To allow us to do this:
$a = array('a' => 1, 'b' => 2);
$b = $a;
$c = $a;
array_merge_recursive($b, $c);
The old check has a problem, because $b & $c are write only copies *src_entry
== *dest_entry condition is met and the function incorrectly terminates
claiming a circular reference. We cann
On June 4, 2003 01:12 pm, Moriyoshi Koizumi wrote:
If (srclen + (chunks + 1) * endlen + 1) overflows and results in a <0 number,
the result of the multiplication inside safe_emalloc would still be negative
and we'll trigger the integer overflow check.
Ilia
> "Ilia Alshanetsky" <[EMAIL PROTECTE
On April 3, 2003 07:35 pm, Marcus Börger wrote:
> >+ if (dstW < 0 || dstH < 0 || srcW < 0 || srcH < 0) {
>
> Why is that not <=
Why not indeed...
I'll make the change.
Ilia
--
PHP CVS Mailing List (http://www.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php
MFH?
On March 28, 2003 08:10 am, David Hill wrote:
> ddhillFri Mar 28 08:10:52 2003 EDT
>
> Modified files:
> /php4/ext/standardparsedate.y
> Log:
> Changing parsedate to be reentrant using pure_parser (Dave)
--
PHP CVS Mailing List (http://www.php.net/)
To uns
Please MFH this & previous fix to the PHP_4_3 branch.
Ilia
On March 27, 2003 12:31 am, Ian Holsman wrote:
> ianh Thu Mar 27 00:31:49 2003 EDT
>
> Modified files:
> /php4/sapi/apache2handler sapi_apache2.c
> Log:
> change to a pool-based bucket instead of a transient one.
>
>
>
On March 25, 2003 08:39 am, Moriyoshi Koizumi wrote:
> Hmm..? my patch has not been committed yet, that was only in the list
> archives. Although I don't see no problem with your fix, I think you'd
> be better off asking the assignee how your patch is before the commit.
Hmm... I must've missed you
On March 25, 2003 03:40 am, Moriyoshi Koizumi wrote:
> Why did you commit another fix though my patch was already there and maybe
> Ian was looking into this issue? :)
Because the existing patch (latest CVS) did not work, when posting 30k or more
data it would usually stop after 7.8k or so. The r
On February 25, 2003 01:51 pm, Moriyoshi Koizumi wrote:
> moriyoshi Tue Feb 25 13:51:23 2003 EDT
>
> Modified files: (Branch: PHP_4_3)
> /php4/ext/standardhtml.c
> Log:
> MFH(r1.72): int / long change
> Fixed bug #22301 (htmlspecialchars crashes on Tru64
On February 11, 2003 02:41 pm, Marcus Börger wrote:
> Why will i hide command line arguments from command line, does this make
> sense?
This is there primarily for daemon and other long term PHP process that run on
shared servers and accept command line arguments. To prevent other users from
see
The scandir should probably be moved into a file of it's own and included when
the OS does not have the function. Recent bug report #21889 appears to
indicate that this functionality is not avaliable on solaris either.
Ilia
On January 25, 2003 08:39 pm, Shane Caraveo wrote:
> shane Sat
On January 23, 2003 01:54 pm, you wrote:
> > > > > extension_loaded("pcre") or die("skip pcre extension is not
> > > available"); ?>
> >
> > This check is pointless, since without pcre the testsuit won't run.
>
> Right, I just followed the way in which other tests are written.
> Should we remove t
On January 23, 2003 01:22 pm, Moriyoshi Koizumi wrote:
> moriyoshi Thu Jan 23 13:22:56 2003 EDT
>
> Added test case for bug #21732
[snip]
> --SKIPIF--
> extension_loaded("pcre") or die("skip pcre extension is not available");
> ?>
This check is pointless, since without pcre the test
On January 18, 2003 03:20 pm, Marcus Börger wrote:
> At 20:57 18.01.2003, you wrote:
> >iliaa Sat Jan 18 14:57:25 2003 EDT
> >
> > Modified files:
> > /php4/ext/swf swf.c
> > Log:
> > Changed php_error to php_error_docref.
>
> Thanks for all this work and one question: did
On January 2, 2003 02:53 pm, Derick Rethans wrote:
> On Thu, 2 Jan 2003, Ilia Alshanetsky wrote:
> > iliaa Thu Jan 2 14:52:19 2003 EDT
> >
> > Modified files: (Branch: PHP_4_3)
> > /php4 NEWS
> > Log:
> > Bug fixing news.
>
> Is it common to have an entry for
On December 27, 2002 06:35 pm, Sascha Schumann wrote:
> On Fri, 27 Dec 2002, Ilia Alshanetsky wrote:
> > iliaa Fri Dec 27 14:59:10 2002 EDT
> >
> > Modified files:
> > /php4/sapi/apache2filtersapi_apache2.c
> > Log:
> > Fixed bug #17098 (make Apache aware that PHP sc
Small correction, it actually should say non-bundled.
On December 11, 2002 06:30 pm, Ilia Alshanetsky wrote:
> iliaa Wed Dec 11 18:30:24 2002 EDT
>
> Modified files: (Branch: PHP_4_3)
> /php4/ext/gd gd.c php_gd.h
> Log:
> MFH (make imagecreatefromxpm() unavailab
On December 6, 2002 12:53 pm, Andrey Hristov wrote:
> is it ok to have NULL keys in the arrays?
It seems to work fine. For example $array[NULL] will return the value of the
NULL key if it is available, so just from consistency standpoint so should
array_key_exists().
Ilia
>
> Best regards
> An
On December 5, 2002 10:37 am, Marcus Börger wrote:
> At 16:20 05.12.2002, Ilia A. wrote:
> >My tests have a if (!GD_BUNDLED) {}, which as I understand should cause
> > the test to be skipped in the event non-bundled GD is being used. The
> > bundled gd
>
> you m
On December 5, 2002 10:26 am, Marcus Börger wrote:
> At 16:23 05.12.2002, Ilia A. wrote:
> >array_init() can no longer fail, because it means that ZE was unable to
> >allocate memory at which point ZE will cease the script execution. So, if
> >there are any existing check
array_init() can no longer fail, because it means that ZE was unable to
allocate memory at which point ZE will cease the script execution. So, if
there are any existing checks for return value of array_init() or any other
e* memory allocation functions, those checks can be safely removed since t
My tests have a if (!GD_BUNDLED) {}, which as I understand should cause the
test to be skipped in the event non-bundled GD is being used. The bundled gd
will always fail and that is NOT an assumption, the external gd has bugs that
were fixed in bundled gd these are known so it is pointless to ma
On December 3, 2002 02:13 pm, Marcus Börger wrote:
> Ah great now it waorks!
> Should we add a test for ImageGd() then, verify it isn't a true color
> image? And what about MFH?
Adding a test for truecolor inside imageGd code seems like the best choice,
that code was not designed with truecolor i
On November 24, 2002 02:06 pm, Derick Rethans wrote:
> On Sun, 24 Nov 2002, Ilia Alshanetsky wrote:
> > iliaa Sat Nov 23 20:15:35 2002 EDT
> >
> > Modified files:
> > /php4/ext/gd/libgd gdft.c
> > Log:
> > Fixed 3 memory leaks.
> > Formatted the code to meet PHP's cod
The current tests (with your patch) work properly on my test machines here.
Could you please MFH the patch to the mhash tests.
Thanks,
Ilia
On November 22, 2002 09:07 am, Marcus Boerger wrote:
> helly Fri Nov 22 09:07:30 2002 EDT
>
> Modified files:
> /php4/ext/mhash/tests 001
uot; should be discouraged, because cli and cgi SAPI
> forcefully set the default CR handling mode to binary with setmode()
> function while apache SAPI doesn't.
>
> Moriyoshi
>
> "Ilia A." <[EMAIL PROTECTED]> wrote:
> > Are you absolutely certain that
The current test allows a clear view of what filters have failed, since the
filer type is being included in the result. Could you please submit the
failed data to QA list so I could see why the tests have failed and fix the
bug.
Ilia
On November 15, 2002 10:42 am, Marcus Börger wrote:
> The te
Are you absolutely certain that it is safe to open the magic file as text and
not binary on Windows?
Ilia
On November 15, 2002 10:13 am, Moriyoshi Koizumi wrote:
> moriyoshi Fri Nov 15 10:13:31 2002 EDT
>
> Modified files:
> /php4/ext/mime_magic mime_magic.c
> Log:
> F
If someone writes tests specifically to test safe_mode functionality, then
they can enable safe_mode inside the script itself via the --INI-- section.
At the moment we have no such tests and safe_mode breaks, needlesly numerous
tests.
Ilia
On November 14, 2002 02:46 pm, Marcus Börger wrote:
>
My mistake, I usually don't compile with TSRM since some modules do not
compile when I use experimental-zts.
Ilia
> marcus
>
> At 16:27 14.11.2002, Ilia A. wrote:
> >Which compiler are you using and with which options that gave you
> > warning(s) on these lines?
>
Which compiler are you using and with which options that gave you warning(s)
on these lines?
Ilia
On November 14, 2002 10:08 am, Marcus Boerger wrote:
> helly Thu Nov 14 10:08:40 2002 EDT
>
> Modified files:
> /php4/ext/gd/libgdgd.c
> Log:
> Fix warnings (Marcus & Pierr
an error we can capture via $php_errormsg, the output
of the message will be not always consistent and depend on numerous
configuration parameters such as enable_debug, error reporting level and so
on. I think it is simply better to let this test go...
Ilia
> Moriyoshi
>
> "Ili
I think this test should be removed. The error that is being generated is
E_ERROR meaning that the script gets terminated and we cannot capture in
$php_errormsg. The result is that the error output is flexible, for example
on my system the error message does not match the expected output and fai
We need to set it to NULL because it may be already set by another PHP
operation prior to htmlentities().
Ilia
On November 8, 2002 12:03 pm, Marcus Boerger wrote:
> helly Fri Nov 8 12:03:06 2002 EDT
>
> Modified files:
> /php4/ext/standard/tests/strings htmlentities06.phpt
> Lo
esolved_path))
>
> Thanks,
> Andi
>
> At 06:49 PM 11/4/2002 -0500, Ilia A. wrote:
> >I've taken you suggestion about adding an additional parameter to
> >file_ex() to
> >address the symlink situation. Attached is a patch that I believe should
> >resolv
I've taken you suggestion about adding an additional parameter to file_ex() to
address the symlink situation. Attached is a patch that I believe should
resolve the problem.
Ilia
On November 4, 2002 02:53 pm, Ilia A. wrote:
> On November 4, 2002 03:43 am, Andi Gutmans wrote:
> >
Okie, I'll revert the virtual_link completely then right now.
Ilia
On November 4, 2002 04:36 am, Andi Gutmans wrote:
> At 02:53 PM 11/4/2002 -0500, Ilia A. wrote:
> >On November 4, 2002 03:43 am, Andi Gutmans wrote:
> > > At 02:41 PM 11/4/2002 -0500, Ilia A. wrote:
> &
On November 4, 2002 03:43 am, Andi Gutmans wrote:
> At 02:41 PM 11/4/2002 -0500, Ilia A. wrote:
> >On November 4, 2002 03:36 am, you wrote:
> > > I'll check it out tomorrow. Your code is problematic in any case as
> > > you're returning a pointer to the stack
nction. Can you revert it in the meantime?
>
Quite correct, not sure why I missed that and the compiler did not catch that.
I'll revert the patch momentarily.
Ilia
> Andi
>
> At 02:31 PM 11/4/2002 -0500, Ilia A. wrote:
> >Certainly. The problem with the old code was quite si
Certainly. The problem with the old code was quite simple. Suppose we have a
file 'original_file' and we make a symlink to it and call it 'my_symlink'. If
inside my PHP code I were to do: unlink('my_symlink');, instead of deleting
the symlink, the unlink function would remove the original file,
rly in
both cases.
Ilia
>
> At 19:19 04.11.2002, Ilia A. wrote:
> >Actually I think bug20175.phpt is a bad test at least the last part. The
> >functionality you are trying to test simply does not exist and has ever
> >existed. I've tested ZE1 and ZE2 and both return wow:2
> marcus
>
> At 18:25 04.11.2002, Ilia A. wrote:
> >The 2nd wow should be '-wow:2' not 'wow:1'. For me it works in both ZE1
> > and ZE2 for value of 2. The value of 2 is correct because the test
> > increments the $wow_global variable twice. Please revert
The 2nd wow should be '-wow:2' not 'wow:1'. For me it works in both ZE1 and
ZE2 for value of 2. The value of 2 is correct because the test increments the
$wow_global variable twice. Please revert your change.
Ilia
On November 4, 2002 12:12 pm, Marcus Börger wrote:
> helly Mon Nov 4 12
have in the same way.
> Andi
>
> At 09:11 AM 11/1/2002 -0500, Ilia A. wrote:
> >Windows can interpret / as well as \, makes no difference to it. In fact
> > when we resolve paths on Windows, we already use / and not \.
> >
> >Ilia
> >
> >On November 1, 20
Windows can interpret / as well as \, makes no difference to it. In fact when
we resolve paths on Windows, we already use / and not \.
Ilia
On November 1, 2002 03:01 am, Andi Gutmans wrote:
> I'm not sure this is the right thing to do. UNIX and Windows are diffferent
> and that's what DEFAULT_SL
& log_errors)
> >
> >Another one here then: +1 for reverting the changes.
> >
> >Derick
> >
> > > On Sun, 27 Oct 2002, Ilia A. wrote:
> > > >I am curios as to your reasoning behind turning on html_errors by
> >
> > default, why
>
Both of those flags are already being set by the run-tests.php script, why
duplicate them?
Ilia
On October 27, 2002 06:56 pm, Marcus Börger wrote:
> helly Sun Oct 27 18:56:08 2002 EDT
>
> Modified files:
> /php4/ext/session/tests 008-php4.2.3.phpt
> Log:
> This test requires
I've just tried the latest version of the xslt tests and
ext/xslt/tests/xslt_setopt.phpt test fails. The failure is the result of
expected output needing some non-ascii character after PHP QA. Where is this
character comming from, the xml & xslt style sheets do not seem to have any
non-ascii da
On October 25, 2002 03:27 pm, Marcus Börger wrote:
> It should not break any test...please send a list of those failed.
This particular option broke the tests where an error message was being
generate by PHP as part of the test and then it's output compared as part of
the PASS/FAIL check. It is
Could you please try the attached patch and let me know if it fixes the
problem.
Ilia
On October 24, 2002 06:27 pm, Melvyn Sopacua wrote:
> At 22:10 24-10-2002, Ilia A. wrote:
> >The lower/upper code uses libc's tolower/toupper functions, as long as the
> >specified locale
On October 23, 2002 08:17 pm, Marcus Börger wrote:
> Nice test and additional skip logic but i think we should use
> characters available in germany like our umlauts.
> I suspect the test fails for me because my locale does not
> recognize the three characters used which are non german
> characters
I will have that test working correctly momentarily. As it stands that test
would only work if a locale supporting the used characters is exported.
Ilia
On October 23, 2002 05:04 pm, Moriyoshi Koizumi wrote:
> Hi,
>
> The first letter in $chars appears to be a small "a" with a grave accent
> in
On October 22, 2002 11:04 pm, Yasuo Ohgaki wrote:
> Ilia A. wrote:
> > On October 22, 2002 10:30 pm, Yasuo Ohgaki wrote:
> >>Ilia A. wrote:
> >>>>What's the point of _not_ specifying certain ini for
> >>>>run-tests.php? In other words, what is
On October 22, 2002 10:30 pm, Yasuo Ohgaki wrote:
> Ilia A. wrote:
> >>What's the point of _not_ specifying certain ini for
> >>run-tests.php? In other words, what is advantages?
> >
> > Because those ini settings may be the ones causing the problems.
On October 22, 2002 01:57 am, Yasuo Ohgaki wrote:
> Jon Parise wrote:
> > On Tue, Oct 22, 2002 at 11:14:58AM +0900, Yasuo Ohgaki wrote:
> >>No objections using php.ini-dist _only for_ run-tests.php?
> >
> > I'm not really in the run-tests.php loop, but why must be
> > configuration file be specifie
On October 18, 2002 04:40 pm, Rasmus Lerdorf wrote:
> > Fixed bug #19971 (optimized the file() function).
> > The file() function is now also binary safe.
>
> That's a bit misleading. By definition file() cannot be binary-safe in
> that \n is treated specially.
Not quite, in the old code (4.2
On October 18, 2002 05:04 pm, Rasmus Lerdorf wrote:
> > On October 18, 2002 04:40 pm, Rasmus Lerdorf wrote:
> > > > Fixed bug #19971 (optimized the file() function).
> > > > The file() function is now also binary safe.
> > >
> > > That's a bit misleading. By definition file() cannot be binary-
On October 18, 2002 05:50 pm, Rasmus Lerdorf wrote:
> > > My point is that the original data is munged once it goes through the
> > > file() function. In that sense it is not binary safe as there is no
> > > way to restore the original data.
> >
> > Unless the user actually goes and modifies the d
I'll correct the name of the function I've added, I applogize for the
oversight. I do not think we should rename existing functions because it
would break BC and force us add to add even more function aliases. We already
have 400+ function alises in PHP, lets try avoiding adding new ones unless
s is what I could think of now and it's
> a valid PHP statement. This patch would add ambiguity to the language.
> We could have an INI directive such as xml_hack=on but I hate adding INI
> stuff too.
>
> I suggest people who are generating XML just turn off short-tags.
>
On October 10, 2002 08:55 pm, Sean R. Bright wrote:
> localeconv is not threadsafe AFAIK, which is why the localeconv_r
> function was introduced.
I'll make the correction.
Also, does HAVE_SETLOCALE necessarily imply
> HAVE_LOCALECONV?
I believe it is, on all the systems I have access to both
60 matches
Mail list logo
