Add support for smudge filters

If defined in .gitattributes call smudge filter during checkout.

To support checkout where current HEAD,index do not contain attributes
we need to also consider attributes from the tree we checkout. Therefore
CanonicalTreeParser has to learn how to provide attributes.

Change-Id: I168fdb81a8e1a9f991587b3e95a36550ea845f0a
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>

2 files changed, 108 insertions, 14 deletions

diff --git a/org.eclipse.jgit/src/org/eclipse/jgit/dircache/DirCacheCheckout.java b/org.eclipse.jgit/src/org/eclipse/jgit/dircache/DirCacheCheckout.java
index 0036ab5089..393a854945 100644
--- a/org.eclipse.jgit/src/org/eclipse/jgit/dircache/DirCacheCheckout.java
+++ b/org.eclipse.jgit/src/org/eclipse/jgit/dircache/DirCacheCheckout.java
@@ -52,15 +52,18 @@ import java.util.HashMap;
 import java.util.List;
 import java.util.Map;
 
+import org.eclipse.jgit.api.errors.FilterFailedException;
 import org.eclipse.jgit.errors.CheckoutConflictException;
 import org.eclipse.jgit.errors.CorruptObjectException;
 import org.eclipse.jgit.errors.IncorrectObjectTypeException;
 import org.eclipse.jgit.errors.IndexWriteException;
 import org.eclipse.jgit.errors.MissingObjectException;
 import org.eclipse.jgit.internal.JGitText;
+import org.eclipse.jgit.lib.Constants;
 import org.eclipse.jgit.lib.CoreConfig.AutoCRLF;
 import org.eclipse.jgit.lib.CoreConfig.SymLinks;
 import org.eclipse.jgit.lib.FileMode;
+import org.eclipse.jgit.lib.NullProgressMonitor;
 import org.eclipse.jgit.lib.ObjectChecker;
 import org.eclipse.jgit.lib.ObjectId;
 import org.eclipse.jgit.lib.ObjectLoader;
@@ -76,6 +79,7 @@ import org.eclipse.jgit.treewalk.WorkingTreeIterator;
 import org.eclipse.jgit.treewalk.WorkingTreeOptions;
 import org.eclipse.jgit.treewalk.filter.PathFilter;
 import org.eclipse.jgit.util.FS;
+import org.eclipse.jgit.util.FS.ExecutionResult;
 import org.eclipse.jgit.util.FileUtils;
 import org.eclipse.jgit.util.RawParseUtils;
 import org.eclipse.jgit.util.SystemReader;
@@ -85,9 +89,10 @@ import org.eclipse.jgit.util.io.AutoCRLFOutputStream;
  * This class handles checking out one or two trees merging with the index.
  */
 public class DirCacheCheckout {
+	private static final int MAX_EXCEPTION_TEXT_SIZE = 10 * 1024;
 	private Repository repo;
 
-	private HashMap<String, ObjectId> updated = new HashMap<String, ObjectId>();
+	private HashMap<String, String> updated = new HashMap<String, String>();
 
 	private ArrayList<String> conflicts = new ArrayList<String>();
 
@@ -112,9 +117,9 @@ public class DirCacheCheckout {
 	private boolean emptyDirCache;
 
 	/**
-	 * @return a list of updated paths and objectIds
+	 * @return a list of updated paths and smudgeFilterCommands
 	 */
-	public Map<String, ObjectId> getUpdated() {
+	public Map<String, String> getUpdated() {
 		return updated;
 	}
 
@@ -447,7 +452,8 @@ public class DirCacheCheckout {
 			for (String path : updated.keySet()) {
 				DirCacheEntry entry = dc.getEntry(path);
 				if (!FileMode.GITLINK.equals(entry.getRawMode()))
-					checkoutEntry(repo, entry, objectReader, false);
+					checkoutEntry(repo, entry, objectReader, false,
+							updated.get(path));
 			}
 
 			// commit the index builder - a new index is persisted
@@ -996,9 +1002,12 @@ public class DirCacheCheckout {
 		removed.add(path);
 	}
 
-	private void update(String path, ObjectId mId, FileMode mode) {
+	private void update(String path, ObjectId mId, FileMode mode)
+			throws IOException {
 		if (!FileMode.TREE.equals(mode)) {
-			updated.put(path, mId);
+			updated.put(path,
+					walk.getFilterCommand(Constants.ATTR_FILTER_TYPE_SMUDGE));
+
 			DirCacheEntry entry = new DirCacheEntry(path, DirCacheEntry.STAGE_0);
 			entry.setObjectId(mId);
 			entry.setFileMode(mode);
@@ -1150,7 +1159,7 @@ public class DirCacheCheckout {
 	 */
 	public static void checkoutEntry(Repository repo, DirCacheEntry entry,
 			ObjectReader or) throws IOException {
-		checkoutEntry(repo, entry, or, false);
+		checkoutEntry(repo, entry, or, false, null);
 	}
 
 	/**
@@ -1186,6 +1195,46 @@ public class DirCacheCheckout {
 	 */
 	public static void checkoutEntry(Repository repo, DirCacheEntry entry,
 			ObjectReader or, boolean deleteRecursive) throws IOException {
+		checkoutEntry(repo, entry, or, deleteRecursive, null);
+	}
+
+	/**
+	 * Updates the file in the working tree with content and mode from an entry
+	 * in the index. The new content is first written to a new temporary file in
+	 * the same directory as the real file. Then that new file is renamed to the
+	 * final filename.
+	 *
+	 * <p>
+	 * <b>Note:</b> if the entry path on local file system exists as a file, it
+	 * will be deleted and if it exists as a directory, it will be deleted
+	 * recursively, independently if has any content.
+	 * </p>
+	 *
+	 * <p>
+	 * TODO: this method works directly on File IO, we may need another
+	 * abstraction (like WorkingTreeIterator). This way we could tell e.g.
+	 * Eclipse that Files in the workspace got changed
+	 * </p>
+	 *
+	 * @param repo
+	 *            repository managing the destination work tree.
+	 * @param entry
+	 *            the entry containing new mode and content
+	 * @param or
+	 *            object reader to use for checkout
+	 * @param deleteRecursive
+	 *            true to recursively delete final path if it exists on the file
+	 *            system
+	 * @param smudgeFilterCommand
+	 *            the filter command to be run for smudging the entry to be
+	 *            checked out
+	 *
+	 * @throws IOException
+	 * @since 4.2
+	 */
+	public static void checkoutEntry(Repository repo, DirCacheEntry entry,
+			ObjectReader or, boolean deleteRecursive,
+			String smudgeFilterCommand) throws IOException {
 		ObjectLoader ol = or.open(entry.getObjectId());
 		File f = new File(repo.getWorkTree(), entry.getPathString());
 		File parentDir = f.getParentFile();
@@ -1210,14 +1259,52 @@ public class DirCacheCheckout {
 		OutputStream channel = new FileOutputStream(tmpFile);
 		if (opt.getAutoCRLF() == AutoCRLF.TRUE)
 			channel = new AutoCRLFOutputStream(channel);
-		try {
-			ol.copyTo(channel);
-		} finally {
-			channel.close();
+		if (smudgeFilterCommand != null) {
+			ProcessBuilder filterProcessBuilder = fs
+					.runInShell(smudgeFilterCommand, new String[0]);
+			filterProcessBuilder.directory(repo.getWorkTree());
+			filterProcessBuilder.environment().put(Constants.GIT_DIR_KEY,
+					repo.getDirectory().getAbsolutePath());
+			ExecutionResult result;
+			int rc;
+			try {
+				// TODO: wire correctly with AUTOCRLF
+				result = fs.execute(filterProcessBuilder, ol.openStream());
+				rc = result.getRc();
+				if (rc == 0) {
+					result.getStdout().writeTo(channel,
+							NullProgressMonitor.INSTANCE);
+				}
+			} catch (IOException | InterruptedException e) {
+				throw new IOException(new FilterFailedException(e,
+						smudgeFilterCommand, entry.getPathString()));
+
+			} finally {
+				channel.close();
+			}
+			if (rc != 0) {
+				throw new IOException(new FilterFailedException(rc,
+						smudgeFilterCommand, entry.getPathString(),
+						result.getStdout().toByteArray(MAX_EXCEPTION_TEXT_SIZE),
+						RawParseUtils.decode(result.getStderr()
+								.toByteArray(MAX_EXCEPTION_TEXT_SIZE))));
+			}
+		} else {
+			try {
+				ol.copyTo(channel);
+			} finally {
+				channel.close();
+			}
+		}
+		// The entry needs to correspond to the on-disk filesize. If the content
+		// was filtered (either by autocrlf handling or smudge filters) ask the
+		// filesystem again for the length. Otherwise the objectloader knows the
+		// size
+		if (opt.getAutoCRLF() == AutoCRLF.TRUE || smudgeFilterCommand != null) {
+			entry.setLength(tmpFile.length());
+		} else {
+			entry.setLength(ol.getSize());
 		}
-		entry.setLength(opt.getAutoCRLF() == AutoCRLF.TRUE ? //
-				tmpFile.length() // AutoCRLF wants on-disk-size
-				: (int) ol.getSize());
 
 		if (opt.isFileMode() && fs.supportsExecute()) {
 			if (FileMode.EXECUTABLE_FILE.equals(entry.getRawMode())) {
diff --git a/org.eclipse.jgit/src/org/eclipse/jgit/lib/Constants.java b/org.eclipse.jgit/src/org/eclipse/jgit/lib/Constants.java
index 1a3111ab49..d30edaf41b 100644
--- a/org.eclipse.jgit/src/org/eclipse/jgit/lib/Constants.java
+++ b/org.eclipse.jgit/src/org/eclipse/jgit/lib/Constants.java
@@ -384,6 +384,13 @@ public final class Constants {
 	 */
 	public static final String ATTR_FILTER_TYPE_CLEAN = "clean";
 
+	/**
+	 * smudge command name, used to call filter driver
+	 *
+	 * @since 4.2
+	 */
+	public static final String ATTR_FILTER_TYPE_SMUDGE = "smudge";
+
 	/** Name of the ignore file */
 	public static final String DOT_GIT_IGNORE = ".gitignore";