This project allows to read UTF-8 encoded UNICODE files in Java. It implements an infinite buffer file reader where one can recover the text word by word, which is converted from UTF-8, skipping blanks. The blanks can be recovered since they are all counted (feature used in the test program). By skipping blanks the reader is able to identify the start and the end of the words because they are different than blank. The fact that the text is encoded is totally transparent to the reader. The test program reads an UTF-8 testing file and prints it over the console. A batch file is provided to help running the jar. Another batch file to build the source files and generate the jar using ava CLI in console window is also given.
Due to incompatibilities between different versions of Java, it's better to build the jar from the source files. In order to build the jar in a Windows platform one should open a command window and set it to the directory containing the contents of this repertory. One should not use a Powershell (Powershell causes a bug with the flag -Dfile.encoding=UTF-8), but just a normal Command Prompt. In this window just type:
build
To run the JAR one needs the files: utf8.jar, UTF8.txt and run.bat. These three files should be in the same directory. If running on Windows it is sufficient to type run in a console in this directory. An out.txt should be created and should be identical to the one supplied in this repository. Alternatively, just double click the run file.
The source files and their locations in the file structure are given by the file sources.txt. This files is used for building the jar.
These files are inside package com.filereader.
Input.java is an interface encapsulating the file. It can be used to encapsulate any source of bytes of any kind including Strings:
public interface Input {
public int nextword();
public String getWord();
public int getnblanks();
public boolean end();
}The method nextword() reads the words (assumed to be separated by spaces), one by one, until end of file is reached. It returns an integer indicating the number of characters in the word or -1 in the case of end of file. The method getWord() returns a string with the word read. Since blanks are just separators they are generally discarded, but in the case they need to be recovered, as in the case of UTF8.txt, the method getnblanks() returns the number of blanks skipped.
In the example, blanks are cached into an array that is used to generate a string with the number of required blanks. This caching scheme seems to be more efficient than looping over System.out.print(' ') to print each blank individually. Finally, the method end() returns true if the end of the file has already been reached.
InputFile is an infinite source of input of bytes
from a file. It repeatedly reads the file to a buffer, skipping blanks (here,
blanks are considered as word separators - another word separator can be configured).
It then scans the file while it contains non blank characters. If the buffer was
totally scanned without a blank being found, the word probably continues in the
contents still not read in the file. The non blank content is then saved in an
Arraylist of bytes. Next, the file is read to fill the buffer again and the
scanning process continues in this manner until a blank character is found,
when the cycle starts over. The number of blanks skipped is maintained (as explained
above), in case an exact copy of the file is required.
This word by word parsing method is focused on free style formats as we usually find in programming languages and text data file formats, thus being able to fit to any application using data in text or mixed file formats. To avoid confusion with the UTF-8 encoding, the binary information is recommended to be dealt using escape sequences of ASCII characters.
This class implements and is supposed to be accessed through the interface Input to allow scanning other different sources of bytes. The other methods that can be found in this class, besides those of Input.java, are the ones below.
This method passes a string containing the name of the file to be read. It opens the file and reads
the first chunk of data into the buffer, as shown, and assigning zero to the initial position of the buffer.
It also sets the charset for decoding UTF-8 contents. This was placed here for future versions of this
method, where the encoding name is passed as a parameter. Since the parser implemented here is generic
and essentially reads only raw bytes it can be easily adapted to read files with other encodings.
public void setFile(String file) throws IOException {
position = 0;
UTF8 = Charset.forName("UTF-8");
filein = new FileInputStream(file);
nbytes = filein.read(buffer);
}This is the method that really reads the file and performs what is described above. The actual code is commented in almost every line, as one can see below:
private boolean readword() {
if ( end = (nbytes == -1) ) return false;
int i, j;
nblanks = 0; // there were no blanks skipped yet
rest.clear(); // list should be empty here
//
// Loops while reading buffer, jumping over all blanks
//
for ( ; ; ) { // loops while it finds blanks
// first skip all the blanks already read
for ( i = position; i < nbytes && buffer[i] == ' '; i++ );
nblanks += i - position; // records number of blanks skipped
if (i < nbytes) break; // if still inside the buffer, get out
try { // otherwise read the buffer from file
nbytes = filein.read(buffer);
if (end = (nbytes == -1)) // end of file -> return false
return false;
} catch (IOException e) { // on a exception, call exit
e.printStackTrace();
System.exit(0);
}
position = 0; // starts from beginning of buffer again
} // go back and continue checking new buffer
//
// Loops while reading buffer, jumping over everything that is not blank
//
for ( ; ; ) { // loops while it doesn't find a blank
// first skip everything that is not blank inside the buffer
for ( j = i; j < nbytes && buffer[j] != ' '; j++);
if ( j < nbytes ) { // if still inside the buffer, get in
position = j; // record last position checked
if (rest.length() == 0) { // if there was nothing from previous buffer
init = i;
return true; // just get the word and return true
}
rest.append(buffer, i, j); // otherwise, append new content to
init = -1; // real length is in rest
return true; // return true because word was found
}
// we reached the end of the buffer without finding a blank
if ( j != i ) rest.append(buffer, i, nbytes); // saves previous contents
try { // reads new content from file into buffer
nbytes = filein.read(buffer);
if (end = (nbytes == -1)) { // end of file and no previous content
if (rest.length() == 0) return false; //-> return false
init = -1; // real length is in rest
return true; // return true because previous content
} // was the last word found
} catch (IOException e) {
e.printStackTrace();
System.exit(0);
}
i = 0; // starts over from the start
} // go back and continue checking new buffer
}