RawParseUtils: utility method to get a header value

The new method takes care of removing the leading blanks on continuation
lines in multi-line headers.

Change-Id: Id5a8063512a2abc3177c104d6ba8fa50d0dc6352

3 files changed, 74 insertions, 8 deletions

diff --git a/org.eclipse.jgit.test/tst/org/eclipse/jgit/util/RawParseUtilsTest.java b/org.eclipse.jgit.test/tst/org/eclipse/jgit/util/RawParseUtilsTest.java
index e80c07509a..d78b372e36 100644
--- a/org.eclipse.jgit.test/tst/org/eclipse/jgit/util/RawParseUtilsTest.java
+++ b/org.eclipse.jgit.test/tst/org/eclipse/jgit/util/RawParseUtilsTest.java
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2011, Leonard Broman <leonard.broman@gmail.com> and others
+ * Copyright (C) 2011, 2024 Leonard Broman <leonard.broman@gmail.com> and others
  *
  * This program and the accompanying materials are made available under the
  * terms of the Eclipse Distribution License v. 1.0 which is available at
@@ -15,6 +15,7 @@ import static org.junit.Assert.fail;
 
 import java.nio.charset.Charset;
 import java.nio.charset.UnsupportedCharsetException;
+import java.util.List;
 
 import org.eclipse.jgit.lib.Constants;
 import org.junit.Test;
@@ -24,7 +25,7 @@ import static java.nio.charset.StandardCharsets.UTF_8;
 public class RawParseUtilsTest {
 	String commit = "tree e3a1035abd2b319bb01e57d69b0ba6cab289297e\n" +
 		"parent 54e895b87c0768d2317a2b17062e3ad9f76a8105\n" +
-		"committer A U Thor <author@xample.com 1528968566 +0200\n" +
+			"committer A U Thor <author@example.com> 1528968566 +0200\n" +
 		"gpgsig -----BEGIN PGP SIGNATURE-----\n" +
 		" \n" +
 		" wsBcBAABCAAQBQJbGB4pCRBK7hj4Ov3rIwAAdHIIAENrvz23867ZgqrmyPemBEZP\n" +
@@ -67,21 +68,24 @@ public class RawParseUtilsTest {
 	public void testHeaderStart() {
 		byte[] headerName = "some".getBytes(UTF_8);
 		byte[] commitBytes = commit.getBytes(UTF_8);
-		assertEquals(625, RawParseUtils.headerStart(headerName, commitBytes, 0));
-		assertEquals(625, RawParseUtils.headerStart(headerName, commitBytes, 4));
+		assertEquals(627,
+				RawParseUtils.headerStart(headerName, commitBytes, 0));
+		assertEquals(627,
+				RawParseUtils.headerStart(headerName, commitBytes, 4));
 
 		byte[] missingHeaderName = "missing".getBytes(UTF_8);
 		assertEquals(-1, RawParseUtils.headerStart(missingHeaderName,
 							   commitBytes, 0));
 
 		byte[] fauxHeaderName = "other".getBytes(UTF_8);
-		assertEquals(-1, RawParseUtils.headerStart(fauxHeaderName, commitBytes, 625 + 4));
+		assertEquals(-1, RawParseUtils.headerStart(fauxHeaderName, commitBytes,
+				627 + 4));
 	}
 
 	@Test
 	public void testHeaderEnd() {
 		byte[] commitBytes = commit.getBytes(UTF_8);
-		int[] expected = new int[] {45, 93, 148, 619, 637};
+		int[] expected = new int[] { 45, 93, 150, 621, 639 };
 		int start = 0;
 		for (int i = 0; i < expected.length; i++) {
 			start = RawParseUtils.headerEnd(commitBytes, start);
@@ -89,4 +93,34 @@ public class RawParseUtilsTest {
 			start += 1;
 		}
 	}
+
+	@Test
+	public void testHeaderValue() {
+		byte[] commitBytes = commit.getBytes(UTF_8);
+		List<String> headers = List.of(
+				"e3a1035abd2b319bb01e57d69b0ba6cab289297e",
+				"54e895b87c0768d2317a2b17062e3ad9f76a8105",
+				"A U Thor <author@example.com> 1528968566 +0200",
+				"-----BEGIN PGP SIGNATURE-----\n"
+						+ "\n"
+						+ "wsBcBAABCAAQBQJbGB4pCRBK7hj4Ov3rIwAAdHIIAENrvz23867ZgqrmyPemBEZP\n"
+						+ "U24B1Tlq/DWvce2buaxmbNQngKZ0pv2s8VMc11916WfTIC9EKvioatmpjduWvhqj\n"
+						+ "znQTFyiMor30pyYsfrqFuQZvqBW01o8GEWqLg8zjf9Rf0R3LlOEw86aT8CdHRlm6\n"
+						+ "wlb22xb8qoX4RB+LYfz7MhK5F+yLOPXZdJnAVbuyoMGRnDpwdzjL5Hj671+XJxN5\n"
+						+ "SasRdhxkkfw/ZnHxaKEc4juMz8Nziz27elRwhOQqlTYoXNJnsV//wy5Losd7aKi1\n"
+						+ "xXXyUpndEOmT0CIcKHrN/kbYoVL28OJaxoBuva3WYQaRrzEe3X02NMxZe9gkSqA=\n"
+						+ "=TClh\n"
+						+ "-----END PGP SIGNATURE-----",
+				"other header");
+		int start = 0;
+		for (String header : headers) {
+			int endOfTag = RawParseUtils.next(commitBytes, start, ' ');
+			int end = RawParseUtils.headerEnd(commitBytes, start);
+
+			assertEquals(header, new String(
+					RawParseUtils.headerValue(commitBytes, endOfTag, end),
+					UTF_8));
+			start = end + 1;
+		}
+	}
 }
diff --git a/org.eclipse.jgit/src/org/eclipse/jgit/revwalk/RevCommit.java b/org.eclipse.jgit/src/org/eclipse/jgit/revwalk/RevCommit.java
index 1f0f13152e..743a8ccce0 100644
--- a/org.eclipse.jgit/src/org/eclipse/jgit/revwalk/RevCommit.java
+++ b/org.eclipse.jgit/src/org/eclipse/jgit/revwalk/RevCommit.java
@@ -17,7 +17,6 @@ import java.io.IOException;
 import java.nio.charset.Charset;
 import java.nio.charset.IllegalCharsetNameException;
 import java.nio.charset.UnsupportedCharsetException;
-import java.util.Arrays;
 import java.util.List;
 
 import org.eclipse.jgit.annotations.Nullable;
@@ -409,7 +408,7 @@ public class RevCommit extends RevObject {
 			return null;
 		}
 		final int end = RawParseUtils.headerEnd(raw, start);
-		return Arrays.copyOfRange(raw, start, end);
+		return RawParseUtils.headerValue(raw, start, end);
 	}
 
 	/**
diff --git a/org.eclipse.jgit/src/org/eclipse/jgit/util/RawParseUtils.java b/org.eclipse.jgit/src/org/eclipse/jgit/util/RawParseUtils.java
index c87a5e60f2..d2ddf2a345 100644
--- a/org.eclipse.jgit/src/org/eclipse/jgit/util/RawParseUtils.java
+++ b/org.eclipse.jgit/src/org/eclipse/jgit/util/RawParseUtils.java
@@ -554,6 +554,39 @@ public final class RawParseUtils {
 	}
 
 	/**
+	 * Extract a part of a buffer as a header value, removing the single blanks
+	 * at the front of continuation lines.
+	 *
+	 * @param b
+	 *            buffer to extract the header from
+	 * @param start
+	 *            of the header value, see
+	 *            {@link #headerStart(byte[], byte[], int)}
+	 * @param end
+	 *            of the header; see
+	 *            {@link #nextLfSkippingSplitLines(byte[], int)}
+	 * @return the header value, with blanks indicating continuation lines
+	 *         stripped
+	 * @since 6.9
+	 */
+	public static final byte[] headerValue(final byte[] b, int start, int end) {
+		byte[] data = new byte[end - start];
+		int out = 0;
+		byte last = '\0';
+		for (int in = start; in < end; in++) {
+			byte ch = b[in];
+			if (ch != ' ' || last != '\n') {
+				data[out++] = ch;
+			}
+			last = ch;
+		}
+		if (out == data.length) {
+			return data;
+		}
+		return Arrays.copyOf(data, out);
+	}
+
+	/**
 	 * Locate the first end of header after the given position. Note that
 	 * headers may be more than one line long.
 	 * <p>