You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When you have tags/notebooks with mixed case and a non ASCII char is uppercase/lowercase you can't search for them with the wrong case in the search filter.
In the normalizeText function of the search engine, I see the diacritics are already removed so in theory it should work. But maybe for some reason certain letters aren't processed properly.
Good point. That might be trickier than I thought then as if we can't use the LOWER function, it means we need to normalize folder and tag titles in a separate table too.
When you have tags/notebooks with mixed case and a non ASCII char is uppercase/lowercase you can't search for them with the wrong case in the search filter.
Affected filter:
Environment
Joplin 2.1.0 (dev, win32)
Sync Version: 2
Profile Version: 39
Keychain Supported: Yes
Steps to reproduce
Tag
AÜW
tag:aüw
= no search resultstag:aÜw
= search resultsNotebook
AÜW
notebook:aüw
= no search resultsnotebook:aÜw
= search resultsDescribe what you expected to happen
You can filter even if you don't know the exact casing.
More info
(18) Case-insensitive matching of Unicode characters does not work.
https://www.sqlite.org/faq.html#q18
SQL Example:
SELECT LOWER('MY TEST STRING with non ASCII ÜÄÖÉ ...') AS test;
One possible solution would be to overload the default LOWER() function.
The text was updated successfully, but these errors were encountered: