https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=16839 David Liddle <david_liddle@sil.org> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |david_liddle@sil.org --- Comment #2 from David Liddle <david_liddle@sil.org> --- Our Koha installation has holdings that use many different scripts, including Thai and Chinese. Issues with the latter caused me to revisit this topic today. In an interesting chat with Google Gemini, it took existing information from sources in Zebra documentation and the Koha community sites and created a seemingly novel solution for sorting. It suggested that I create a similar file to those mentioned in the instructions on the Koha Wiki, words-icu.xml and phrases-icu.xml. This file, named sort-icu.xml, replaced sort-string-utf.chr in default.idx. Additionally, all three XML files were reconfigured to use the Thai locale, which apparently resolves some distinct challenges with sorting while remaining compatible with other languages. I put this in place on our test server, did a full reindex (without errors), and asked our staff to compare search result sorting between it and our production server. We'll see if Gemini was right – or full of hooey. -- You are receiving this mail because: You are watching all bug changes.

https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=16839 --- Comment #4 from David Liddle <david_liddle@sil.org> --- Initial feedback from our librarian who manages some of the libraries in southeast Asia indicates that the changes seem to work well for Thai, Chinese, and Burmese. I suppose the next step would be for me to post our configuration in the ICU Chains Library? -- You are receiving this mail because: You are watching all bug changes.