5 replies to this topic

[Guntis]

Just downloaded the latest Accordance 10.0.4 Demo build. Added Latvian user Bible. When I open Latvian Bible in another tab, I can enter Latvian words in the search field and get results, but when I copy/paste the same search text string into the default text tab, and if KJV is selected as a search target, although Bible text below is Latvian Bible, Latvian (Unicode) characters in the search field are replaced with something else, and there are no search results. Please fix this bug.

Attached Files

•  Screen Shot 2012-10-18 at 00.05.05.png   84.45K   34 downloads
•  Screen Shot 2012-10-18 at 00.05.20.png   83.08K   35 downloads

[Joel Brown]

This is not a bug, you are searching the KJVS, which is not a unicode text, so it does not support unicode search strings. Why not just search your latvian text directly?

[Ken Simpson]

Hi Joel,
which texts are unicode texts? Is there a way of telling?

[Joel Brown]

Any user bibles that are imported that need to be represented as unicode, will be. The only official texts that are unicode are the Chinese, Japanese, Korean, and Russian texts, I believe.

[evhatfield]

Is there a reason that all Texts can't be treated as Unicode? Other languages语言 including English英语 can be stored in Unicode strings as well. See, I just did it. At least the search text box should handle Unicode characters at all times. That would give a user an opportunity to paste the Unicode word into text box first and then change the Search Text to the appropriate Text. I run into this same problem with Chinese all the time. I have to change the Search Text to my Chinese Text and then repaste the word I want to search for. Whether you want to call it a bug or not, I feel it is an unnecessary step the user has to do in order to make it work. Worst case, if the search text box can contain Unicode words all the time, if I forget to change my Search Text from an English text to a Chinese text, I should just get no results.

[Joel Brown]

When searching original languages, such as Hebrew, Greek, Syriac, etc, we don't use Unicode as we render explicitly with our fonts. However, I agree that when rendering English we could at least allow unicode characters, even if they won't search. We'll look into this for 10.1.2. For the record, though, even if the text looks like jibberish when in English, if you do change to Chinese (or other unicode bible) without deleting/repasting, it will work just fine.

Thanks for the feedback.