DiffFormatter: support core.quotePath = false

core.quotePath = false means that "bytes higher than 0x80 are not
considered "unusal" anymore"[1], i.e., they are not escaped. In
essence this preserves non-ASCII characters in path names in output.

Note that control characters and other special characters in the
ASCII range will still be escaped.

Add a new QuotedString.GIT_PATH_MINIMAL singleton implementing this.
Change the normal GIT_PATH algorithm to use bytes instead of characters
so it can be re-used. Provide a setter in DiffFormatter for the quoting
style so that an application can override the default, which is the
setting from the git config (and by default "true"). Use the new
QuotedString.GIT_PATH_MINIMAL when core.quotePath == false.

[1] https://git-scm.com/docs/git-config#Documentation/git-config.txt-corequotePath

Bug: 552467
Change-Id: Ifcb233e7d10676333bf42011e32d01a4e1138059
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>

4 files changed, 144 insertions, 19 deletions

diff --git a/org.eclipse.jgit.test/tst/org/eclipse/jgit/util/QuotedStringGitPathStyleTest.java b/org.eclipse.jgit.test/tst/org/eclipse/jgit/util/QuotedStringGitPathStyleTest.java
index 9a0c96e171..c09b136ca6 100644
--- a/org.eclipse.jgit.test/tst/org/eclipse/jgit/util/QuotedStringGitPathStyleTest.java
+++ b/org.eclipse.jgit.test/tst/org/eclipse/jgit/util/QuotedStringGitPathStyleTest.java
@@ -45,6 +45,7 @@ package org.eclipse.jgit.util;
 
 import static java.nio.charset.StandardCharsets.ISO_8859_1;
 import static org.eclipse.jgit.util.QuotedString.GIT_PATH;
+import static org.eclipse.jgit.util.QuotedString.GIT_PATH_MINIMAL;
 import static org.junit.Assert.assertEquals;
 import static org.junit.Assert.assertFalse;
 import static org.junit.Assert.assertNotSame;
@@ -67,6 +68,12 @@ public class QuotedStringGitPathStyleTest {
 		assertEquals(exp, r);
 	}
 
+	private static void assertDequoteMinimal(String exp, String in) {
+		final byte[] b = ('"' + in + '"').getBytes(ISO_8859_1);
+		final String r = GIT_PATH_MINIMAL.dequote(b, 0, b.length);
+		assertEquals(exp, r);
+	}
+
 	@Test
 	public void testQuote_Empty() {
 		assertEquals("\"\"", GIT_PATH.quote(""));
@@ -206,4 +213,75 @@ public class QuotedStringGitPathStyleTest {
 		assertSame("abc@2x.png", GIT_PATH.quote("abc@2x.png"));
 		assertDequote("abc@2x.png", "abc\\1002x.png");
 	}
+
+	@Test
+	public void testNoQuote() {
+		assertSame("\u00c5ngstr\u00f6m",
+				GIT_PATH_MINIMAL.quote("\u00c5ngstr\u00f6m"));
+	}
+
+	@Test
+	public void testQuoteMinimal() {
+		assertEquals("\"\u00c5n\\\\str\u00f6m\"",
+				GIT_PATH_MINIMAL.quote("\u00c5n\\str\u00f6m"));
+	}
+
+	@Test
+	public void testDequoteMinimal() {
+		assertEquals("\u00c5n\\str\u00f6m", GIT_PATH_MINIMAL
+				.dequote(GIT_PATH_MINIMAL.quote("\u00c5n\\str\u00f6m")));
+
+	}
+
+	@Test
+	public void testRoundtripMinimal() {
+		assertEquals("\u00c5ngstr\u00f6m", GIT_PATH_MINIMAL
+				.dequote(GIT_PATH_MINIMAL.quote("\u00c5ngstr\u00f6m")));
+
+	}
+
+	@Test
+	public void testQuoteMinimalDequoteNormal() {
+		assertEquals("\u00c5n\\str\u00f6m", GIT_PATH
+				.dequote(GIT_PATH_MINIMAL.quote("\u00c5n\\str\u00f6m")));
+
+	}
+
+	@Test
+	public void testQuoteNormalDequoteMinimal() {
+		assertEquals("\u00c5n\\str\u00f6m", GIT_PATH_MINIMAL
+				.dequote(GIT_PATH.quote("\u00c5n\\str\u00f6m")));
+
+	}
+
+	@Test
+	public void testRoundtripMinimalDequoteNormal() {
+		assertEquals("\u00c5ngstr\u00f6m",
+				GIT_PATH.dequote(GIT_PATH_MINIMAL.quote("\u00c5ngstr\u00f6m")));
+
+	}
+
+	@Test
+	public void testRoundtripNormalDequoteMinimal() {
+		assertEquals("\u00c5ngstr\u00f6m",
+				GIT_PATH_MINIMAL.dequote(GIT_PATH.quote("\u00c5ngstr\u00f6m")));
+
+	}
+
+	@Test
+	public void testDequote_UTF8_Minimal() {
+		assertDequoteMinimal("\u00c5ngstr\u00f6m",
+				"\\303\\205ngstr\\303\\266m");
+	}
+
+	@Test
+	public void testDequote_RawUTF8_Minimal() {
+		assertDequoteMinimal("\u00c5ngstr\u00f6m", "\303\205ngstr\303\266m");
+	}
+
+	@Test
+	public void testDequote_RawLatin1_Minimal() {
+		assertDequoteMinimal("\u00c5ngstr\u00f6m", "\305ngstr\366m");
+	}
+
 }
diff --git a/org.eclipse.jgit/src/org/eclipse/jgit/diff/DiffFormatter.java b/org.eclipse.jgit/src/org/eclipse/jgit/diff/DiffFormatter.java
index fa552aabe1..d764499159 100644
--- a/org.eclipse.jgit/src/org/eclipse/jgit/diff/DiffFormatter.java
+++ b/org.eclipse.jgit/src/org/eclipse/jgit/diff/DiffFormatter.java
@@ -145,6 +145,8 @@ public class DiffFormatter implements AutoCloseable {
 
 	private Repository repository;
 
+	private Boolean quotePaths;
+
 	/**
 	 * Create a new formatter with a default level of context.
 	 *
@@ -199,6 +201,11 @@ public class DiffFormatter implements AutoCloseable {
 		this.closeReader = closeReader;
 		this.reader = reader;
 		this.diffCfg = cfg.get(DiffConfig.KEY);
+		if (quotePaths == null) {
+			quotePaths = Boolean
+					.valueOf(cfg.getBoolean(ConfigConstants.CONFIG_CORE_SECTION,
+							ConfigConstants.CONFIG_KEY_QUOTE_PATH, true));
+		}
 
 		ContentSource cs = ContentSource.create(reader);
 		source = new ContentSource.Pair(cs, cs);
@@ -379,6 +386,21 @@ public class DiffFormatter implements AutoCloseable {
 	}
 
 	/**
+	 * Sets whether or not path names should be quoted.
+	 * <p>
+	 * By default the setting of git config {@code core.quotePath} is active,
+	 * but this can be overridden through this method.
+	 * </p>
+	 *
+	 * @param quote
+	 *            whether to quote path names
+	 * @since 5.6
+	 */
+	public void setQuotePaths(boolean quote) {
+		quotePaths = Boolean.valueOf(quote);
+	}
+
+	/**
 	 * Set the filter to produce only specific paths.
 	 *
 	 * If the filter is an instance of
@@ -726,8 +748,11 @@ public class DiffFormatter implements AutoCloseable {
 		return id.name();
 	}
 
-	private static String quotePath(String name) {
-		return QuotedString.GIT_PATH.quote(name);
+	private String quotePath(String path) {
+		if (quotePaths == null || quotePaths.booleanValue()) {
+			return QuotedString.GIT_PATH.quote(path);
+		}
+		return QuotedString.GIT_PATH_MINIMAL.quote(path);
 	}
 
 	/**
diff --git a/org.eclipse.jgit/src/org/eclipse/jgit/lib/ConfigConstants.java b/org.eclipse.jgit/src/org/eclipse/jgit/lib/ConfigConstants.java
index f2f1d5a48f..e0bd592194 100644
--- a/org.eclipse.jgit/src/org/eclipse/jgit/lib/ConfigConstants.java
+++ b/org.eclipse.jgit/src/org/eclipse/jgit/lib/ConfigConstants.java
@@ -155,6 +155,12 @@ public final class ConfigConstants {
 	 */
 	public static final String CONFIG_KEY_HOOKS_PATH = "hooksPath";
 
+	/**
+	 * The "quotePath" key.
+	 * @since 5.6
+	 */
+	public static final String CONFIG_KEY_QUOTE_PATH = "quotePath";
+
 	/** The "algorithm" key */
 	public static final String CONFIG_KEY_ALGORITHM = "algorithm";
 
diff --git a/org.eclipse.jgit/src/org/eclipse/jgit/util/QuotedString.java b/org.eclipse.jgit/src/org/eclipse/jgit/util/QuotedString.java
index a55cad3705..2b2358abe9 100644
--- a/org.eclipse.jgit/src/org/eclipse/jgit/util/QuotedString.java
+++ b/org.eclipse.jgit/src/org/eclipse/jgit/util/QuotedString.java
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2008, Google Inc.
+ * Copyright (C) 2008, 2019 Google Inc.
  * and other copyright owners as documented in the project's IP log.
  *
  * This program and the accompanying materials are made available
@@ -54,7 +54,15 @@ import org.eclipse.jgit.lib.Constants;
  */
 public abstract class QuotedString {
 	/** Quoting style that obeys the rules Git applies to file names */
-	public static final GitPathStyle GIT_PATH = new GitPathStyle();
+	public static final GitPathStyle GIT_PATH = new GitPathStyle(true);
+
+	/**
+	 * Quoting style that obeys the rules Git applies to file names when
+	 * {@code core.quotePath = false}.
+	 *
+	 * @since 5.6
+	 */
+	public static final QuotedString GIT_PATH_MINIMAL = new GitPathStyle(false);
 
 	/**
 	 * Quoting style used by the Bourne shell.
@@ -256,40 +264,48 @@ public abstract class QuotedString {
 			quote['"'] = '"';
 		}
 
+		private final boolean quoteHigh;
+
 		@Override
 		public String quote(String instr) {
-			if (instr.length() == 0)
+			if (instr.isEmpty()) {
 				return "\"\""; //$NON-NLS-1$
+			}
 			boolean reuse = true;
 			final byte[] in = Constants.encode(instr);
-			final StringBuilder r = new StringBuilder(2 + in.length);
-			r.append('"');
+			final byte[] out = new byte[4 * in.length + 2];
+			int o = 0;
+			out[o++] = '"';
 			for (int i = 0; i < in.length; i++) {
 				final int c = in[i] & 0xff;
 				if (c < quote.length) {
 					final byte style = quote[c];
 					if (style == 0) {
-						r.append((char) c);
+						out[o++] = (byte) c;
 						continue;
 					}
 					if (style > 0) {
 						reuse = false;
-						r.append('\\');
-						r.append((char) style);
+						out[o++] = '\\';
+						out[o++] = style;
 						continue;
 					}
+				} else if (!quoteHigh) {
+					out[o++] = (byte) c;
+					continue;
 				}
 
 				reuse = false;
-				r.append('\\');
-				r.append((char) (((c >> 6) & 03) + '0'));
-				r.append((char) (((c >> 3) & 07) + '0'));
-				r.append((char) (((c >> 0) & 07) + '0'));
+				out[o++] = '\\';
+				out[o++] = (byte) (((c >> 6) & 03) + '0');
+				out[o++] = (byte) (((c >> 3) & 07) + '0');
+				out[o++] = (byte) (((c >> 0) & 07) + '0');
 			}
-			if (reuse)
+			if (reuse) {
 				return instr;
-			r.append('"');
-			return r.toString();
+			}
+			out[o++] = '"';
+			return new String(out, 0, o, UTF_8);
 		}
 
 		@Override
@@ -375,8 +391,8 @@ public abstract class QuotedString {
 			return RawParseUtils.decode(UTF_8, r, 0, rPtr);
 		}
 
-		private GitPathStyle() {
-			// Singleton
+		private GitPathStyle(boolean doQuote) {
+			quoteHigh = doQuote;
 		}
 	}
 }