From a8b29afd82795b3d98b42bf214fea27ab61984cc Mon Sep 17 00:00:00 2001 From: Jeff Schumacher Date: Thu, 8 Jul 2010 09:59:36 -0700 Subject: [PATCH] Create FileHeader from DiffEntry Added support for converting DiffEntrys to FileHeaders. FileHeaders are DiffEntrys with a buffer containing the diff output as well as a list of HunkHeaders. The HunkHeaders contain EditLists. The createFileHeader(DiffEntry) method in DiffFormatter performs a Myers Diff on the files refered to by the DiffEntry, then puts the returned EditList into a single HunkHeader, which is then put into the FileHeader to be returned. It also generates the appropriate diff header an puts it into the FileHeader's buffer. The rest of the diff output, which would normally be parsed to generate the HunkHeaders, is not generated. In fact, the purpose of this method is to avoid the costly diff output generation and parsing normally required to create a FileHeader. Change-Id: I7d8b18c0f6c85e3d02ad58995d3d231e69af5887 --- .../eclipse/jgit/diff/DiffFormatterTest.java | 186 ++++++++++++++ .../src/org/eclipse/jgit/diff/DiffEntry.java | 7 + .../org/eclipse/jgit/diff/DiffFormatter.java | 226 ++++++++++++------ .../org/eclipse/jgit/patch/FileHeader.java | 19 ++ .../org/eclipse/jgit/patch/HunkHeader.java | 93 ++++--- .../src/org/eclipse/jgit/patch/Patch.java | 2 +- 6 files changed, 416 insertions(+), 117 deletions(-) create mode 100644 org.eclipse.jgit.test/tst/org/eclipse/jgit/diff/DiffFormatterTest.java diff --git a/org.eclipse.jgit.test/tst/org/eclipse/jgit/diff/DiffFormatterTest.java b/org.eclipse.jgit.test/tst/org/eclipse/jgit/diff/DiffFormatterTest.java new file mode 100644 index 000000000..996ee35a1 --- /dev/null +++ b/org.eclipse.jgit.test/tst/org/eclipse/jgit/diff/DiffFormatterTest.java @@ -0,0 +1,186 @@ +/* + * Copyright (C) 2010, Google Inc. + * and other copyright owners as documented in the project's IP log. + * + * This program and the accompanying materials are made available + * under the terms of the Eclipse Distribution License v1.0 which + * accompanies this distribution, is reproduced below, and is + * available at http://www.eclipse.org/org/documents/edl-v10.php + * + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials provided + * with the distribution. + * + * - Neither the name of the Eclipse Foundation, Inc. nor the + * names of its contributors may be used to endorse or promote + * products derived from this software without specific prior + * written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND + * CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, + * INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES + * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR + * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT + * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER + * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + * STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF + * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +package org.eclipse.jgit.diff; + +import org.eclipse.jgit.diff.DiffEntry.ChangeType; +import org.eclipse.jgit.junit.TestRepository; +import org.eclipse.jgit.lib.FileMode; +import org.eclipse.jgit.lib.ObjectId; +import org.eclipse.jgit.lib.RepositoryTestCase; +import org.eclipse.jgit.patch.FileHeader; +import org.eclipse.jgit.patch.HunkHeader; +import org.eclipse.jgit.util.RawParseUtils; +import org.eclipse.jgit.util.io.DisabledOutputStream; + +public class DiffFormatterTest extends RepositoryTestCase { + private static final String DIFF = "diff --git "; + + private static final String REGULAR_FILE = "100644"; + + private static final String GITLINK = "160000"; + + private static final String PATH_A = "src/a"; + + private static final String PATH_B = "src/b"; + + private DiffFormatter df; + + private TestRepository testDb; + + @Override + public void setUp() throws Exception { + super.setUp(); + testDb = new TestRepository(db); + df = new DiffFormatter(DisabledOutputStream.INSTANCE); + df.setRepository(db); + } + + public void testCreateFileHeader_Modify() throws Exception { + ObjectId adId = blob("a\nd\n"); + ObjectId abcdId = blob("a\nb\nc\nd\n"); + + String diffHeader = makeDiffHeader(PATH_A, PATH_A, adId, abcdId); + + DiffEntry ad = DiffEntry.delete(PATH_A, adId); + DiffEntry abcd = DiffEntry.add(PATH_A, abcdId); + + DiffEntry mod = DiffEntry.pair(ChangeType.MODIFY, ad, abcd, 0); + + FileHeader fh = df.createFileHeader(mod); + + assertEquals(diffHeader, RawParseUtils.decode(fh.getBuffer())); + assertEquals(0, fh.getStartOffset()); + assertEquals(fh.getBuffer().length, fh.getEndOffset()); + assertEquals(FileHeader.PatchType.UNIFIED, fh.getPatchType()); + + assertEquals(1, fh.getHunks().size()); + + HunkHeader hh = fh.getHunks().get(0); + assertEquals(1, hh.toEditList().size()); + + EditList el = hh.toEditList(); + assertEquals(1, el.size()); + + Edit e = el.get(0); + assertEquals(1, e.getBeginA()); + assertEquals(1, e.getEndA()); + assertEquals(1, e.getBeginB()); + assertEquals(3, e.getEndB()); + assertEquals(Edit.Type.INSERT, e.getType()); + } + + public void testCreateFileHeader_Binary() throws Exception { + ObjectId adId = blob("a\nd\n"); + ObjectId binId = blob("a\nb\nc\n\0\0\0\0d\n"); + + String diffHeader = makeDiffHeader(PATH_A, PATH_B, adId, binId) + + "Binary files differ\n"; + + DiffEntry ad = DiffEntry.delete(PATH_A, adId); + DiffEntry abcd = DiffEntry.add(PATH_B, binId); + + DiffEntry mod = DiffEntry.pair(ChangeType.MODIFY, ad, abcd, 0); + + FileHeader fh = df.createFileHeader(mod); + + assertEquals(diffHeader, RawParseUtils.decode(fh.getBuffer())); + assertEquals(FileHeader.PatchType.BINARY, fh.getPatchType()); + + assertEquals(1, fh.getHunks().size()); + + HunkHeader hh = fh.getHunks().get(0); + assertEquals(0, hh.toEditList().size()); + } + + public void testCreateFileHeader_GitLink() throws Exception { + ObjectId aId = blob("a\n"); + ObjectId bId = blob("b\n"); + + String diffHeader = makeDiffHeaderModeChange(PATH_A, PATH_A, aId, bId, + GITLINK, REGULAR_FILE) + + "-Subproject commit " + aId.name() + "\n"; + + DiffEntry ad = DiffEntry.delete(PATH_A, aId); + ad.oldMode = FileMode.GITLINK; + DiffEntry abcd = DiffEntry.add(PATH_A, bId); + + DiffEntry mod = DiffEntry.pair(ChangeType.MODIFY, ad, abcd, 0); + + FileHeader fh = df.createFileHeader(mod); + + assertEquals(diffHeader, RawParseUtils.decode(fh.getBuffer())); + + assertEquals(1, fh.getHunks().size()); + + HunkHeader hh = fh.getHunks().get(0); + assertEquals(0, hh.toEditList().size()); + } + + private String makeDiffHeader(String pathA, String pathB, ObjectId aId, + ObjectId bId) { + String a = aId.abbreviate(db).name(); + String b = bId.abbreviate(db).name(); + return DIFF + "a/" + pathA + " " + "b/" + pathB + "\n" + // + "index " + a + ".." + b + " " + REGULAR_FILE + "\n" + // + "--- a/" + pathA + "\n" + // + "+++ b/" + pathB + "\n"; + } + + private String makeDiffHeaderModeChange(String pathA, String pathB, + ObjectId aId, ObjectId bId, String modeA, String modeB) { + String a = aId.abbreviate(db).name(); + String b = bId.abbreviate(db).name(); + return DIFF + "a/" + pathA + " " + "b/" + pathB + "\n" + // + "old mode " + modeA + "\n" + // + "new mode " + modeB + "\n" + // + "index " + a + ".." + b + "\n" + // + "--- a/" + pathA + "\n" + // + "+++ b/" + pathB + "\n"; + } + + private ObjectId blob(String content) throws Exception { + return testDb.blob(content).copy(); + } + +} diff --git a/org.eclipse.jgit/src/org/eclipse/jgit/diff/DiffEntry.java b/org.eclipse.jgit/src/org/eclipse/jgit/diff/DiffEntry.java index e41ec5670..4d25dbd02 100644 --- a/org.eclipse.jgit/src/org/eclipse/jgit/diff/DiffEntry.java +++ b/org.eclipse.jgit/src/org/eclipse/jgit/diff/DiffEntry.java @@ -81,6 +81,13 @@ public static enum ChangeType { COPY; } + /** + * Create an empty DiffEntry + */ + protected DiffEntry(){ + // reduce the visibility of the default constructor + } + /** * Convert the TreeWalk into DiffEntry headers. * 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 c12c54b05..cdcc5e63e 100644 --- a/org.eclipse.jgit/src/org/eclipse/jgit/diff/DiffFormatter.java +++ b/org.eclipse.jgit/src/org/eclipse/jgit/diff/DiffFormatter.java @@ -48,18 +48,24 @@ import static org.eclipse.jgit.lib.Constants.encodeASCII; import static org.eclipse.jgit.lib.FileMode.GITLINK; +import java.io.ByteArrayOutputStream; import java.io.IOException; import java.io.OutputStream; import java.util.List; import org.eclipse.jgit.JGitText; +import org.eclipse.jgit.errors.CorruptObjectException; +import org.eclipse.jgit.errors.MissingObjectException; import org.eclipse.jgit.lib.AbbreviatedObjectId; import org.eclipse.jgit.lib.Constants; import org.eclipse.jgit.lib.FileMode; import org.eclipse.jgit.lib.ObjectLoader; import org.eclipse.jgit.lib.Repository; import org.eclipse.jgit.patch.FileHeader; +import org.eclipse.jgit.patch.HunkHeader; +import org.eclipse.jgit.patch.FileHeader.PatchType; import org.eclipse.jgit.util.QuotedString; +import org.eclipse.jgit.util.io.DisabledOutputStream; /** * Format an {@link EditList} as a Git style unified patch script. @@ -185,87 +191,10 @@ public void format(List entries) throws IOException { * be written to. */ public void format(DiffEntry ent) throws IOException { - String oldName = quotePath("a/" + ent.getOldName()); - String newName = quotePath("b/" + ent.getNewName()); - out.write(encode("diff --git " + oldName + " " + newName + "\n")); - - switch (ent.getChangeType()) { - case ADD: - out.write(encodeASCII("new file mode ")); - ent.getNewMode().copyTo(out); - out.write('\n'); - break; - - case DELETE: - out.write(encodeASCII("deleted file mode ")); - ent.getOldMode().copyTo(out); - out.write('\n'); - break; - - case RENAME: - out.write(encodeASCII("similarity index " + ent.getScore() + "%")); - out.write('\n'); - - out.write(encode("rename from " + quotePath(ent.getOldName()))); - out.write('\n'); - - out.write(encode("rename to " + quotePath(ent.getNewName()))); - out.write('\n'); - break; - - case COPY: - out.write(encodeASCII("similarity index " + ent.getScore() + "%")); - out.write('\n'); - - out.write(encode("copy from " + quotePath(ent.getOldName()))); - out.write('\n'); - - out.write(encode("copy to " + quotePath(ent.getNewName()))); - out.write('\n'); - - if (!ent.getOldMode().equals(ent.getNewMode())) { - out.write(encodeASCII("new file mode ")); - ent.getNewMode().copyTo(out); - out.write('\n'); - } - break; - } - - switch (ent.getChangeType()) { - case RENAME: - case MODIFY: - if (!ent.getOldMode().equals(ent.getNewMode())) { - out.write(encodeASCII("old mode ")); - ent.getOldMode().copyTo(out); - out.write('\n'); - - out.write(encodeASCII("new mode ")); - ent.getNewMode().copyTo(out); - out.write('\n'); - } - } - - out.write(encodeASCII("index " // - + format(ent.getOldId()) // - + ".." // - + format(ent.getNewId()))); - if (ent.getOldMode().equals(ent.getNewMode())) { - out.write(' '); - ent.getNewMode().copyTo(out); - } - out.write('\n'); - out.write(encode("--- " + oldName + '\n')); - out.write(encode("+++ " + newName + '\n')); + writeDiffHeader(out, ent); if (ent.getOldMode() == GITLINK || ent.getNewMode() == GITLINK) { - if (ent.getOldMode() == GITLINK) { - out.write(encodeASCII("-Subproject commit " - + ent.getOldId().name() + "\n")); - } - if (ent.getNewMode() == GITLINK) { - out.write(encodeASCII("+Subproject commit " - + ent.getNewId().name() + "\n")); - } + writeGitLinkDiffText(out, ent); } else { byte[] aRaw = open(ent.getOldMode(), ent.getOldId()); byte[] bRaw = open(ent.getNewMode(), ent.getNewId()); @@ -281,6 +210,93 @@ public void format(DiffEntry ent) throws IOException { } } + private void writeGitLinkDiffText(OutputStream o, DiffEntry ent) + throws IOException { + if (ent.getOldMode() == GITLINK) { + o.write(encodeASCII("-Subproject commit " + ent.getOldId().name() + + "\n")); + } + if (ent.getNewMode() == GITLINK) { + o.write(encodeASCII("+Subproject commit " + ent.getNewId().name() + + "\n")); + } + } + + private void writeDiffHeader(OutputStream o, DiffEntry ent) + throws IOException { + String oldName = quotePath("a/" + ent.getOldName()); + String newName = quotePath("b/" + ent.getNewName()); + o.write(encode("diff --git " + oldName + " " + newName + "\n")); + + switch (ent.getChangeType()) { + case ADD: + o.write(encodeASCII("new file mode ")); + ent.getNewMode().copyTo(o); + o.write('\n'); + break; + + case DELETE: + o.write(encodeASCII("deleted file mode ")); + ent.getOldMode().copyTo(o); + o.write('\n'); + break; + + case RENAME: + o.write(encodeASCII("similarity index " + ent.getScore() + "%")); + o.write('\n'); + + o.write(encode("rename from " + quotePath(ent.getOldName()))); + o.write('\n'); + + o.write(encode("rename to " + quotePath(ent.getNewName()))); + o.write('\n'); + break; + + case COPY: + o.write(encodeASCII("similarity index " + ent.getScore() + "%")); + o.write('\n'); + + o.write(encode("copy from " + quotePath(ent.getOldName()))); + o.write('\n'); + + o.write(encode("copy to " + quotePath(ent.getNewName()))); + o.write('\n'); + + if (!ent.getOldMode().equals(ent.getNewMode())) { + o.write(encodeASCII("new file mode ")); + ent.getNewMode().copyTo(o); + o.write('\n'); + } + break; + } + + switch (ent.getChangeType()) { + case RENAME: + case MODIFY: + if (!ent.getOldMode().equals(ent.getNewMode())) { + o.write(encodeASCII("old mode ")); + ent.getOldMode().copyTo(o); + o.write('\n'); + + o.write(encodeASCII("new mode ")); + ent.getNewMode().copyTo(o); + o.write('\n'); + } + } + + o.write(encodeASCII("index " // + + format(ent.getOldId()) // + + ".." // + + format(ent.getNewId()))); + if (ent.getOldMode().equals(ent.getNewMode())) { + o.write(' '); + ent.getNewMode().copyTo(o); + } + o.write('\n'); + o.write(encode("--- " + oldName + '\n')); + o.write(encode("+++ " + newName + '\n')); + } + private String format(AbbreviatedObjectId oldId) { if (oldId.isComplete() && db != null) oldId = oldId.toObjectId().abbreviate(db, abbreviationLength); @@ -513,6 +529,58 @@ protected void writeLine(final char prefix, final RawText text, out.write('\n'); } + /** + * Creates a {@link FileHeader} representing the given {@link DiffEntry} + *

+ * This method does not use the OutputStream associated with this + * DiffFormatter instance. It is therefore safe to instantiate this + * DiffFormatter instance with a {@link DisabledOutputStream} if this method + * is the only one that will be used. + * + * @param ent + * the DiffEntry to create the FileHeader for + * @return a FileHeader representing the DiffEntry. The FileHeader's buffer + * will contain only the header of the diff output. It will also + * contain one {@link HunkHeader}. + * @throws IOException + * the stream threw an exception while writing to it, or one of + * the blobs referenced by the DiffEntry could not be read. + * @throws CorruptObjectException + * one of the blobs referenced by the DiffEntry is corrupt. + * @throws MissingObjectException + * one of the blobs referenced by the DiffEntry is missing. + */ + public FileHeader createFileHeader(DiffEntry ent) throws IOException, + CorruptObjectException, MissingObjectException { + ByteArrayOutputStream buf = new ByteArrayOutputStream(); + final EditList editList; + final FileHeader.PatchType type; + + writeDiffHeader(buf, ent); + + if (ent.getOldMode() == GITLINK || ent.getNewMode() == GITLINK) { + writeGitLinkDiffText(buf, ent); + editList = new EditList(); + type = PatchType.UNIFIED; + } else { + byte[] aRaw = open(ent.getOldMode(), ent.getOldId()); + byte[] bRaw = open(ent.getNewMode(), ent.getNewId()); + + if (RawText.isBinary(aRaw) || RawText.isBinary(bRaw)) { + buf.write(encodeASCII("Binary files differ\n")); + editList = new EditList(); + type = PatchType.BINARY; + } else { + RawText a = rawTextFactory.create(aRaw); + RawText b = rawTextFactory.create(bRaw); + editList = new MyersDiff(a, b).getEdits(); + type = PatchType.UNIFIED; + } + } + + return new FileHeader(buf.toByteArray(), editList, type); + } + private int findCombinedEnd(final List edits, final int i) { int end = i + 1; while (end < edits.size() diff --git a/org.eclipse.jgit/src/org/eclipse/jgit/patch/FileHeader.java b/org.eclipse.jgit/src/org/eclipse/jgit/patch/FileHeader.java index f2f2f2e74..0c24fc6be 100644 --- a/org.eclipse.jgit/src/org/eclipse/jgit/patch/FileHeader.java +++ b/org.eclipse.jgit/src/org/eclipse/jgit/patch/FileHeader.java @@ -134,6 +134,25 @@ public static enum PatchType { /** If {@link #patchType} is {@link PatchType#GIT_BINARY}, the old image */ BinaryHunk reverseBinaryHunk; + /** + * Constructs a new FileHeader + * + * @param headerLines + * buffer holding the diff header for this file + * @param edits + * the edits for this file + * @param type + * the type of patch used to modify this file + */ + public FileHeader(final byte[] headerLines, EditList edits, PatchType type) { + this(headerLines, 0); + endOffset = headerLines.length; + int ptr = parseGitFileName(Patch.DIFF_GIT.length, headerLines.length); + ptr = parseGitHeaders(ptr, headerLines.length); + this.patchType = type; + addHunk(new HunkHeader(this, edits)); + } + FileHeader(final byte[] b, final int offset) { buf = b; startOffset = offset; diff --git a/org.eclipse.jgit/src/org/eclipse/jgit/patch/HunkHeader.java b/org.eclipse.jgit/src/org/eclipse/jgit/patch/HunkHeader.java index bfb20b64e..a6ea74eb1 100644 --- a/org.eclipse.jgit/src/org/eclipse/jgit/patch/HunkHeader.java +++ b/org.eclipse.jgit/src/org/eclipse/jgit/patch/HunkHeader.java @@ -116,6 +116,8 @@ public int getLinesAdded() { /** Total number of lines of context appearing in this hunk */ int nContext; + private EditList editList; + HunkHeader(final FileHeader fh, final int offset) { this(fh, offset, new OldImage() { @Override @@ -131,6 +133,21 @@ public AbbreviatedObjectId getId() { old = oi; } + HunkHeader(final FileHeader fh, final EditList editList) { + this(fh, fh.buf.length); + this.editList = editList; + endOffset = startOffset; + nContext = 0; + if (editList.isEmpty()) { + newStartLine = 0; + newLineCount = 0; + } else { + newStartLine = editList.get(0).getBeginB(); + Edit last = editList.get(editList.size() - 1); + newLineCount = last.getEndB() - newStartLine; + } + } + /** @return header for the file this hunk applies to */ public FileHeader getFileHeader() { return file; @@ -173,48 +190,50 @@ public int getLinesContext() { /** @return a list describing the content edits performed within the hunk. */ public EditList toEditList() { - final EditList r = new EditList(); - final byte[] buf = file.buf; - int c = nextLF(buf, startOffset); - int oLine = old.startLine; - int nLine = newStartLine; - Edit in = null; + if (editList == null) { + editList = new EditList(); + final byte[] buf = file.buf; + int c = nextLF(buf, startOffset); + int oLine = old.startLine; + int nLine = newStartLine; + Edit in = null; - SCAN: for (; c < endOffset; c = nextLF(buf, c)) { - switch (buf[c]) { - case ' ': - case '\n': - in = null; - oLine++; - nLine++; - continue; + SCAN: for (; c < endOffset; c = nextLF(buf, c)) { + switch (buf[c]) { + case ' ': + case '\n': + in = null; + oLine++; + nLine++; + continue; - case '-': - if (in == null) { - in = new Edit(oLine - 1, nLine - 1); - r.add(in); + case '-': + if (in == null) { + in = new Edit(oLine - 1, nLine - 1); + editList.add(in); + } + oLine++; + in.extendA(); + continue; + + case '+': + if (in == null) { + in = new Edit(oLine - 1, nLine - 1); + editList.add(in); + } + nLine++; + in.extendB(); + continue; + + case '\\': // Matches "\ No newline at end of file" + continue; + + default: + break SCAN; } - oLine++; - in.extendA(); - continue; - - case '+': - if (in == null) { - in = new Edit(oLine - 1, nLine - 1); - r.add(in); - } - nLine++; - in.extendB(); - continue; - - case '\\': // Matches "\ No newline at end of file" - continue; - - default: - break SCAN; } } - return r; + return editList; } void parseHeader() { diff --git a/org.eclipse.jgit/src/org/eclipse/jgit/patch/Patch.java b/org.eclipse.jgit/src/org/eclipse/jgit/patch/Patch.java index ce006dadb..cf42b665a 100644 --- a/org.eclipse.jgit/src/org/eclipse/jgit/patch/Patch.java +++ b/org.eclipse.jgit/src/org/eclipse/jgit/patch/Patch.java @@ -60,7 +60,7 @@ /** A parsed collection of {@link FileHeader}s from a unified diff patch file */ public class Patch { - private static final byte[] DIFF_GIT = encodeASCII("diff --git "); + static final byte[] DIFF_GIT = encodeASCII("diff --git "); private static final byte[] DIFF_CC = encodeASCII("diff --cc ");