From: Sarah Hoffmann <lonvia@denofr.de>
Date: Tue, 11 Jan 2022 08:41:07 +0000 (+0100)
Subject: Merge pull request #2571 from lonvia/ukrainian-apostrophe
X-Git-Tag: v4.1.0~99
X-Git-Url: https://git.openstreetmap.org./nominatim.git/commitdiff_plain/d6140d6d54b57ae664db0ffa5bbcac20dc6b937f?hp=a486ee347a747edab07fae71906f1e630411b64e

Merge pull request #2571 from lonvia/ukrainian-apostrophe

Consider "modifier letter apostrophe" to be punctuation
---

diff --git a/settings/icu_tokenizer.yaml b/settings/icu_tokenizer.yaml
index e5cbeb6f..a3c62e67 100644
--- a/settings/icu_tokenizer.yaml
+++ b/settings/icu_tokenizer.yaml
@@ -7,7 +7,7 @@ normalization:
     - "'nº' > 'no'"
     - "ª > a"
     - "º > o"
-    - "[[:Punctuation:][:Symbol:]]  > ' '"
+    - "[[:Punctuation:][:Symbol:]\u02bc]  > ' '"
     - "ß > 'ss'" # German szet is unimbigiously equal to double ss
     - "[^[:Letter:] [:Number:] [:Space:]] >"
     - "[:Lm:] >"