tags URL wrong forming

[guest*]

All my tags with special characters (german) are forming a wrong URL.

F.ex.

keyword: Ausströmer = /search.php?tag=Ausstr-and here strange characters instead of-ö mer
keyword: Kürbis = /search.php?tag=K-and here strange characters instead of-ü rbis

and so on. Also the canonical are forming with the strange characters like % C 3 % B 6 for ö.

I take a look into /classes/tag.php and also in modules/blocktag/blocktags.php on both files there is nothing specified with UTF-8, shall this the problem ? If yes, how to fix ?

[shokinro]

Try to use urlencode($str) function to encode the keyword
I think you may need to change inside the modules/bolcktags/bolcktags.php file

[guest*]

Sorry I'm not a programmer. What shall I ammend in the file ?

[shokinro]

I just checked the code in blocktags.tpl file.
The urlencode has been already used.
I am not sure what is the cause of your problem.

Could you please share me your site link here or by PM so that I may be able to take a look.

[guest*]

Yes, sorry. I checked it again and it ist used:

    {$tag.name|escape:html:'UTF-8'}     

Project is in coming, but has no public access now (is a clone from an existing domain with another e-commerce software, but with different IP's). To have access you must change your hosts file in windows to do not ping the online project. What do you need to evaluate ?

[inveostore.com]

There is no way how to avoid such strange characters for non English chars (only if you replace such chars with ASCII equivalents - such as ü -> u etc. and then recover them back.

[guest*]

As this module comes as standard I will pass this as a bug. I've checked in DB, the special characters are writing correct, so it must be a change in code. I'm not a coder only presta user.

[inveostore.com]

The current behavior is not a bug. It works exactly as specified by related RFC.

[shokinro]

I re-read again the original post. It seems that I misunderstood the question.
Just like "inveostore.com" pointed out, it is correct to covert special char to something %C. It is the work of urlencode().

If there is no problem with URL link (for example, broken link), then everything is normal, sorry for the confusion.

[guest*]

The current behavior is not a bug. It works exactly as specified by related RFC.

So how to change for other standard presta lanaugages which uses special characters (Finnland, Denmark, Germany, and so on ?)

[Artizzz]

I had the same problem with Latvian charters - ā,ē,č.... . I solve it by deleting lines 64-66 in themes/(theme)/pagination.tlp

                        {if $requestKey != 'requestUrl'}
                           <input type="hidden" name="{$requestKey|escape:'htmlall':'UTF-8'}" value="{$requestValue|escape:'htmlall':'UTF-8'}" />
                       {/if}

what does this line do anyway? maybe someone can fix it without deleting it?

P.S I had this problem when I choose tags and then tried to change products per page.

[guest*]

For me it not solved. Tags remain as this one:

/search.php?tag=St % C 3 % B 6 rfutter

I also deleted the files in folder /compile and browser cache. No change.

Perhaps somebody can fix this module ?

[Artizzz]

Hi, cd2500!
I had problem with tags with specific letters(ā,ē,ī etc) and I had problem with the choosing producs per pagein those tag lists. I recomend you to edit pagination.tlp (in version 1.3.7) lines 48-73

{if $nb_products > 10}
       <form action="{if !is_array($requestNb)}{$requestNb}{else}{$requestNb.requestUrl}{/if}" method="get" class="pagination">



               {if isset($query) AND $query}<input type="hidden" name="search_query" value="{$query|escape:'htmlall':'UTF-8'}" />{/if}
               {if isset($tag) AND $tag AND !is_array($tag)}<input type="hidden" name="tag" value="{$tag|escape:'htmlall':'UTF-8'}" />
               {else is_array($requestNb)}
                   {foreach from=$requestNb item=requestValue key=requestKey}
                       {if $requestKey != 'requestUrl'}
                           <input type="hidden" name="{$requestKey|escape:'htmlall':'UTF-8'}" value="{$requestValue|escape:'htmlall':'UTF-8'}" />
                       {/if}
                   {/foreach}
               {/if}
               <input type="submit" class="button_mini" value="{l s='OK'}" />
{l s='items:'}

               {foreach from=$nArray item=nValue}
                   {if $nValue <= $nb_products}
{$nValue|escape:'htmlall':'UTF-8'}
                   {/if}
               {/foreach}



       </form>
   {/if}

[guest*]

Hi Artizzz,
thxs for code. But no change... I'm still getting an wrong UTF-8 encoding like this:

D % C 3 % B C n g e w % C 3 % B C rfel instead the word Düngewürfel as URL

[Artizzz]

Hi Artizzz,
thxs for code. But no change... I'm still getting an wrong UTF-8 encoding like this:

D % C 3 % B C n g e w % C 3 % B C rfel instead the word Düngewürfel as URL

Hi, cd2500!

Did you find some solution? Now on 1.4.0.17 version I have this url problem in tag list when I choose different number of products per page it shows word - Grāmata like "Gr % 2 5 C 4 % 2 5 8 1 matas"(what I can't with method what posted before).

[guest*]

No- no solution. According to a SEO proficient the SERP's also can read these UTF-encoding and translating into right characters. If you go with mousover on the link browsers translate them. The same do google & co.

I think we can let it, if on header the encoding is UTF-8 (presta standard)

[Artizzz]

No- no solution. According to a SEO proficient the SERP's also can read these UTF-encoding and translating into right characters. If you go with mousover on the link browsers translate them. The same do google & co.

I think we can let it, if on header the encoding is UTF-8 (presta standard)

yea, but my problem isn't only url looking. Look at my test site list of tag(with specific charter) - http://artis.noooby.lv/en/search?tag=ārprātš and when you are in try to change products per page.
This is version 1.4.0.17 and there is something different because in version 1.3.7 I fix it with that code what I posted before.
Any ideas how to fix to make it working?
And strangest thing is that the page selecting from pagination works with words with specific charters.

[guest*]

Yes it works because the strange characters are a readable UTF-encoding. No I don't know how to fix. As developer of module garanted to me, that google & co. also can read/translate the characters into readable one, I let it so...
For me the strange characters are only in source.code, but not when you go with your mouse over the link. It's only on source...