Normalizer creating a valid branch name from a string

Generic normalization method for a possible invalid branch name.
The method compresses dividers between spaces, then replaces spaces
and non word characters with underscores.

This method is needed in preparation for subsequent EGit changes.

Bug: 509878
Change-Id: Ic0d12f098f90f912a45bcc5693d6accf751d4e58
Signed-off-by: Wim Jongman <wim.jongman@remainsoftware.com>

2 files changed, 119 insertions, 0 deletions

diff --git a/org.eclipse.jgit.test/tst/org/eclipse/jgit/lib/ValidRefNameTest.java b/org.eclipse.jgit.test/tst/org/eclipse/jgit/lib/ValidRefNameTest.java
index e9d46bb582..9c85fbe1ea 100644
--- a/org.eclipse.jgit.test/tst/org/eclipse/jgit/lib/ValidRefNameTest.java
+++ b/org.eclipse.jgit.test/tst/org/eclipse/jgit/lib/ValidRefNameTest.java
@@ -247,4 +247,86 @@ public class ValidRefNameTest {
 		assertValid(true, "refs/heads/conx");
 		assertValid(true, "refs/heads/xcon");
 	}
+
+	@Test
+	public void testNormalizeBranchName() {
+
+		assertEquals(true, Repository.normalizeBranchName("").equals(""));
+
+		assertEquals(true,
+				Repository.normalizeBranchName("__@#$@#$@$____   _")
+						.equals(""));
+
+		assertEquals(true,
+				Repository.normalizeBranchName("~`!@#$%^&*()_+}]{[|\\\";?>.<,/")
+						.equals(""));
+
+		assertEquals(true,
+				Repository.normalizeBranchName("Bug 12345 :::: Hello World")
+						.equals("Bug_12345-Hello_World"));
+
+		assertEquals(true,
+				Repository.normalizeBranchName("Bug 12345 :::: Hello::: World")
+						.equals("Bug_12345-Hello-_World"));
+
+		assertEquals(true,
+				Repository.normalizeBranchName(":::Bug 12345 - Hello World")
+						.equals("Bug_12345-Hello_World"));
+
+		assertEquals(true,
+				Repository.normalizeBranchName("---Bug 12345 - Hello World")
+						.equals("Bug_12345-Hello_World"));
+
+		assertEquals(true,
+				Repository.normalizeBranchName("Bug 12345 ---- Hello --- World")
+						.equals("Bug_12345-Hello-World"));
+
+		assertEquals(true, Repository.normalizeBranchName(null) == null);
+
+		assertEquals(true,
+				Repository.normalizeBranchName("Bug 12345 - Hello World!")
+						.equals("Bug_12345-Hello_World"));
+
+		assertEquals(true,
+				Repository.normalizeBranchName("Bug 12345 : Hello World!")
+						.equals("Bug_12345-Hello_World"));
+
+		assertEquals(true,
+				Repository.normalizeBranchName("Bug 12345 _ Hello World!")
+						.equals("Bug_12345_Hello_World"));
+
+		assertEquals(true,
+				Repository
+						.normalizeBranchName("Bug 12345   -       Hello World!")
+						.equals("Bug_12345-Hello_World"));
+
+		assertEquals(true,
+				Repository.normalizeBranchName(" Bug 12345   -   Hello World! ")
+						.equals("Bug_12345-Hello_World"));
+
+		assertEquals(true,
+				Repository
+						.normalizeBranchName(" Bug 12345   -   Hello World!   ")
+						.equals("Bug_12345-Hello_World"));
+
+		assertEquals(true,
+				Repository
+						.normalizeBranchName(
+								"Bug 12345   -   Hello______ World!")
+						.equals("Bug_12345-Hello_World"));
+
+		assertEquals(true,
+				Repository.normalizeBranchName("_Bug 12345 - Hello World!")
+						.equals("Bug_12345-Hello_World"));
+
+		assertEquals(true,
+				Repository
+						.normalizeBranchName(
+								"Bug 12345 - Hello Wo!@#$%^&*(rld {@")
+						.equals("Bug_12345-Hello_World_"));
+
+		assertEquals(true,
+				Repository.normalizeBranchName("Bug 1#$  2345 - Hello World")
+						.equals("Bug_12345-Hello_World"));
+	}
 }
diff --git a/org.eclipse.jgit/src/org/eclipse/jgit/lib/Repository.java b/org.eclipse.jgit/src/org/eclipse/jgit/lib/Repository.java
index c5b2ef8e5b..641262cd50 100644
--- a/org.eclipse.jgit/src/org/eclipse/jgit/lib/Repository.java
+++ b/org.eclipse.jgit/src/org/eclipse/jgit/lib/Repository.java
@@ -4,6 +4,7 @@
  * Copyright (C) 2006-2010, Robin Rosenberg <robin.rosenberg@dewire.com>
  * Copyright (C) 2006-2012, Shawn O. Pearce <spearce@spearce.org>
  * Copyright (C) 2012, Daniel Megert <daniel_megert@ch.ibm.com>
+ * Copyright (C) 2017, Wim Jongman <wim.jongman@remainsoftware.com>
  * and other copyright owners as documented in the project's IP log.
  *
  * This program and the accompanying materials are made available
@@ -1887,4 +1888,40 @@ public abstract class Repository implements AutoCloseable {
 	public void autoGC(ProgressMonitor monitor) {
 		// default does nothing
 	}
+
+	/**
+	 * Normalizes the passed branch name into a possible valid branch name. The
+	 * validity of the returned name should be checked by a subsequent call to
+	 * {@link #isValidRefName(String)}.
+	 * <p/>
+	 * Future implementations of this method could be more restrictive or more
+	 * lenient about the validity of specific characters in the returned name.
+	 * <p/>
+	 * The current implementation returns a trimmed string only containing word
+	 * characters ([a-zA-Z_0-9]) and hyphens ('-'). Colons are replaced by
+	 * hyphens. Repeating underscores and hyphens are replaced by a single
+	 * occurrence. Underscores and hyphens at the beginning of the string are
+	 * removed.
+	 *
+	 * @param name
+	 *            The name to normalize.
+	 *
+	 * @return The normalized String or null if null was passed.
+	 * @since 4.7
+	 * @see #isValidRefName(String)
+	 */
+	public static String normalizeBranchName(String name) {
+		if (name == null || name.length() == 0) {
+			return name;
+		}
+		String result = name.trim();
+		return result.replaceAll("\\s+([_:-])*?\\s+", "$1") //$NON-NLS-1$//$NON-NLS-2$
+				.replaceAll(":", "-") //$NON-NLS-1$//$NON-NLS-2$
+				.replaceAll("\\s+", "_") //$NON-NLS-1$//$NON-NLS-2$
+				.replaceAll("_{2,}", "_") //$NON-NLS-1$//$NON-NLS-2$
+				.replaceAll("-{2,}", "-") //$NON-NLS-1$//$NON-NLS-2$
+				.replaceAll("[^\\w-]", "") //$NON-NLS-1$ //$NON-NLS-2$
+				.replaceAll("^_+", "") //$NON-NLS-1$//$NON-NLS-2$
+				.replaceAll("^-+", ""); //$NON-NLS-1$//$NON-NLS-2$
+	}
 }