5.3CVS-2008-07-27 (snap) New Comment: This bug has been fixed in CVS. Snapshots of the sources are packaged every three hours; this change will be in the next snapshot. You can grab the snapshot at http://snaps.php.net/. Thank you for the report, and for helping us make PHP better. Previous Comments: [2008-08-06 17:05:53] Fedora at FamilleCollet dot com Check with lastest snapshot (200808061630) Build ok. Thanks for the fix. Regards [2008-08-06 14:43:21] [EMAIL PROTECTED] Committed in CVS, could you please verify that it works for you too now? [2008-07-29 21:03:13] [EMAIL PROTECTED] Assigned to the maintainer. [2008-07-27 15:59:37] Fedora at FamilleCollet dot com ok, thanks for the information about system lib. So, error occur with bundled lib. [2008-07-27 15:30:26] [EMAIL PROTECTED] FYI: This extension only compiles with the bundled library. There is no way to use external libmagic anymore. 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/45636 -- Edit this bug report at http://bugs.php.net/?id=45636&edit=1
#45406 [Csd]: [PATCH] session.serialize_handler declared by shared extension fails
ID: 45406 Updated by: [EMAIL PROTECTED] Reported By: oleg dot grenrus at dynamoid dot com Status: Closed Bug Type: Session related Operating System: * PHP Version: 5.2.6 New Comment: Backported to 5.2.7 Previous Comments: [2008-08-04 06:22:05] [EMAIL PROTECTED] This bug has been fixed in CVS. Snapshots of the sources are packaged every three hours; this change will be in the next snapshot. You can grab the snapshot at http://snaps.php.net/. Thank you for the report, and for helping us make PHP better. [2008-07-01 16:29:05] oleg dot grenrus at dynamoid dot com Description: PHP warnings are issued as PHP Warning: session_start(): Unknown session.serialize_handler. Failed todecode session object. in - on line 2 PHP Warning: Unknown: Unknown session.serialize_handler. Failed to encode session object. in Unknown on line 0 These happen because session module try's to find serialize_handler before serializer module is loaded (and registered itself as valid serialize_handler) There are check for save_handler (PS(mod) != NULL) in the PHP_RINIT_FUNCTION(session), but not for PS(serializer) I wrote a short patch http://phadej.daug.net/php-5.2.6-serialize_handler.patch against php-5.2.6. PHP_RINIT_FUNCTION seems to be same in php-5.2 and php-5.3 brances. Reproduce code: --- php -d "extension=wddx.so" -d 'session.serialize_handler=wddx' <<\EOF EOF Expected result: bool(true) Actual result: -- PHP Warning: session_start(): Unknown session.serialize_handler. Failed to decode session object. in /home/phadej/dev/php-sess-bug/- on line 2 bool(true) PHP Warning: Unknown: Unknown session.serialize_handler. Failed to encode session object. in Unknown on line 0 -- Edit this bug report at http://bugs.php.net/?id=45406&edit=1
#45737 [NEW]: mysql_fetch_assoc returns null on duplicate columns in a join and select *
From: quixote at toysmakeuspowerful dot com Operating system: Windows and Linux PHP version: 5.2.6 PHP Bug Type: MySQL related Bug description: mysql_fetch_assoc returns null on duplicate columns in a join and select * Description: select a.*,b.* from a join b on a.id=b.id; mysql_fetch_assoc() returns null for id. select a.*,b.*,a.id from a join b on a.id=b.id; Works, since last instance overrides previous as per documentation. However it should be in the first case that b.id would get returned automatically, since it is the second instance. -- Edit bug report at http://bugs.php.net/?id=45737&edit=1 -- Try a CVS snapshot (PHP 5.2): http://bugs.php.net/fix.php?id=45737&r=trysnapshot52 Try a CVS snapshot (PHP 5.3): http://bugs.php.net/fix.php?id=45737&r=trysnapshot53 Try a CVS snapshot (PHP 6.0): http://bugs.php.net/fix.php?id=45737&r=trysnapshot60 Fixed in CVS: http://bugs.php.net/fix.php?id=45737&r=fixedcvs Fixed in release: http://bugs.php.net/fix.php?id=45737&r=alreadyfixed Need backtrace: http://bugs.php.net/fix.php?id=45737&r=needtrace Need Reproduce Script:http://bugs.php.net/fix.php?id=45737&r=needscript Try newer version:http://bugs.php.net/fix.php?id=45737&r=oldversion Not developer issue: http://bugs.php.net/fix.php?id=45737&r=support Expected behavior:http://bugs.php.net/fix.php?id=45737&r=notwrong Not enough info: http://bugs.php.net/fix.php?id=45737&r=notenoughinfo Submitted twice: http://bugs.php.net/fix.php?id=45737&r=submittedtwice register_globals: http://bugs.php.net/fix.php?id=45737&r=globals PHP 4 support discontinued: http://bugs.php.net/fix.php?id=45737&r=php4 Daylight Savings: http://bugs.php.net/fix.php?id=45737&r=dst IIS Stability:http://bugs.php.net/fix.php?id=45737&r=isapi Install GNU Sed: http://bugs.php.net/fix.php?id=45737&r=gnused Floating point limitations: http://bugs.php.net/fix.php?id=45737&r=float No Zend Extensions: http://bugs.php.net/fix.php?id=45737&r=nozend MySQL Configuration Error:http://bugs.php.net/fix.php?id=45737&r=mysqlcfg
#45738 [NEW]: php stream error
From: ajai73 at hotmail dot com Operating system: vista 32 PHP version: 4.4.8 PHP Bug Type: Streams related Bug description: php stream error Description: i have an error when loading csv widget on my site. -partial widget -navbar not show Error state. Warning: fopen(/path_to/uploads/content.csv) [function.fopen]: failed to open stream: No such file or directory in /mnt/w0500/d41/s13/b02e8f66/www/lnaelite.org/widget/libs/Content.php on line 6 Warning: fgetcsv(): supplied argument is not a valid stream resource in /mnt/w0500/d41/s13/b02e8f66/www/lnaelite.org/widget/libs/Content.php on line 7 Warning: fclose(): supplied argument is not a valid stream resource in /mnt/w0500/d41/s13/b02e8f66/www/lnaelite.org/widget/libs/Content.php on line 18 Reproduce code: --- http://bugs.php.net/?id=45738&edit=1 -- Try a CVS snapshot (PHP 5.2): http://bugs.php.net/fix.php?id=45738&r=trysnapshot52 Try a CVS snapshot (PHP 5.3): http://bugs.php.net/fix.php?id=45738&r=trysnapshot53 Try a CVS snapshot (PHP 6.0): http://bugs.php.net/fix.php?id=45738&r=trysnapshot60 Fixed in CVS: http://bugs.php.net/fix.php?id=45738&r=fixedcvs Fixed in release: http://bugs.php.net/fix.php?id=45738&r=alreadyfixed Need backtrace: http://bugs.php.net/fix.php?id=45738&r=needtrace Need Reproduce Script:http://bugs.php.net/fix.php?id=45738&r=needscript Try newer version:http://bugs.php.net/fix.php?id=45738&r=oldversion Not developer issue: http://bugs.php.net/fix.php?id=45738&r=support Expected behavior:http://bugs.php.net/fix.php?id=45738&r=notwrong Not enough info: http://bugs.php.net/fix.php?id=45738&r=notenoughinfo Submitted twice: http://bugs.php.net/fix.php?id=45738&r=submittedtwice register_globals: http://bugs.php.net/fix.php?id=45738&r=globals PHP 4 support discontinued: http://bugs.php.net/fix.php?id=45738&r=php4 Daylight Savings: http://bugs.php.net/fix.php?id=45738&r=dst IIS Stability:http://bugs.php.net/fix.php?id=45738&r=isapi Install GNU Sed: http://bugs.php.net/fix.php?id=45738&r=gnused Floating point limitations: http://bugs.php.net/fix.php?id=45738&r=float No Zend Extensions: http://bugs.php.net/fix.php?id=45738&r=nozend MySQL Configuration Error:http://bugs.php.net/fix.php?id=45738&r=mysqlcfg
#34381 [Opn]: nl2br() should have an option for XHTML/HTML compatible BR element
ID: 34381 Updated by: [EMAIL PROTECTED] Reported By: jasper at bryant-greene dot name Status: Open Bug Type: Feature/Change Request Operating System: N/A PHP Version: 5.0.4 New Comment: Try this patch (for PHP_5_3 cvs): http://www.phpfi.com/340833 This adds a new parameter to nl2br() to enable nl2br() to return "" insted of "" Previous Comments: [2008-03-13 15:28:15] holliwell at gmx dot net I second jasper's suggestion of an optional switch, if one wants plain old html 4.01 , default should be xhtml-compliant . [2007-10-25 17:25:29] nospam at wanted dot by dot me That tags generated by the NL2BR() function are not compliant with an HTML 4.01 DOCTYPE is now commented on otherwise legitimate pages by the W3C validator. I fail to understand why a new function wasn't invented to handle output intended for XHTML instead of simply amending this function intended for HTML output as of PHP 4.0.5 onwards - did you assume that everyone would immediately migrate to XHTML at that time? The only alternative for PHP authors who wish to generate HTML compliant markup at present are left with is to write their own version of NL2BR() and replace it's use throughout their script bases and libraries when generating HTML. A solution PHP could offer is a new function that enables PHP authors to specify whether an HTML page is being output or whether it is something else, like XHTML or XML. This function (or method) could be invoked at the same time as the DOCTYPE is output. It could also be defaulted by a php.ini switch quite handily. [2005-09-05 22:56:35] jasper at bryant-greene dot name Description: The nl2br() function, as of PHP 4.0.5, always inserts . However, in HTML a complaint parser should understand this as > (it is the SGML SHORTTAG syntax). For example, this code: | This isa paragraph should be rendered as this in an HTML document[1]: | This is | >a paragraph Therefore, in order to maintain compatibility with both HTML and XHTML specifications, I request that an option be added to nl2br() that allows XHTML or HTML compliant br elements to be inserted. [1] for more information, see http://hixie.ch/advocacy/xhtml -- Edit this bug report at http://bugs.php.net/?id=34381&edit=1
#45730 [Opn->Bgs]: missing dll files
ID: 45730 Updated by: [EMAIL PROTECTED] Reported By: jwebb at obeliskinternational dot com -Status: Open +Status: Bogus Bug Type: *General Issues Operating System: windows xp PHP Version: 5.2.6 New Comment: You need the windows zip releases. Previous Comments: [2008-08-06 10:23:48] jwebb at obeliskinternational dot com Description: i just downloaded php-5.2.6.tar.gz and have been through the manual setup for iis5 on my xp machine But there are no dll files in my download So how can i finish setting up IIS to use isapi ?? -- Edit this bug report at http://bugs.php.net/?id=45730&edit=1
#45739 [NEW]: what is this for?
From: zhangguiyin at gmail dot com
Operating system: linux
PHP version: 5.2.6
PHP Bug Type: Regexps related
Bug description: what is this for?
Description:
array preg_grep ( string pattern, array input [, int flags] )
it wont work when array input is a Multi-dimensional array
Reproduce code:
---
array preg_grep ( string pattern, array input [, int flags] )
it wont work when array input is a Multi-dimensional array
Expected result:
array(1) { [2]=> string(3) "cpu" }
Actual result:
--
array(0) { }
--
Edit bug report at http://bugs.php.net/?id=45739&edit=1
--
Try a CVS snapshot (PHP 5.2):
http://bugs.php.net/fix.php?id=45739&r=trysnapshot52
Try a CVS snapshot (PHP 5.3):
http://bugs.php.net/fix.php?id=45739&r=trysnapshot53
Try a CVS snapshot (PHP 6.0):
http://bugs.php.net/fix.php?id=45739&r=trysnapshot60
Fixed in CVS: http://bugs.php.net/fix.php?id=45739&r=fixedcvs
Fixed in release:
http://bugs.php.net/fix.php?id=45739&r=alreadyfixed
Need backtrace: http://bugs.php.net/fix.php?id=45739&r=needtrace
Need Reproduce Script:http://bugs.php.net/fix.php?id=45739&r=needscript
Try newer version:http://bugs.php.net/fix.php?id=45739&r=oldversion
Not developer issue: http://bugs.php.net/fix.php?id=45739&r=support
Expected behavior:http://bugs.php.net/fix.php?id=45739&r=notwrong
Not enough info:
http://bugs.php.net/fix.php?id=45739&r=notenoughinfo
Submitted twice:
http://bugs.php.net/fix.php?id=45739&r=submittedtwice
register_globals: http://bugs.php.net/fix.php?id=45739&r=globals
PHP 4 support discontinued: http://bugs.php.net/fix.php?id=45739&r=php4
Daylight Savings: http://bugs.php.net/fix.php?id=45739&r=dst
IIS Stability:http://bugs.php.net/fix.php?id=45739&r=isapi
Install GNU Sed: http://bugs.php.net/fix.php?id=45739&r=gnused
Floating point limitations: http://bugs.php.net/fix.php?id=45739&r=float
No Zend Extensions: http://bugs.php.net/fix.php?id=45739&r=nozend
MySQL Configuration Error:http://bugs.php.net/fix.php?id=45739&r=mysqlcfg
#45739 [Opn]: what is this for?
ID: 45739
User updated by: zhangguiyin at gmail dot com
Reported By: zhangguiyin at gmail dot com
Status: Open
Bug Type: Regexps related
Operating System: linux
PHP Version: 5.2.6
New Comment:
The following is the examplt code
Previous Comments:
[2008-08-07 03:55:25] zhangguiyin at gmail dot com
Description:
array preg_grep ( string pattern, array input [, int flags] )
it wont work when array input is a Multi-dimensional array
Reproduce code:
---
array preg_grep ( string pattern, array input [, int flags] )
it wont work when array input is a Multi-dimensional array
Expected result:
array(1) { [2]=> string(3) "cpu" }
Actual result:
--
array(0) { }
--
Edit this bug report at http://bugs.php.net/?id=45739&edit=1
#45132 [Com]: escapeshellcmd removes swedish characters
ID: 45132
Comment by: tstarling at wikimedia dot org
Reported By: tom at collegit dot se
Status: Assigned
Bug Type: Strings related
Operating System: Linux www2 2.6.24.7-grsec
PHP Version: 5.2.6
Assigned To: iliaa
New Comment:
The issue is that previously 8-bit clean locales, like "C", are now
being validated for whatever character set they supposedly are, with
characters above 127 being removed.
The suggested fix, here and on
https://bugzilla.wikimedia.org/show_bug.cgi?id=14944 , appears to reopen
whatever security vulnerability it was that the patch fixed in the first
place.
$ LANG=C php eval.php
> setlocale(LC_CTYPE, 'en_US.UTF-8')
> echo escapeshellarg("\xC3\x96")
'Ö'
> passthru('locale')
LANG=C
LC_CTYPE="C"
LC_NUMERIC="C"
LC_TIME="C"
LC_COLLATE="C"
LC_MONETARY="C"
LC_MESSAGES="C"
LC_PAPER="C"
LC_NAME="C"
LC_ADDRESS="C"
LC_TELEPHONE="C"
LC_MEASUREMENT="C"
LC_IDENTIFICATION="C"
LC_ALL=
Because the environment variable LC_CTYPE is not set by setlocale(),
the spawned shell sees the old character set, not the new one. So the
shell can be passed an argument escaped for the wrong character set,
potentially opening a vulnerability.
I'm assuming that the attack scenario for this vulnerability is where
an attacker can set environment variables such as LANG to a vulnerable
character set, before starting PHP. Because if an attacker can set
environment variables during execution of a script, the bug is not
fixed. But in that case you're probably screwed anyway.
Previous Comments:
[2008-05-29 23:09:34] [EMAIL PROTECTED]
Try using:
setlocale(LC_CTYPE, "UTF8", "en_US.UTF-8");
[2008-05-29 22:55:58] [EMAIL PROTECTED]
Ilia, did that fix get into actual release..?
[2008-05-29 22:53:54] tom at collegit dot se
Description:
Since my web server got updated to 5.2.6 escapeshellcmd() has started
removing all Swedish characters from user input. Looks a lot like bug
#44564.
'./configure' '--with-apxs2=/usr/local/apache/bin/apxs' '--with-
mysql=/usr/local/mysql' '--enable-debug=no' '--enable-track-vars' '--
with-imap=/usr/local/imap' '--with-mcrypt' '--with-mhash' '--with-dom'
'--with-gd' '--with-jpeg-dir' '--enable-ftp' '--with-freetype-
dir=/usr' '--with-zlib-dir=/usr' '--with-png-dir=/usr' '--enable-
dbase' '--with-gettext' '--with-zlib' '--enable-inline-optimization'
'--with-openssl' '--with-curl' '--with-curlwrappers' '--enable-exif'
'--with-imap-ssl' '--enable-bcmath' '--enable-mbstring' '--enable-
wddx' '--with-xsl' '--with-pdo-mysql=/usr/local/mysql' '--with-
mysqli=/usr/local/mysql/bin/mysql_config' '--enable-soap' '--enable-
sockets' '--with-mssql' '--with-ming' '--with-pspell' '--enable-
calendar' '--enable-bcmath' '--with-t1lib' '--with-
pgsql=/usr/local/pgsql' '--with-tidy' '--enable-zip'
Reproduce code:
---
$test = "åäö";
$test = (escapeshellcmd($test));
print $test;
Expected result:
"åäö"
Actual result:
--
""
--
Edit this bug report at http://bugs.php.net/?id=45132&edit=1
#36398 [Opn]: "make test" should be fully scriptable [Do not prompt for email address]
ID: 36398
User updated by: nickj-phpbugs at nickj dot org
Reported By: nickj-phpbugs at nickj dot org
Status: Open
Bug Type: Feature/Change Request
Operating System: Linux
PHP Version: 5.1.2
New Comment:
Yes, it would be nice if after calling for a patch, if the submitted
patch could be either accepted or rejected, rather than ignored for 2.5
years ;-)
I gather run-tests.php may be rewritten sometime (I think I saw it
somewhere as a possible Google SoC project), and if this happens
including functionality to send the results using an email address
specified on the command line could be nice to include. Presumably then
someone (or some bit of software) can collate the output and help to
pinpoint consistent problems, or when things break on particular
platforms, and give a summary of the health of PHP on different
platforms, and so forth.
Previous Comments:
[2008-08-05 12:21:28] Bjorn dot Wiberg at its dot uu dot se
Shouldn't the patch be incorporated into run-tests.php? As it has been
around for a long time.
[2006-02-17 06:53:53] nickj-phpbugs at nickj dot org
No worries! There is a patch now available for this at:
http://www.files.nickj.org/php/run-tests-patch.txt
Also, the run-test.php script is not currently E_STRICT clean. I have
included in the patch some simple changes I had to make to get it to run
with E_STRICT enabled.
Note that there are 3 remaining things I did not know how to fix for
clean E_STRICT output, so even with this patch run-tests.php is still
not totally E_STRICT clean. The three remaining things were:
1) Was this error:
Error! type: 8; File:
/root/php-5.1-dev/php5.1-200602150330/run-tests.php; Line: 93; Message:
ob_end_clean(): failed to delete buffer. No buffer to delete.
... which was generated by this line under E_STRICT:
while(@ob_end_clean());
2) Was the use of date() in this line:
$output_file = $CUR_DIR . '/php_test_results_' . date('Ymd_Hi') .
'.txt';
Which gives this error under E_STRICT:
Error! type: 2048; File:
/root/php-5.1-dev/php5.1-200602150330/run-tests.php; Line: 246; Message:
date(): It is not safe to rely on the system's timezone settings. Please
use the date.timezone setting, the TZ environment variable or the
date_default_timezone_set() function. In case you used any of those
methods and you are still getting this warning, you most likely
misspelled the timezone identifier. We selected 'Australia/Melbourne'
for 'EST/11.0/DST' instead
Not sure how to say "just use the system default" nowadays, given that
we probably shouldn't hard-code a timezone, as users from anywhere could
be running this test script.
3) Was the "select_stream" call in the "system_with_timeout" function,
which generates this error under E_STRICT:
Error! type: 2; File:
/root/php-5.1-dev/php5.1-200602150330/run-tests.php; Line: 893; Message:
stream_select(): 230 is not a valid stream resource
[2006-02-16 07:35:06] [EMAIL PROTECTED]
Thanks, I've just fixed the typo.
As for the request - I'm pretty sure you can add it to run-tests.php
and send a patch for us to review.
Though, I don't know if we really want php-qa list to get filled with
results of the tests executed automatically every X minutes.
[2006-02-16 06:19:59] nickj-phpbugs at nickj dot org
Thank you - that sort of helps, and I realise now that I should have
phrased my request better (so I have reopened, and rephrased it to
hopefully be clearer).
What I mostly wanted was to run the tests, and then email the results
off, without prompting.
I can enable/disable the prompt now with:
TEST_PHP_ARGS="-q" make test
And I can see the available options with:
TEST_PHP_EXECUTABLE="/root/tmp/php-5.1-dev/php5.1-200602150330/sapi/cli/php"
sapi/cli/php run-tests.php --help
However, unless I am mistaken, I can't see any option to specify the
email address to use, such as for example:
TEST_PHP_ARGS="--email-results [EMAIL PROTECTED]" make test
My underlying assumption here is that you folks want and use the output
of "make test" in some way. If that's not the case, then of course,
don't make it an option. However, if you are using this, then wouldn't
it be good to be able to script it so that it could automatically email
off the results of make test, without having to prompt the user? Then
you could (for example) have the build farm automatically email the
qa.php.net list with the results of "make test" after every build of
every snapshot. Maybe you already do this in some other way, but if not,
it seems like it could be useful to me.
You could also update http://qa.php.net/running-tests.php with this
information, so that people could "set and forget" to run the tests and
email the results, wi
