X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=datamodel%2Fcompbio%2Fdata%2Fsequence%2FFastaReader.java;h=6c3e9432fb416ae9abd1c4c5c59ac92df6f41f64;hb=8ce6db56df6c0a1c2baca2fc2d33bcb10067c6cd;hp=10eec8f60c50377ce548fb4b61dfb398bd8167c7;hpb=26166c1ca1dd3761305024cef6ee6ae67c6c135f;p=jabaws.git

diff --git a/datamodel/compbio/data/sequence/FastaReader.java b/datamodel/compbio/data/sequence/FastaReader.java
index 10eec8f..6c3e943 100644
--- a/datamodel/compbio/data/sequence/FastaReader.java
+++ b/datamodel/compbio/data/sequence/FastaReader.java
@@ -1,9 +1,30 @@
+/* Copyright (c) 2011 Peter Troshin
+ *  
+ *  JAva Bioinformatics Analysis Web Services (JABAWS) @version: 2.0
+ * 
+ *  This library is free software; you can redistribute it and/or modify it under the terms of the
+ *  Apache License version 2 as published by the Apache Software Foundation
+ * 
+ *  This library is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without
+ *  even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the Apache 
+ *  License for more details.
+ * 
+ *  A copy of the license is in apache_license.txt. It is also available here:
+ * @see: http://www.apache.org/licenses/LICENSE-2.0.txt
+ * 
+ * Any republication or derived work distributed in source code form
+ * must include this copyright and license notice.
+ */
 package compbio.data.sequence;
 
 import java.io.File;
 import java.io.FileNotFoundException;
+import java.io.InputStream;
 import java.util.Iterator;
 import java.util.Scanner;
+import java.util.regex.MatchResult;
+
+import javax.vecmath.MismatchedSizeException;
 
 import compbio.util.Util;
 
@@ -49,23 +70,53 @@ import compbio.util.Util;
 public class FastaReader implements Iterator<FastaSequence> {
 
 	private final Scanner input;
-
+	/**
+	 * Delimiter for the scanner
+	 */
+	private final String DELIM=">";
 	/**
 	 * Header data can contain non-ASCII symbols and read in UTF8
 	 * 
-	 * @param input
+	 * @param inputFile
 	 *            the file containing the list of FASTA formatted sequences to
 	 *            read from
 	 * @throws FileNotFoundException
 	 *             if the input file is not found
+	 * @throws IllegalStateException
+	 *             if the close method was called on this instance
+	 * 
 	 */
-	public FastaReader(final String input) throws FileNotFoundException {
-		this.input = new Scanner(new File(input), "UTF8");
-		this.input.useDelimiter("\\s*>");
+	public FastaReader(final String inputFile) throws FileNotFoundException {
+		input = new Scanner(new File(inputFile), "UTF8");
+		input.useDelimiter(DELIM);
+		Runtime.getRuntime().addShutdownHook(new Thread() {
+
+			@Override
+			public void run() {
+				if (input != null) {
+					input.close();
+				}
+			}
+		});
 	}
 
 	/**
+	 * This class will not close the incoming stream! So the client should do
+	 * so.
+	 * 
+	 * @param inputStream
+	 * @throws FileNotFoundException
+	 */
+	public FastaReader(final InputStream inputStream)
+			throws FileNotFoundException {
+		input = new Scanner(inputStream);
+		input.useDelimiter(DELIM);
+	}
+	/**
 	 * {@inheritDoc}
+	 * 
+	 * @throws IllegalStateException
+	 *             if the close method was called on this instance
 	 */
 	@Override
 	public boolean hasNext() {
@@ -77,10 +128,19 @@ public class FastaReader implements Iterator<FastaSequence> {
 	 * 
 	 * @throws AssertionError
 	 *             if the header or the sequence is missing
+	 * @throws IllegalStateException
+	 *             if the close method was called on this instance
+	 *             @throws MismatchException - if there were no more FastaSequence's.
 	 */
 	@Override
 	public FastaSequence next() {
-		return FastaReader.toFastaSequence(input.next());
+		String fastaHeader=input.next();
+		while (fastaHeader.indexOf("\n")<0 && input.hasNext())
+		{
+			fastaHeader = fastaHeader.concat(">");
+			fastaHeader = fastaHeader.concat(input.next());
+		}
+		return FastaReader.toFastaSequence(fastaHeader);
 	}
 
 	/**
@@ -91,27 +151,35 @@ public class FastaReader implements Iterator<FastaSequence> {
 		throw new UnsupportedOperationException();
 	}
 
+	/**
+	 * Call this method to close the connection to the input file if you want to
+	 * free up the resources. The connection will be closed on the JVM shutdown
+	 * if this method was not called explicitly. No further reading on this
+	 * instance of the FastaReader will be possible after calling this method.
+	 */
+	public void close() {
+		input.close();
+	}
+
 	private static FastaSequence toFastaSequence(final String singleFastaEntry) {
-		final Scanner sc = new Scanner(singleFastaEntry);
-		// Use new line delimiter
-		sc.useDelimiter("\n");
-		if (!sc.hasNext()) {
+
+		assert !Util.isEmpty(singleFastaEntry) : "Empty String where FASTA sequence is expected!";
+
+		int nlineidx = singleFastaEntry.indexOf("\n");
+		if (nlineidx < 0) {
 			throw new AssertionError(
 					"The FASTA sequence must contain the header information"
 							+ " separated by the new line from the sequence. Given sequence does not appear to "
 							+ "contain the header! Given data:\n "
 							+ singleFastaEntry);
 		}
-		String header = sc.next();
+		String header = singleFastaEntry.substring(0, nlineidx);
+
 		// Get rid of the new line chars (should cover common cases)
-		header = header.replaceAll("\n", "").replaceAll("\r", "");
+		header = header.replaceAll("\r", "");
+
+		String sequence = singleFastaEntry.substring(nlineidx);
 
-		sc.useDelimiter("\\s*");
-		final StringBuilder sb = new StringBuilder();
-		while (sc.hasNext()) {
-			sb.append(sc.next().trim());
-		}
-		final String sequence = sb.toString();
 		if (Util.isEmpty(sequence)) {
 			throw new AssertionError(
 					"Empty sequences are not allowed! Please make sure the "