Glad it worked, if you have any further questions feel free to ask. I'm
using it extensively and it and Clojure seem to be a perfect match these
days, especially on very large datasets.
--Joseph
On Friday, December 6, 2013 12:56:40 AM UTC-6, Himakshi Mangal wrote:
>
> Hi Joseph Guhlin,
>
> Thanks your idea helped and i could send some sample data to my neo4j
> database.
>
> Thank you very much... :)
>
>
> On Monday, December 2, 2013 3:41:53 PM UTC+5:30, Himakshi Mangal wrote:
>>
>> Hi...
>>
>>
>> I am using clojure to import dbpedia data into neo4j.
>>
>> Here's the code:
>> (ns opal.dbpedia
>> (:use [clojure.tools.logging :only [log]])
>> (:require [clojure.java.io :as io])
>> (:import [uk.ac.manchester.cs.owl.owlapi.turtle.parser TurtleParser]
>> [org.neo4j.unsafe.batchinsert BatchInserters]
>> [org.neo4j.graphdb DynamicRelationshipType]))
>>
>> ;; PARSING METHODS
>>
>> (defn get-next-tuple
>> [parser]
>> (let [last-item (atom nil)
>> tuple (atom [])]
>> (while (and (not= "." @last-item)
>> (not= "" @last-item))
>> (reset! last-item
>> (-> parser
>> (.getNextToken)
>> (.toString)))
>> (swap! tuple conj @last-item))
>> (when-not (empty? (first @tuple)) ; .getNextToken returns "" once you
>> are out of data
>> @tuple)))
>>
>> (defn seq-of-parser
>> [parser]
>> (if-let [next-tuple (get-next-tuple parser)]
>> (lazy-cat [next-tuple]
>> (seq-of-parser parser))))
>>
>> (defn parse-file
>> [filename]
>> (seq-of-parser
>> (TurtleParser.
>> (io/input-stream filename))))
>>
>> ;; BATCH UPSERT METHODS
>>
>> (def id-map (atom nil))
>> (defn insert-resource-node!
>> [inserter res]
>> (if-let [id (get @id-map res)]
>> ; If the resource has aleady been added, just return the id.
>> id
>> ; Otherwise, add the node for the node, and remember its id for later.
>> (let [id (.createNode inserter {"resource" res})]
>> (swap! id-map #(assoc! % res id))
>> id)))
>>
>> (defn connect-resource-nodes!
>> [inserter node1 node2 label]
>> (let [relationship (DynamicRelationshipType/withName label)]
>> (.createRelationship inserter node1 node2 relationship nil)))
>>
>> (defn insert-tuple!
>> [inserter tuple]
>> ; Get the resource and label names out of the tuple.
>> (let [[resource-1 label resource-2 & _ ] tuple
>> ; Upsert the resource nodes.
>> node-1 (insert-resource-node! inserter resource-1)
>> node-2 (insert-resource-node! inserter resource-2)]
>> ; Connect the nodes with an edge.
>> (connect-resource-nodes! inserter node-1 node-2 label)))
>>
>> (defn -main [graph-path & files]
>> (let [inserter (BatchInserters/inserter graph-path)]
>> (doseq [file files]
>> (log :debug (str "Loading file: " file))
>> (let [c (atom 0)]
>> (doseq [tuple (parse-file file)]
>> (if (= (mod @c 10000) 0)
>> (log :debug (str file ": " @c)))
>> (swap! c inc)
>> (insert-tuple! inserter tuple))))
>> (log :debug "Loading complete.")
>> (log :debug "Shutting down.")
>> (.shutdown inserter)
>> (log :debug "Shutdown complete!")))
>>
>> I am getting the following errors:
>>
>> IllegalAccessError Transient used by non-owner thread
>> clojure.lang.PersistentArrayMap$TransientArrayMap.ensureEditable
>> (PersistentArrayMap.java:449)
>>
>> &&
>>
>>
>> IllegalArgumentException No matching method found: createNode for class
>> org.neo4j.unsafe.batchinsert.BatchInserterImpl
>> clojure.lang.Reflector.invokeMatchingMethod
>>
>>
>> Can anyone please help me in this.. Am doing something wrong or am i
>> missing something.. I am completely new to clojure. Is there a working
>> example for this?
>>
>>
>> Please help..
>>
>> Thanks
>>
>>
>>
--
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to [email protected]
Note that posts from new members are moderated - please be patient with your
first post.
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/clojure?hl=en
---
You received this message because you are subscribed to the Google Groups
"Clojure" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.
