grep puzzle

[KhanyBoy]

Hi,

this should test you guru's. I want a function that accepts text as an
argument and converts all & into & except where it is a html
character already such as  , ", and of course &.

If there is already a php function for this I would like to know, but
if not, what is the GREP equivilent?

Thanks

[Alvaro G Vicario]

Re: grep puzzle

*** KhanyBoy wrote/escribió (10 Jun 2004 14:22:24 -0700):[color=blue]
> this should test you guru's. I want a function that accepts text as an
> argument and converts all & into & except where it is a html
> character already such as  , ", and of course &.[/color]

I can't figure out how you manage to get such a garbled input. Anyway, I
guess a combination of html_entity_dec ode() and html_entities() should
help.


--
--
-- Álvaro G. Vicario - Burgos, Spain
--

[Justin Koivisto]

Re: grep puzzle

KhanyBoy wrote:[color=blue]
> this should test you guru's. I want a function that accepts text as an
> argument and converts all & into & except where it is a html
> character already such as  , ", and of course &.
>
> If there is already a php function for this I would like to know, but
> if not, what is the GREP equivilent?[/color]

OK, this does that, but it may not be a very elegant soltion. I recently
needed the same functionality for a project involving oscommerce.

function ampersandFix($x ){
$x=str_replace( '&','&',$x) ;
$pattern='`&amp ;(#[0-9]{2,3}|aacute|ac irc|acute|aelig |agrave|amp'.
'|aring|atilde| auml|brvbar|brk bar|ccedil|cedi l|cent'.
'|copy|curren|d eg|divide|eacut e|ecirc|egrave| eth|euml'.
'|frac12|frac14 |frac34|gt|iacu te|icirc|iexcl| igrave'.
'|iquest|iuml|l aquo|lt|macr|hi bar|micro|middo t|nbsp|not'.
'|ntilde|oacute |ocirc|ograve|o rdf|ordm|oslash |otilde'.
'|ouml|para|plu smn|pound|quot| raquo|reg|sect| shy|sup1|sup2'.
'|sup3|szlig|th orn|times|uacut e|ucirc|ugrave| uml'.
'|die|uuml|yacu te|yen|yuml);`i ';
$replace='&$1;' ;
return preg_reacple($p attern,$replace ,$x);
}


--
Justin Koivisto - spam@koivi.com
PHP POSTERS: Please use comp.lang.php for PHP related questions,
alt.php* groups are not recommended.

[Christian Fersch]

Re: grep puzzle

Justin Koivisto wrote:[color=blue]
> KhanyBoy wrote:[color=green]
>> If there is already a php function for this I would like to know, but
>> if not, what is the GREP equivilent?[/color]
>
> OK, this does that, but it may not be a very elegant soltion. I recently
> needed the same functionality for a project involving oscommerce.
>
> function ampersandFix($x ){
> $x=str_replace( '&','&',$x) ;
> $pattern='`&amp ;(#[0-9]{2,3}|aacute|ac irc|acute|aelig |agrave|amp'.
> '|aring|atilde| auml|brvbar|brk bar|ccedil|cedi l|cent'.
> '|copy|curren|d eg|divide|eacut e|ecirc|egrave| eth|euml'.
> '|frac12|frac14 |frac34|gt|iacu te|icirc|iexcl| igrave'.
> '|iquest|iuml|l aquo|lt|macr|hi bar|micro|middo t|nbsp|not'.
> '|ntilde|oacute |ocirc|ograve|o rdf|ordm|oslash |otilde'.
> '|ouml|para|plu smn|pound|quot| raquo|reg|sect| shy|sup1|sup2'.
> '|sup3|szlig|th orn|times|uacut e|ucirc|ugrave| uml'.
> '|die|uuml|yacu te|yen|yuml);`i ';
> $replace='&$1;' ;
> return preg_reacple($p attern,$replace ,$x);
> }[/color]

A lot faster, but not as accurate:
$text = preg_replace('! &(?![#a-z0-9]{1,7};)!i','&am p',$text);

You could also use this method (it's a lookahead assertion) with
Justin's function, which will still be a lot faster than his hack ;)

Greetings Christian.

[John Dunlop]

Re: grep puzzle

KhanyBoy wrote:
[color=blue]
> this should test you guru's. I want a function that accepts text as an
> argument and converts all & into & except where it is a html
> character already such as  , ", and of course &.[/color]

Why are entity references recognised in text?

--
Jock

[FLEB]

Re: grep puzzle

Regarding this well-known quote, often attributed to KhanyBoy's famous "10
Jun 2004 14:22:24 -0700" speech:
[color=blue]
> Hi,
>
> this should test you guru's. I want a function that accepts text as an
> argument and converts all & into & except where it is a html
> character already such as  , ", and of course &.
>
> If there is already a php function for this I would like to know, but
> if not, what is the GREP equivilent?
>
> Thanks[/color]

It might be a different direction, but here are some functions to determine
whether something is an HTML entity, using the built-in PHP entity
functions. It's sort of a recycled reply to an earlier question:

A simple (rough) test of the concept is online at:

http://php.pixelsaredead.com/htmlentities.php

<?php
/*
How to determine whether a given string decodes to an HTML entity
*/

function contains_entiti es($raw)
{
// $raw can be a string of any length
$raw = trim($raw);
return (strlen(htmlent ities($raw)) > strlen($raw));
}

function is_entity_refer ence($raw)
{
// $raw should be a string with only the entity ref in it,
// in the form "&...;"

return (preg_match('/&.+;/', $raw)) &&
(strlen(html_en tity_decode(tri m($raw))) == 1);
}
?>


--
-- Rudy Fleminger
-- sp@mmers.and.ev il.ones.will.bo w-down-to.us
(put "Hey!" in the Subject line for priority processing!)
-- http://www.pixelsaredead.com

[Justin Koivisto]

Re: grep puzzle

Christian Fersch wrote:
[color=blue]
> Justin Koivisto wrote:
>[color=green]
>> KhanyBoy wrote:
>>[color=darkred]
>>> If there is already a php function for this I would like to know, but
>>> if not, what is the GREP equivilent?[/color]
>>
>>
>> OK, this does that, but it may not be a very elegant soltion. I
>> recently needed the same functionality for a project involving
>> oscommerce.
>>
>> function ampersandFix($x ){
>> $x=str_replace( '&','&amp;',$x) ;
>> $pattern='`&amp ;(#[0-9]{2,3}|aacute|ac irc|acute|aelig |agrave|amp'.
>> '|aring|atilde| auml|brvbar|brk bar|ccedil|cedi l|cent'.
>> '|copy|curren|d eg|divide|eacut e|ecirc|egrave| eth|euml'.
>> '|frac12|frac14 |frac34|gt|iacu te|icirc|iexcl| igrave'.
>> '|iquest|iuml|l aquo|lt|macr|hi bar|micro|middo t|nbsp|not'.
>> '|ntilde|oacute |ocirc|ograve|o rdf|ordm|oslash |otilde'.
>> '|ouml|para|plu smn|pound|quot| raquo|reg|sect| shy|sup1|sup2'.
>> '|sup3|szlig|th orn|times|uacut e|ucirc|ugrave| uml'.
>> '|die|uuml|yacu te|yen|yuml);`i ';
>> $replace='&$1;' ;
>> return preg_reacple($p attern,$replace ,$x);
>> }[/color]
>
>
> A lot faster, but not as accurate:
> $text = preg_replace('! &(?![#a-z0-9]{1,7};)!i','&am p',$text);
>
> You could also use this method (it's a lookahead assertion) with
> Justin's function, which will still be a lot faster than his hack ;)[/color]

Let's fix the hack then, shall we?

function ampersandFix($x ){
$pattern='`&(?! (#[0-9]{2,3}|aacute|ac irc|acute|aelig |agrave|amp'.
'|aring|atilde| auml|brvbar|brk bar|ccedil|cedi l|cent'.
'|copy|curren|d eg|divide|eacut e|ecirc|egrave| eth|euml'.
'|frac12|frac14 |frac34|gt|iacu te|icirc|iexcl| igrave'.
'|iquest|iuml|l aquo|lt|macr|hi bar|micro|middo t|nbsp|not'.
'|ntilde|oacute |ocirc|ograve|o rdf|ordm|oslash |otilde'.
'|ouml|para|plu smn|pound|quot| raquo|reg|sect| shy|sup1|sup2'.
'|sup3|szlig|th orn|times|uacut e|ucirc|ugrave| uml'.
'|die|uuml|yacu te|yen|yuml);)` i';
return preg_replace($p attern,'&amp;', $x);
}

Now it's faster AND accurate. ;)

--
Justin Koivisto - spam@koivi.com
PHP POSTERS: Please use comp.lang.php for PHP related questions,
alt.php* groups are not recommended.