hadoop-hdfs-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "David Parks" <davidpark...@yahoo.com>
Subject RE: Hadoop 101
Date Wed, 12 Dec 2012 06:04:09 GMT
You use TextInputFormat, you'll get the following key<LongWritable>,
value<Text> pairs in your mapper:

file_position, your_input

Example:
0,
"0\t[356:0.3481597,359:0.3481597,358:0.3481597,361:0.3481597,360:0.3481597]"
100,
"8\t[356:0.34786037,359:0.34786037,358:0.34786037,361:0.34786037,360:0.34786
037]"
200,
"25\t[284:0.34821576,286:0.34821576,287:0.34821576,288:0.34821576,289:0.3482
1576]"

Then just parse it out in your mapper.


-----Original Message-----
From: Pat Ferrel [mailto:pat.ferrel@gmail.com] 
Sent: Wednesday, December 12, 2012 7:50 AM
To: user@hadoop.apache.org
Subject: Hadoop 101

Stupid question for the day.

I have a file created by a mahout job of the form:

0
[356:0.3481597,359:0.3481597,358:0.3481597,361:0.3481597,360:0.3481597]
8
[356:0.34786037,359:0.34786037,358:0.34786037,361:0.34786037,360:0.34786037]
25
[284:0.34821576,286:0.34821576,287:0.34821576,288:0.34821576,289:0.34821576]
28
[452:0.34802154,454:0.34802154,453:0.34802154,456:0.34802154,455:0.34802154]
.

If this were a SequenceFile I could read it and be merrily on my way but
it's a text file. The classes written are key, value pairs <LongWritable,
VectorWritable> but the file is tab delimited text. 

I was hoping to do something like:

SequenceFile.Reader reader = new SequenceFile.Reader(fs, inputFile, conf);
Writable userId = new LongWritable(); VectorWritable recommendations = new
VectorWritable(); while (reader.next(userId, recommendations)) {
	//do something with each pair
}

But alas Google fails me. How do you read in key, values pairs from text
files outside of a map or reduce? 


Mime
View raw message