Pagination seems to hang

Sorry but the title of this is misleading. Upon further investigation it appears to be an issue with a PDF that had Chinese characters. The Chinese shows up as random characters in the AJAX response when calling return_solr_results. Is there a work around for this?

get_search_results is not returning $content in UTF-8?

Further investigation leads to class-wp-solr.php > get_search_results (…) function. Around line 488 $cont = ucfirst(trim($count)); seems to be causing the problem. After that function call the UTF-8 Chinese characters turn to gibberish. I think ucfirst does not work properly with non English characters?

I tested with all sorts of unicode characters (chinese included), and found no problems.

ucfirst() does not work as expected (no uppercase of first letter), but no error neither.

The initial results after clicking search seem to display properly. I get the error when I paginate to the page that has the UTF-8 characters. The ajax call returns nothing.
Maybe something odd with my installation but the same thing is happening on 3 of our dev servers.
Thanks for looking into this.

Please send me the pdf. I’ll try to index it and search it.

Can test with any AJAX call not just pagination i.e. change sort order.

FAQ-Mandarin.pdf

It’s working fine for me.

I could index it, search it, by latin characters (“1”), or chinese characters.

I can see it’s excerpt in results, I can navigate back and forward in results pages, I can filter by facets, I can order by.

Okay thanks for the update. It may be something odd in my environment.