Home› General support

Problem on seraching with germen mutated vowels

MarkusDippold Member
in General support
Hi,

first the link to my gallery:
http://www.dippold.org/zenphoto/

Ok, the simplest way to describe the problem is, to try it yourself.

Enter "Südwind" in the searchbox, nothing will be found.
Enter "`Südwind`" in the searchbox, you will get the right output.
But this is not very usable.

If mutated vowels like ä, ö, ü, ß are in the description text of the pictures, this words (or pictures) will not be found.
If words with ä, ö, ... are used as buzzwords, pictures with this buzzword will be found.

Is there any solution for this problem?
I havn't found the right location, where I have to change the code, maybe to add a htmlentity-call. The question is: Does this work correct, so the buzzwords still will be found?

Greets,
Markus

Comments

  • sbillard Member
    Thank you for the report. Using your information I have determined that the TinyMCE editor is creating these HTML Special Characters when it edits your descriptions. The data is being saved in the database with these encodings.

    I will ask my collegue acrylian to look into the configuration of TinyMCE. (He is the expert.) Perhaps it can be convinced not to do this.

    In the mean time, you can disable TinyMCE editing by deleting the `editor_config.js.php` file from the zp-core/js folder. You will have to re-save any descriptions (or other fields) that have accented characters and were edited with TinyMCE as the database has the encoded characters.
  • MarkusDippold Member
    Hi sbillard,

    thanks for the answer.
    Now, it's funny. In moment, I have with "Südwind" exakt the contrary behaviour.
    "Südwind" has been found (all picutres), but not "`Südwind`". That will be ok.

    But: Searching for "Märklin" has other results as a search for "`Märklin`". In both results are partial the same pictures, partial different pictures.

    In moment, I can't look into the database (I'm at work).
    Yesterday I have changed some texts directly over the user-frontend and not with the admin-frontend. Is it possible, that the handling with the vowels is different in this both ways of editing the texts?

    For the first, I will try deleting the file.
    If this will work fine, for it's ok.

    Greets,
    Markus
  • acrylian Administrator, Developer
    I will take a look at that.
  • acrylian Administrator, Developer
    This is a TinyMCE configuration thing: Please open the `editor_config.js.php` file and add `entity_encoding: "raw"` to it.
  • MarkusDippold Member
    Uhhhh ... editing texts in the user-frontend does this with a list-element (with not modfied `editor_config.js.php`):
    `
    • Hersteller: Märklin
    • Artikelnummer: vermutlich 4083 (1)
    `
    The new entry in the file seems to work. I get the results with ä, ö, ü, ... that I expect.
    Now I have to re-save all texts.

    Thanks a lot.

    Greets,
    Markus
Sign In or Register to comment.