Custom thesaurus of czech synonyms.

Dockerfile

@@ -4,3 +4,4 @@ RUN elasticsearch-plugin remove --purge x-pack
 RUN elasticsearch-plugin install analysis-icu
 
 ADD hunspell/ /usr/share/elasticsearch/config/hunspell/
+ADD analysis/ /usr/share/elasticsearch/config/analysis/

README.md

@@ -30,7 +30,11 @@ Create Index with these settings:
                     'type': 'hunspell',
                     'locale': 'cs_CZ',
                     'dedup': True,
-                }
+                },
+                'czech_synonym': {
+                    'type': 'synonym',
+                    'synonyms_path': 'analysis/cs_CZ/synonym.txt',
+                },
             },
             'analyzer': {
                 'czech': {
@@ -38,6 +42,7 @@ Create Index with these settings:
                     'filter': [
                         'icu_folding',
                         'lowercase',
+                        'czech_synonym',
                         'czech_stop',
                         'czech_stemmer',
                         'cs_CZ',
@@ -51,6 +56,9 @@ Create Index with these settings:
 
 You can use `'czech'` analyzer on text fields now.
 
+There is custom dictionary of synonyms included. You can adjust it for your
+needs or remove it from analyzer settings.
+
 ## Build
 
 If you don't want to use pre-built container from

analysis/cs_CZ/synonym.txt

+# Czech Synonyms in Solr format
+# =============================
+#
+# Write comma separated list of words without diacritic in lowercase.
+# One set of synonyms per line.
+
+kava, kafe
+tramvaj, salina