Everything

php

[RFC]strn(case)cmp supporting negative length as third parameter

php.internals

Laruence

15 years ago

Hi:

strn(case)cmp dosen't  support a negative length as its third
paramter,  while substr dose.

I wrote a patch to make strn(case)cmp supporting negative length,
make following script works as expect:

<?php
if (strncmp("prefix_num", "num", -3) === 0) {
echo "they have same suffix\n";
}
?>

here is the rfc: https://wiki.php.net/rfc/strncmpnegativelen

any question? plz worte me back.

thanks

--
Laruence  Xinchen Hui
http://www.laruence.com/

Derick Rethans

15 years ago

On Thu, 28 Jul 2011, Laruence wrote:

> Hi:
>
>   strn(case)cmp dosen't  support a negative length as its third
> paramter,  while substr dose.
>
>  I wrote a patch to make strn(case)cmp supporting negative length,
> make following script works as expect:
>
>     <?php
>    if (strncmp("prefix_num", "num", -3) === 0) {
>         echo "they have same suffix\n";
>    }
>   ?>
>
>  here is the rfc: https://wiki.php.net/rfc/strncmpnegativelen

- You're changing binary API of some API functions, so this can't go
into PHP 5.3:

-ZEND_API int zend_binary_strncmp(const char *s1, uint len1, const char *s2, uint len2, uint length);
+ZEND_API int zend_binary_strncmp(const char *s1, uint len1, const char *s2, uint len2, int length);

cheers,
Derick

--
http://derickrethans.nl | http://xdebug.org
Like Xdebug? Consider a donation: http://xdebug.org/donate.php
twitter: @derickr and @xdebug

Laruence

15 years ago

Hi:

why?

I think this change would not cause any other side-effects.

and if you are worry about INT_MAX,  I think it's ok to declare with
long(in fact, I already did so).

thanks

2011/7/28 Derick Rethans <derick@php.net>:

> On Thu, 28 Jul 2011, Laruence wrote:
>
>> Hi:
>>
>>   strn(case)cmp dosen't  support a negative length as its third
>> paramter,  while substr dose.
>>
>>  I wrote a patch to make strn(case)cmp supporting negative length,
>> make following script works as expect:
>>
>>     <?php
>>    if (strncmp("prefix_num", "num", -3) === 0) {
>>         echo "they have same suffix\n";
>>    }
>>   ?>
>>
>>  here is the rfc: https://wiki.php.net/rfc/strncmpnegativelen
>
> - You're changing binary API of some API functions, so this can't go
>  into PHP 5.3:
>
>        -ZEND_API int zend_binary_strncmp(const char *s1, uint len1, const char *s2, uint len2, uint length);
>        +ZEND_API int zend_binary_strncmp(const char *s1, uint len1, const char *s2, uint len2, int length);
>
> cheers,
> Derick
>
> --
> http://derickrethans.nl | http://xdebug.org
> Like Xdebug? Consider a donation: http://xdebug.org/donate.php
> twitter: @derickr and @xdebug
>

--
Laruence  Xinchen Hui
http://www.laruence.com/

Pierre Joye

15 years ago

the signature is not the same anymore, some extension may not build
anymore (depending on the compiler used and its flag).

Cheers,

On Thu, Jul 28, 2011 at 1:32 PM, Laruence <laruence@php.net> wrote:

> Hi:
>
> why?
>
> I think this change would not cause any other side-effects.
>
> and if you are worry about INT_MAX,  I think it's ok to declare with
> long(in fact, I already did so).
>
> thanks
>
> 2011/7/28 Derick Rethans <derick@php.net>:
>> On Thu, 28 Jul 2011, Laruence wrote:
>>
>>> Hi:
>>>
>>>   strn(case)cmp dosen't  support a negative length as its third
>>> paramter,  while substr dose.
>>>
>>>  I wrote a patch to make strn(case)cmp supporting negative length,
>>> make following script works as expect:
>>>
>>>     <?php
>>>    if (strncmp("prefix_num", "num", -3) === 0) {
>>>         echo "they have same suffix\n";
>>>    }
>>>   ?>
>>>
>>>  here is the rfc: https://wiki.php.net/rfc/strncmpnegativelen
>>
>> - You're changing binary API of some API functions, so this can't go
>>  into PHP 5.3:
>>
>>        -ZEND_API int zend_binary_strncmp(const char *s1, uint len1, const char *s2, uint len2, uint length);
>>        +ZEND_API int zend_binary_strncmp(const char *s1, uint len1, const char *s2, uint len2, int length);
>>
>> cheers,
>> Derick
>>
>> --
>> http://derickrethans.nl | http://xdebug.org
>> Like Xdebug? Consider a donation: http://xdebug.org/donate.php
>> twitter: @derickr and @xdebug
>>
>
>
>
> --
> Laruence  Xinchen Hui
> http://www.laruence.com/
>
> --
> PHP Internals - PHP Runtime Development Mailing List
> To unsubscribe, visit: http://www.php.net/unsub.php
>
>

--
Pierre

@pierrejoye | http://blog.thepimp.net | http://www.libgd.org

Laruence

15 years ago

Hi:
As Derick and Pierre  said(in IRC),  I change the patch base from
PHP 5.3 to PHP5.4,

and avoid changing ZEND_API declaration.

RFC updated

new patch : http://www.laruence.com/php-dev/php-5.4-trunk-strncmp-supproting-negative-len.patch

thanks

2011/7/28 Pierre Joye <pierre.php@gmail.com>:

> the signature is not the same anymore, some extension may not build
> anymore (depending on the compiler used and its flag).
>
> Cheers,
>
> On Thu, Jul 28, 2011 at 1:32 PM, Laruence <laruence@php.net> wrote:
>> Hi:
>>
>> why?
>>
>> I think this change would not cause any other side-effects.
>>
>> and if you are worry about INT_MAX,  I think it's ok to declare with
>> long(in fact, I already did so).
>>
>> thanks
>>
>> 2011/7/28 Derick Rethans <derick@php.net>:
>>> On Thu, 28 Jul 2011, Laruence wrote:
>>>
>>>> Hi:
>>>>
>>>>   strn(case)cmp dosen't  support a negative length as its third
>>>> paramter,  while substr dose.
>>>>
>>>>  I wrote a patch to make strn(case)cmp supporting negative length,
>>>> make following script works as expect:
>>>>
>>>>     <?php
>>>>    if (strncmp("prefix_num", "num", -3) === 0) {
>>>>         echo "they have same suffix\n";
>>>>    }
>>>>   ?>
>>>>
>>>>  here is the rfc: https://wiki.php.net/rfc/strncmpnegativelen
>>>
>>> - You're changing binary API of some API functions, so this can't go
>>>  into PHP 5.3:
>>>
>>>        -ZEND_API int zend_binary_strncmp(const char *s1, uint len1, const char *s2, uint len2, uint length);
>>>        +ZEND_API int zend_binary_strncmp(const char *s1, uint len1, const char *s2, uint len2, int length);
>>>
>>> cheers,
>>> Derick
>>>
>>> --
>>> http://derickrethans.nl | http://xdebug.org
>>> Like Xdebug? Consider a donation: http://xdebug.org/donate.php
>>> twitter: @derickr and @xdebug
>>>
>>
>>
>>
>> --
>> Laruence  Xinchen Hui
>> http://www.laruence.com/
>>
>> --
>> PHP Internals - PHP Runtime Development Mailing List
>> To unsubscribe, visit: http://www.php.net/unsub.php
>>
>>
>
>
>
> --
> Pierre
>
> @pierrejoye | http://blog.thepimp.net | http://www.libgd.org
>

--
Laruence  Xinchen Hui
http://www.laruence.com/

Keisial

15 years ago

Laruence wrote:

> Hi:
>
>    strn(case)cmp dosen't  support a negative length as its third
> paramter,  while substr dose.

>   here is the rfc: https://wiki.php.net/rfc/strncmpnegativelen
>
>   any question? plz worte me back.
>
>   thanks

What do you exactly mean by "in the reverse order" in

> if the abs of the negative length is greater than any strlen of the
> first two parameters, then strn(case)cmp will work as strncmp in the
> reverse order with the abs value:

Looking at

> var_dump  <http://www.php.net/var_dump>(strncmp  <http://www.php.net/strncmp>("prefix_num",  "num",  -10));
>    //outpu: int(7)

it seems to mean that it would be equivalent to

> var_dump(strncmp("mun_xiferp", "mun", 10));

However, if I'm reading the patch correctly, it would
be taking the smaller string. Ie. it would be equivalent
to

> var_dump  <http://www.php.net/var_dump>(strncmp  <http://www.php.net/strncmp>("prefix_num",  "num",  -3));

Seems an implementation bug.

Laruence

15 years ago

Hi:
first of all, yes , you are right about the reverse order thing:

and then, they are not equal:

<?php
echo "strncmp:\n";
var_dump(strncmp("laruence", "nce", -10));
var_dump(strncmp("ecneural", "ecn", 10));

var_dump(strncmp("ecneural", "ecn", 3));
var_dump(strncmp("laruence", "nce", -3));
?>

will output:
strncmp:
int(5)
int(5)
int(0)
int(0)


thanks

2011/7/29 Ángel González <keisial@gmail.com>:

> Laruence wrote:
>>
>> Hi:
>>
>>   strn(case)cmp dosen't  support a negative length as its third
>> paramter,  while substr dose.
>
>>  here is the rfc: https://wiki.php.net/rfc/strncmpnegativelen
>>
>>  any question? plz worte me back.
>>
>>  thanks
>
> What do you exactly mean by "in the reverse order" in
>>
>> if the abs of the negative length is greater than any strlen of the first
>> two parameters, then strn(case)cmp will work as strncmp in the reverse order
>> with the abs value:
>
> Looking at
>>
>> var_dump  <http://www.php.net/var_dump>(strncmp
>>  <http://www.php.net/strncmp>("prefix_num",  "num",  -10));
>>   //outpu: int(7)
>
> it seems to mean that it would be equivalent to
>>
>> var_dump(strncmp("mun_xiferp", "mun", 10));
>
> However, if I'm reading the patch correctly, it would
> be taking the smaller string. Ie. it would be equivalent
> to
>
>> var_dump  <http://www.php.net/var_dump>(strncmp
>>  <http://www.php.net/strncmp>("prefix_num",  "num",  -3));
>
> Seems an implementation bug.
>
>

--
Laruence  Xinchen Hui
http://www.laruence.com/