The Hortonworks Community Connection is now live. A completely rebuilt Q&A forum, Knowledge Base, Code Hub and more, backed by the experts in the industry.

You will be redirected here in 10 seconds. If your are not redirected, click here to visit the new site.

The legacy Hortonworks Forum is now closed. You can view a read-only version of the former site by clicking here. The site will be taken offline on January 31,2016

HDFS Forum

Loading a Fixed Width File

  • #43477
    Seth Ramey

    I’m trying to figure out how to load the file below. I’ve heard that the best way is something to do with streaming but I don’t know to actually do this.

    Rows that begin with a “01” denote a new record. There can be many (variable number of) rows between “01” rows, but we are only interested in a handful (starts with “02” or “03” or “23” for example). For those rows, we may need to grab 10 characters beginning at position 7 for example. Another row may require 7 characters at position 140.

    My question is – how would we iterate through that via the query tools in Pig or Hive or some other method?

    I’ve uploaded a picture of a sample of the file below:

  • Author
  • #43487
    Seth Ramey

    If it helps, here are the rows we are interested in that I’m currently parsing via a small custom c# program. So to obtain the well district, I look on row (that starts with) 01, go to char 14 and get the next 2 characters. Assumption is that rows positions are zero based:

    public struct PositionLength
    public int position, length;
    public string rowbeginswith;

    public PositionLength(string row, int pos, int len)
    rowbeginswith = row;
    position = pos;
    length = len;

    static PositionLength pl_district = new PositionLength(“01”, 14, 2);
    static PositionLength pl_apinumber = new PositionLength(“01”, 2, 9);
    static PositionLength pl_total_depth = new PositionLength(“01”, 28, 5);
    static PositionLength pl_isplugged = new PositionLength(“01”, 90, 1);

    static PositionLength pl_oilgas = new PositionLength(“02”, 2, 1);

    static PositionLength pl_wellcompletiondate = new PositionLength(“03”, 39, 8);

    static PositionLength pl_lat = new PositionLength(“13”, 132, 9);
    static PositionLength pl_lng = new PositionLength(“13”, 142, 9);

    static PositionLength pl_oilnumber = new PositionLength(“23”, 50, 6);
    static PositionLength pl_gasnumber = new PositionLength(“23”, 56, 6);
    static PositionLength pl_operator = new PositionLength(“23”, 11, 6);

    Carter Shanklin


    Pig might work, I’m not an expert. Hive — no don’t use that for this use case.

    Streaming refers to Hadoop streaming. It will allow you to use your .NET application in conjunction with Map/Reduce. This requires Hadoop on Windows. See — that’s all I know about it unfortunately.

The forum ‘HDFS’ is closed to new topics and replies.

Support from the Experts

A HDP Support Subscription connects you experts with deep experience running Apache Hadoop in production, at-scale on the most demanding workloads.

Enterprise Support »

Become HDP Certified

Real world training designed by the core architects of Hadoop. Scenario-based training courses are available in-classroom or online from anywhere in the world

Training »

Hortonworks Data Platform
The Hortonworks Data Platform is a 100% open source distribution of Apache Hadoop that is truly enterprise grade having been built, tested and hardened with enterprise rigor.
Get started with Sandbox
Hortonworks Sandbox is a self-contained virtual machine with Apache Hadoop pre-configured alongside a set of hands-on, step-by-step Hadoop tutorials.
Modern Data Architecture
Tackle the challenges of big data. Hadoop integrates with existing EDW, RDBMS and MPP systems to deliver lower cost, higher capacity infrastructure.