Added ICU chain component - which used to be part of Pazpar2.
[yaz-moved-to-github.git] / util / yaz-icu-example.xml
diff --git a/util/yaz-icu-example.xml b/util/yaz-icu-example.xml
new file mode 100644 (file)
index 0000000..ea7e7d2
--- /dev/null
@@ -0,0 +1,30 @@
+<icu>
+  <icu_chain id="en:field" locale="en">
+    <display/>
+  </icu_chain>
+  <icu_chain id="en:sentence" locale="en">
+    <normalize rule="[:Control:] Any-Remove"/>
+    <tokenize rule="s"/>
+    <display/>
+    <normalize rule="[:Punctuation:] Remove"/>
+    <casemap rule="l"/>
+    <normal/>
+    <sort/>
+  </icu_chain>
+  <icu_chain id="en:word" locale="en">
+    <normalize rule="[:Control:] Any-Remove"/>
+    <tokenize rule="w"/>
+    <normalize rule="[[:WhiteSpace:][:Punctuation:]] Remove"/>
+    <display/>
+    <normalize rule="Lower"/>
+    <normal/>
+  </icu_chain>
+  <icu_chain id="en:number" locale="en">
+    <normalize rule="[:Letter:] Remove"/>
+    <tokenize rule="w"/>
+    <normalize rule="[:^Number:] Remove"/>
+    <display/>
+    <normal/>
+    <sort/>
+  </icu_chain>
+</icu>
\ No newline at end of file