masayang · March 9, 2013 07:28 · Mar 9, 2013
diff --git a/mapper.py b/mapper.py
@@ -0,0 +1,15 @@
+#! /usr/bin/env python
+# -*- coding: utf-8 -*- 
+
+from mrjob.job import MRJob
+
+class MRWordCounter(MRJob):
+    def mapper(self, key, line):
+        for word in line.split():
+            yield word, 1
+
+    def reducer(self, word, occurrences):
+        yield word, sum(occurrences)
+
+if __name__ == '__main__':
+    MRWordCounter.run()
diff --git a/mrjob_local.sh b/mrjob_local.sh
@@ -0,0 +1,2 @@
+#ローカルでの実行
+python wc.py < creativecommons.txt
diff --git a/mrjob_s3.sh b/mrjob_s3.sh
@@ -0,0 +1,7 @@
+#s3上ファイルを使ってEMR上で実行
+
+
+export AWS_ACCESS_KEY_ID=<your aws access key>
+export AWS_SECRET_ACCESS_KEY=<your secret access key>
+
+python wc.py -r emr s3://masayang-bootcamp/bootcamp4/EMRconsole/creativecommons.txt s3://masayang-bootcamp/bootcamp4/EMRconsole/creativecommons.txt -o s3://masayang-bootcamp/bootcamp4/EMRconsole/<your account>
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1,2 @@
		#ローカルでの実行
		python wc.py < creativecommons.txt
No results found