reinstate word column in icu word table

[nominatim.git] / settings / env.defaults

diff --git a/settings/env.defaults b/settings/env.defaults
index 53efb3f79f016ac835bad7944249b3383c80eeaa..cf1f5108c4ac1ff839c6a4336e9070833af00cc5 100644 (file)
--- a/settings/env.defaults
+++ b/settings/env.defaults
@@ -18,6 +18,12 @@ NOMINATIM_DATABASE_WEBUSER="www-data"
 # Changing this value requires to run 'nominatim refresh --functions'.
 NOMINATIM_DATABASE_MODULE_PATH=
 
+# Tokenizer used for normalizing and parsing queries and names.
+# The tokenizer is set up during import and cannot be changed afterwards
+# without a reimport.
+# Currently available tokenizers: legacy
+NOMINATIM_TOKENIZER="legacy"
+
 # Number of occurances of a word before it is considered frequent.
 # Similar to the concept of stop words. Frequent partial words get ignored
 # or handled differently during search.
@@ -40,6 +46,12 @@ NOMINATIM_LANGUAGES=
 # Changing this value requires a reimport.
 NOMINATIM_TERM_NORMALIZATION=":: NFD (); [[:Nonspacing Mark:] [:Cf:]] >;  :: lower (); [[:Punctuation:][:Space:]]+ > ' '; :: NFC ();"
 
+# Configuration file for the tokenizer.
+# The content depends on the tokenizer used. If left empty the default settings
+# for the chooseen tokenizer will be used. The configuration can only be set
+# on import and not be changed afterwards.
+NOMINATIM_TOKENIZER_CONFIG=
+
 # Search in the Tiger house number data for the US.
 # Note: The tables must already exist or queries will throw errors.
 # Changing this value requires to run ./utils/setup --create-functions --setup-website.
@@ -77,7 +89,7 @@ NOMINATIM_TIGER_DATA_PATH=
 NOMINATIM_WIKIPEDIA_DATA_PATH=
 
 # Configuration file for special phrase import.
-# When unset, the internal default settings from 'settings/phrase_settings.php'
+# When unset, the internal default settings from 'settings/phrase-settings.json'
 # are used.
 NOMINATIM_PHRASE_CONFIG=
 
@@ -150,14 +162,14 @@ NOMINATIM_REPLICATION_URL="https://planet.openstreetmap.org/replication/minute"
 # Size is in MB.
 NOMINATIM_REPLICATION_MAX_DIFF=50
 
-# Publication interval of the replication service.
+# Publication interval of the replication service (in seconds).
 # Determines when Nominatim will attempt again to download again a new
 # update. The time is computed from the publication date of the last diff
 # downloaded. Setting this to a slightly higher value than the actual
 # publication interval avoids unnecessary rechecks.
 NOMINATIM_REPLICATION_UPDATE_INTERVAL=75
 
-# Wait time to recheck for a pending update.
+# Wait time to recheck for a pending update (in seconds).
 # Time to wait after an expected update was not available on the server.
 NOMINATIM_REPLICATION_RECHECK_INTERVAL=60