Add workflow for csv

Author: TobiasNx

Concordance-RVK-Verbundbibliothek/bulk.csv

@@ -0,0 +1 @@
+"HT013166356, HT018625006, TT000577460","CI 5310","CI 5603","CI 1100","CI 1125","CI 5603","CI 5604","EC 2430","IH 34381"

Concordance-RVK-Verbundbibliothek/bulk.json

@@ -0,0 +1 @@
+"HT013166356, HT018625006, TT000577460","CI 5310","CI 5603","CI 1100","CI 1125","CI 5603","CI 5604","EC 2430","IH 34381"

Concordance-RVK-Verbundbibliothek/culturegraph_to_Rvk-Verbundbibliothek_concordance_csv.flux

@@ -0,0 +1,27 @@
+// Die flux filtert mittels morph-cg-to-es.xml die Isil DE-605 aus
+// This flux uses morph-cg-to-es.xml to filter records with holdings
+// by Isil DE-605 from culturegraph aggregate marcxml. It then builds
+// a concordance Id<->RVK which can be indexed directly into elasticsearch.
+// Snippet from the output json:
+//
+//{"index":{"_index":"cgrvk","_type":"rvk"}}
+//{"rvk":["CI 1100","5,1"],"hbzId":"HT018839495, HT018625006"}
+//
+// Use curl to bulk load the file:
+//
+//  curl -XPOST --header 'Content-Type: application/x-ndjson' -d @bulk.ndjson 'http://localhost:9200/_bulk'
+
+default outfile = FLUX_DIR + "bulk.json";
+default infile = FLUX_DIR + "aggregate_auslieferung_20191212.small.marcxml.gz";
+default fixfile = FLUX_DIR + "fix-cg-to-es.fix";
+
+
+infile
+| open-file
+| decode-xml
+| handle-marcxml
+| fix(fixfile)
+| encode-csv
+//encode-json
+| write(outfile)
+;

Concordance-RVK-Verbundbibliothek/fix-cg-to-es.fix

@@ -1,5 +1,4 @@
-add_field("id","")
-set_array("@id")
+set_array("id")
 set_array("rvk[]")
 
 do list(path: "084??", "var": "$i")
@@ -8,14 +7,14 @@ do list(path: "084??", "var": "$i")
     end
 end
 
+
 do list(path: "035??", "var": "$i")
     if any_match("$i.a", "^\\(DE-605\\)(.*)")
-        copy_field("$i.a","@id.$append")
+        copy_field("$i.a","id.$append")
     end
 end
-replace_all("@id.*","^\\(DE-605\\)(.*)","$1")
-join_field("@id",", ")
-move_field("@id","id")
+replace_all("id.*","^\\(DE-605\\)(.*)","$1")
+join_field("id",", ")
 
 retain("rvk[]","id")
 vacuum()
@@ -25,7 +24,7 @@ unless exists("rvk[]")
     reject()
 end
 
-# Filter records without hbz @ids
+# Filter records without hbz ids
 unless exists("id")
     reject()
 end