[pmwiki-devel] lower casing diacritics

[Simon]

Can anyone suggest a means of converting diacritic characters to lower case,

e.g. Ā to ā, Ê to ê, Į to į, etc

other than creating a translation table?

thanks

Simon

[Petko Yotov]

You can use mb_strtolower():

https://php.net/mb_strtolower

Here is an example from the PHP interactive shell:

php > $str = "e.g. Ā to ā, Ê to ê, Į to į, etc";
php > print_r(mb_strtolower($str));
e.g. ā to ā, ê to ê, į to į, etc
php > print_r(mb_strtoupper($str));
E.G. Ā TO Ā, Ê TO Ê, Į TO Į, ETC

Petko

--
If you upgrade : https://www.pmwiki.org/Upgrades

On 18/10/2022 06:46, Simon wrote:
> Can anyone suggest a means of converting diacritic [1]characters to

> lower case,
> e.g. Ā to ā, Ê to ê, Į to į, etc
> other than creating a translation table?
>
> thanks
>
> Simon
>
>
>

> Links:
> ------
> [1] https://en.wikipedia.org/wiki/Diacritic
> _______________________________________________
> pmwiki-devel mailing list
> pmwiki...@pmichaud.com
> http://www.pmichaud.com/mailman/listinfo/pmwiki-devel

_______________________________________________
pmwiki-devel mailing list
pmwiki...@pmichaud.com
http://www.pmichaud.com/mailman/listinfo/pmwiki-devel

[Simon]

Again, thanks heaps for answering these newbie questions, that works. 

What I think I have found is that while html_entity_decode('Ē') gives "Ē"

htmlentities ("Ē") doesn't convert Ē back to  Ē

Simon

[Petko Yotov]

You may be able to use:

$entity = mb_convert_encoding($decoded, 'HTML');

You may or may not need to specify a $from_encoding argument. From the
documentation it seems before PHP 8.0 $from_encoding was required.
Documentation:

https://php.net/mb_convert_encoding

Petko

--
If you upgrade : https://www.pmwiki.org/Upgrades

[Simon]

Some background.

I am trying to update the SearchCloud recipe.

The recipe grabs the q parameter of a search action.

I want it to 

* make the search terms insensitive

* handle characters with diacritics.

Here is some debug output

2022-10-19 21:26:01 

q="SĀÉÎÖŬ-àęiøűd" 

$SCrq="SĀÉÎÖŬ-àęiøűd" 

tkey1="SÄ€ÉÎÖŬ-àÄ™iøűd" 

tkey2="sÄ ???Å­-?Ä™i?űd" 

tkey3="sÄ ???Å­-?Ä™i?űd"

Generated from debug code

      $convmap = array (0x80, 0xffff);
      $q     = strval($_REQUEST['q']); # get search term
      $SCrq  = trim (\stripmagic($q));
      $tkey1 = html_entity_decode($SCrq); # remove html entities to allow lower case conversion
      $tkey2 = mb_strtolower($tkey1); # convert to lower case
      $tkey3 = mb_encode_numericentity ($tkey2, $convmap); # convert non-ascii to htmlentities
      $fwritestatus = fwrite($logfilehandle, $logfiletime
      . 'q="' . $q
      . '" $SCrq="' . $SCrq
      . '" tkey1="' . $tkey1 
      . '" tkey2="' . $tkey2 
      . '" tkey3="' . $tkey3 . '"'

As you can see in the debug output it seems to fall apart at tkey2.

I'd welcome more suggestions

[Dominique Faure]

You should perhaps specify the 'UTF-8' encoding to the mb_strtolower call.