Merge remote-tracking branch 'upstream/master'

[nominatim.git] / settings / icu_tokenizer.yaml

diff --git a/settings/icu_tokenizer.yaml b/settings/icu_tokenizer.yaml
index a3c62e67b8d2a500baee37223f56b151bce14324..bf51f56344e12ae4e6986d8fc60683fa77a8d002 100644 (file)
--- a/settings/icu_tokenizer.yaml
+++ b/settings/icu_tokenizer.yaml
@@ -27,6 +27,11 @@ transliteration:
 sanitizers:
     - step: split-name-list
     - step: strip-brace-terms
+    - step: clean-housenumbers
+      filter-kind:
+        - housenumber
+        - conscriptionnumber
+        - streetnumber
     - step: tag-analyzer-by-language
       filter-kind: [".*name.*"]
       whitelist: [bg,ca,cs,da,de,el,en,es,et,eu,fi,fr,gl,hu,it,ja,mg,ms,nl,no,pl,pt,ro,ru,sk,sl,sv,tr,uk,vi]
@@ -59,6 +64,13 @@ token-analysis:
       mode: variant-only
       variants:
           - !include icu-rules/variants-de.yaml
+      mutations:
+          - pattern: ä
+            replacements: ["ä", "ae"]
+          - pattern: ö
+            replacements: ["ö", "oe"]
+          - pattern: ü
+            replacements: ["ü", "ue"]
     - id: el
       analyzer: generic
       mode: variant-only