From dcf85f9d5ec456bf8f202147fb75f4eb193e5a38 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jan=20B=C3=B6hmer?= Date: Tue, 16 Jan 2024 22:10:29 +0100 Subject: [PATCH] Add hint about bad unicode treatment in sqlite to dabase choosing guide Related to issue #465 --- docs/installation/choosing_database.md | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/docs/installation/choosing_database.md b/docs/installation/choosing_database.md index 1b4ca223..4bd9a93e 100644 --- a/docs/installation/choosing_database.md +++ b/docs/installation/choosing_database.md @@ -30,7 +30,10 @@ automatically However, SQLite does not support certain operations like regex search, which has to be emulated by PHP and therefore are pretty slow compared to the same operation at MySQL. In future there might be features that may only be available, when -using MySQL. +using MySQL. Also SQLite has limitations in comparisons and sorting of unicode characters, which might lead to unexpected +behavior when using non-ASCII characters in your data. For example `µ` (micro sign) is not seen as equal to `μ`(greek minuscle mu), +therefore searching for `µ` (micro sign) will not find parts containing `μ` (mu) and vice versa. In MySQL identical +looking characters are seen as equal, which is more intuitive in most cases. In general MySQL might perform better for big Part-DB instances with many entries, lots of users and high activity, than SQLite.