Special chars not showing up right in cgi search

[dbuck]

Hi,
I have the enterprise edition and I am indexing UTF-8 html content with the Unicode char set selected in the Language config tab.

This source content sometimes has a lot of ® and © symbols in it. Serving the html will show the chars just fine. But when searching using the CGI search, and a result has one of these chars., it will show up as a "�" or a "?" in internet explorer, firefox, and opera; Windows and Linux.

I have used both indexer 5.0 build 1000 and 1001. Is there some way to have these show up fine?

Thanks,
D

[David]

Do the copyright symbols appear in the HTML source as character entities or as ordinary characters? A HTML character entity looks like this, "©".

Are you sure that the search template file is also UTF-8?

What font set are you using on the result page? Arial or something more obscure?

Can you post the URL to the search function and / or the source document, so that we can see the problem. It makes it much easier to investigate.

[dbuck]

Hi again,
Those were some really good suggestions. We tried them all, but to no avail:

Used character entities and ordinary chars all in UTF-8.
Our font was Arial.
The template was UTF-8; we also tried the default template.
We used a different web server. (first was shttpd, second apache)
The encoding under the Launguage tab in the indexer was still set to UTF-8

We set up a test here:
http://www.gogroundline.com/cgi-bin/Linux_search.cgi

Most the links will not work because we only indexed one page. If you do a search for 'Groupwise', the problem will show up.

I am attaching a piece of one of the documents that show this error. It is found as 'test.html' in the accompanying zip file. The index files are also attached.

Thanks,
D

[Ray]

We've confirmed the problem in the latest build (5.0.1001) where copyright, trademark and various symbols may not appear correctly in the context description on an UTF-8 page. This will be fixed in the upcoming new build (5.0.1002).

[dbuck]

Thanks!

D