Hadoop Learning-URL method to access HDFS data-day04
Import java.io.ByteArrayOutputStream
Import java.io.InputStream
Import java.net.URL
Import org.apache.hadoop.fs.FsUrlStreamHandlerFactory
Import org.apache.hadoop.io.IOUtils
Import org.junit.Test
Public class TestFileSystemURL {
Static {
URL.setURLStreamHandlerFactory (new FsUrlStreamHandlerFactory ())
}
/ * *
* read HDFS file data through URL object
*
* @ author Administrator
*
* to read data from HDFS, the first thing we consider is to extract data from WEBUI.
* 1. Define a URL object that encapsulates the URL address information of the HDFS
* 2. Call the openStream () method to open an input stream
* 3. Copy the input stream to the byte array output stream through the IOUtils utility class, and then print the information of the output byte output stream
*
, /
@ Test
Public void readByURL () throws Exception {
/ / configure files with URL addresses to be read
String urlStr = "hdfs://hadoop01:9000/user/hadoop/data/hello.txt"
/ / create URL objects through URL's constructor with string
URL url = new URL (urlStr)
/ / call the openStream () method of the URL object to get an inputstream object
InputStream is = url.openStream ()
/ / define an array of characters
/ / byte [] bys = new byte [1024]
/ / int len = 0
/ / there are three parts, read, assign, judge, read one character array at a time, and the return value is the length of the read character array.
/ / at the end of the file, return-1
/ / while ((len=is.read (bys))! =-1) {
/ / outputs an array of characters, part of the array of output characters, from 0 to the end of the read length, with no addition to print.
/ / ln
/ / System.out.print (new String (bys,0,len))
/ /}
/ / create byte array output stream object
ByteArrayOutputStream baos = new ByteArrayOutputStream ()
/ / IOUtils method provided by hadoop to realize stream copy
IOUtils.copyBytes (is, baos, 1024)
/ / close the stream object
IOUtils.closeStream (is)
System.out.println (new String (baos.toByteArray ()
}
}
