Example usage for org.apache.hadoop.util UTF8ByteArrayUtils findBytes

Introduction

In this page you can find the example usage for org.apache.hadoop.util UTF8ByteArrayUtils findBytes.

Prototype

public static int findBytes(byte[] utf, int start, int end, byte[] b) 

Source Link

Document

Find the first occurrence of the given bytes b in a UTF-8 encoded string

Usage

From source file:edu.stolaf.cs.wmrserver.streaming.PipeMapRed.java

License:Apache License

/**
 * Split a line into key and value./*from w ww  .ja  v a  2 s.c  o m*/
 * @param line: a byte array of line containing UTF-8 bytes
 * @param key: key of a record
 * @param val: value of a record
 * @throws IOException
 */
void splitKeyVal(byte[] line, int length, Text key, Text val) throws IOException {
    int numKeyFields = getNumOfKeyFields();
    byte[] separator = getFieldSeparator();

    // Need to find numKeyFields separators
    int pos = UTF8ByteArrayUtils.findBytes(line, 0, length, separator);
    for (int k = 1; k < numKeyFields && pos != -1; k++) {
        pos = UTF8ByteArrayUtils.findBytes(line, pos + separator.length, length, separator);
    }
    try {
        if (pos == -1) {
            key.set(line, 0, length);
            val.set("");
        } else {
            StreamKeyValUtil.splitKeyVal(line, 0, length, key, val, pos, separator.length);
        }
    } catch (CharacterCodingException e) {
        LOG.warn(StringUtils.stringifyException(e));
    }
}

---

From source file:fm.last.darling.hbase.HBaseJSONOutputReader.java

License:Apache License

private void interpretKeyandValue(byte[] line, int length) throws IOException {
    // Need to find numKeyFields separators
    int pos = UTF8ByteArrayUtils.findBytes(line, 0, length, separator);
    for (int k = 1; k < numKeyFields && pos != -1; k++) {
        pos = UTF8ByteArrayUtils.findBytes(line, pos + separator.length, length, separator);
    }/* ww  w. ja  va 2  s . c o  m*/

    Text k = new Text();
    Text v = new Text();
    try {
        if (pos == -1) {
            k.set(line, 0, length);
            v.set("");
        } else {
            StreamKeyValUtil.splitKeyVal(line, 0, length, k, v, pos, separator.length);
        }
    } catch (CharacterCodingException e) {
        throw new IOException(e);
    }

    // removing a ' at the start and end of the key
    byte[] keyBytes = trimOuterBytes(k);

    rowkey = new ImmutableBytesWritable(keyBytes);
    put = new Put(keyBytes);

    String tmpV = v.toString();
    String json = tmpV.substring(1, tmpV.length() - 1);
    Map<String, Map> payload;
    try {
        payload = (Map<String, Map>) ObjectBuilder.fromJSON(json); // the 'erased' type?
    } catch (Exception e) {
        throw new IOException("error, fromJson: ", e);
    }

    Set<Map.Entry<String, Map>> entries = payload.entrySet();
    for (Map.Entry<String, Map> entry : entries) {
        String cfq = entry.getKey(); // let's consider not joining family and qualifier at emitter.
        String[] parts = cfq.split(":");
        if (parts.length < 2)
            continue;
        String family = parts[0];
        String qualifier = parts[1];

        Map dict = entry.getValue(); // unchecked.

        // expecting dict to carry 'value',
        Object value = dict.get("value");
        if (value == null)
            continue; // no good.

        // ..and possibly 'timestamp'.
        //Object ts = 0;
        //if (dict.containsKey("timestamp"))
        //ts = dict.get("timestamp");

        put.add(family.getBytes("UTF-8"), qualifier.getBytes("UTF-8"), value.toString().getBytes("UTF-8"));
    }
}

---