It would be hard to avoid using ArrayList or something similar. If you know the file is ASCII, you could do
int partSize = 500;
File f = new File("file.txt");
String[] parts = new String[(f.length() + partSize - 1) / partSize];
But if the file uses a variable-width encoding like UTF-8, this won't work. This code will do the job.
static String[] readFileInParts(String fname) throws IOException {
int partSize = 500;
FileReader fr = new FileReader(fname);
List<String> parts = new ArrayList<String>();
char[] buf = new char[partSize];
int pos = 0;
for (;;) {
int nRead = fr.read(buf, pos, partSize - pos);
if (nRead == -1) {
if (pos > 0)
parts.add(new String(buf, 0, pos));
break;
}
pos += nRead;
if (pos == partSize) {
parts.add(new String(buf));
pos = 0;
}
}
return parts.toArray(new String[parts.size()]);
}
Note that FileReader uses the platform default encoding. To specify a specific encoding, replace it with new InputStreamReader(new FileInputStream(fname), charSet). It bit ugly, but that's the best way to do it.
ready()fails. We still have to check the number of characters read.