Add mb13, robust04, and wash18 collections. (#7)

castorini · Nov 5, 2018 · 34abd17d0567fa81ed8726ae656f8c5f8831bfbe · 34abd17
1 parent 7935981
commit 34abd17d0567fa81ed8726ae656f8c5f8831bfbe
diff --git a/Dockerfile b/Dockerfile
@@ -21,6 +21,9 @@ COPY --chown=solr configsets/. server/solr/configsets
 # Create cores
 RUN precreate-core core17 server/solr/configsets/core17
 RUN precreate-core mb11 server/solr/configsets/mb11
+  
+RUN precreate-core mb13 server/solr/configsets/mb13
+  
+RUN precreate-core robust04 server/solr/configsets/robust04
+  
+RUN precreate-core wash18 server/solr/configsets/wash18
 # Start the server.
 CMD solr-foreground
diff --git a/config.json b/config.json
@@ -4,13 +4,28 @@
  "indexes": [
    {
      "name": "core17",
-  
-      "index_path": "../index/lucene-index.core17.pos+docvectors+rawdocs",
+  
+      "index_path": "/tuna1/indexes/lucene-index.core17.pos+docvectors+rawdocs",
      "config_path": "configsets/core17"
    },
    {
      "name": "mb11",
-  
-      "index_path": "../index/lucene-index.mb11.pos+docvectors+rawdocs",
+  
+      "index_path": "/tuna1/indexes/lucene-index.mb11.pos+docvectors+rawdocs",
      "config_path": "configsets/mb11"
+  
+    },
+  
+    {
+  
+      "name": "mb13",
+  
+      "index_path": "/tuna1/indexes/lucene-index.mb13.pos+docvectors+rawdocs",
+  
+      "config_path": "configsets/mb13"
+  
+    },
+  
+    {
+  
+      "name": "robust04",
+  
+      "index_path": "/tuna1/indexes/lucene-index.robust04.pos+docvectors+rawdocs",
+  
+      "config_path": "configsets/robust04"
+  
+    },
+  
+    {
+  
+      "name": "wash18",
+  
+      "index_path": "/tuna1/indexes/lucene-index.wash18.pos+docvectors+rawdocs",
+  
+      "config_path": "configsets/wash18"
    }
  ]
-  
-}
+  
+}
diff --git a/configsets/mb13/conf/lang/contractions_ca.txt b/configsets/mb13/conf/lang/contractions_ca.txt
@@ -0,0 +1,8 @@
+  
+# Set of Catalan contractions for ElisionFilter
+  
+# TODO: load this as a resource from the analyzer and sync it in build.xml
+  
+d
+  
+l
+  
+m
+  
+n
+  
+s
+  
+t
diff --git a/configsets/mb13/conf/lang/contractions_fr.txt b/configsets/mb13/conf/lang/contractions_fr.txt
@@ -0,0 +1,15 @@
+  
+# Set of French contractions for ElisionFilter
+  
+# TODO: load this as a resource from the analyzer and sync it in build.xml
+  
+l
+  
+m
+  
+t
+  
+qu
+  
+n
+  
+s
+  
+j
+  
+d
+  
+c
+  
+jusqu
+  
+quoiqu
+  
+lorsqu
+  
+puisqu
diff --git a/configsets/mb13/conf/lang/contractions_ga.txt b/configsets/mb13/conf/lang/contractions_ga.txt
@@ -0,0 +1,5 @@
+  
+# Set of Irish contractions for ElisionFilter
+  
+# TODO: load this as a resource from the analyzer and sync it in build.xml
+  
+d
+  
+m
+  
+b
diff --git a/configsets/mb13/conf/lang/contractions_it.txt b/configsets/mb13/conf/lang/contractions_it.txt
@@ -0,0 +1,23 @@
+  
+# Set of Italian contractions for ElisionFilter
+  
+# TODO: load this as a resource from the analyzer and sync it in build.xml
+  
+c
+  
+l 
+  
+all 
+  
+dall 
+  
+dell 
+  
+nell 
+  
+sull 
+  
+coll 
+  
+pell 
+  
+gl 
+  
+agl 
+  
+dagl 
+  
+degl 
+  
+negl 
+  
+sugl 
+  
+un 
+  
+m 
+  
+t 
+  
+s 
+  
+v 
+  
+d
diff --git a/configsets/mb13/conf/lang/hyphenations_ga.txt b/configsets/mb13/conf/lang/hyphenations_ga.txt
@@ -0,0 +1,5 @@
+  
+# Set of Irish hyphenations for StopFilter
+  
+# TODO: load this as a resource from the analyzer and sync it in build.xml
+  
+h
+  
+n
+  
+t
diff --git a/configsets/mb13/conf/lang/stemdict_nl.txt b/configsets/mb13/conf/lang/stemdict_nl.txt
@@ -0,0 +1,6 @@
+  
+# Set of overrides for the dutch stemmer
+  
+# TODO: load this as a resource from the analyzer and sync it in build.xml
+  
+fiets	fiets
+  
+bromfiets	bromfiets
+  
+ei	eier
+  
+kind	kinder