/* Copyright 2010 Dmitry Naumenko (dm.naumenko@gmail.com) Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License. */ package difflib; import difflib.myers.MyersDiff; import java.util.ArrayList; import java.util.List; import java.util.regex.Matcher; import java.util.regex.Pattern; /** * Implements the difference and patching engine * * @author Dmitry Naumenko * @version 0.4.1 */ public class DiffUtils { private static Pattern unifiedDiffChunkRe = Pattern.compile("^@@\\s+-(?:(\\d+)(?:,(\\d+))?)\\s+\\+(?:(\\d+)(?:,(\\d+))?)\\s+@@$"); /** * Compute the difference between the original and revised texts with default diff algorithm * * @param original the original text * @param revised the revised text * @return the patch describing the difference between the original and revised texts */ public static Patch diff(Iterable original, Iterable revised) { return DiffUtils.diff(original, revised, new MyersDiff()); } /** * Compute the difference between the original and revised texts with given diff algorithm * * @param original the original text * @param revised the revised text * @param algorithm the given algorithm * @return the patch describing the difference between the original and revised texts */ public static Patch diff(Iterable original, Iterable revised, DiffAlgorithm algorithm) { return algorithm.diff(original, revised); } /** * Patch the original text with given patch * * @param original the original text * @param patch the given patch * @return the revised text * @throws PatchFailedException if can't apply patch */ public static List patch(List original, Patch patch) throws PatchFailedException { return patch.applyTo(original); } /** * Unpatch the revised text for a given patch * * @param revised the revised text * @param patch the given patch * @return the original text */ public static List unpatch(List revised, Patch patch) { return patch.restore(revised); } /** * Parse the given text in unified format and creates the list of deltas for it. * * @param diff the text in unified format * @return the patch with deltas. */ public static Patch parseUnifiedDiff(List diff) { boolean inPrelude = true; List rawChunk = new ArrayList(); Patch patch = new Patch(); int old_ln = 0, new_ln = 0; String tag; String rest; for (String line : diff) { // Skip leading lines until after we've seen one starting with '+++' if (inPrelude) { if (line.startsWith("+++")) { inPrelude = false; } continue; } Matcher m = unifiedDiffChunkRe.matcher(line); if (m.find()) { // Process the lines in the previous chunk if (rawChunk.size() != 0) { List oldChunkLines = new ArrayList(); List newChunkLines = new ArrayList(); for (Object[] raw_line : rawChunk) { tag = (String) raw_line[0]; rest = (String) raw_line[1]; if (tag.equals(" ") || tag.equals("-")) { oldChunkLines.add(rest); } if (tag.equals(" ") || tag.equals("+")) { newChunkLines.add(rest); } } patch.addDelta(new ChangeDelta(new Chunk(old_ln - 1, oldChunkLines), new Chunk(new_ln - 1, newChunkLines))); rawChunk.clear(); } // Parse the @@ header old_ln = m.group(1) == null ? 1 : Integer.parseInt(m.group(1)); new_ln = m.group(3) == null ? 1 : Integer.parseInt(m.group(3)); if (old_ln == 0) { old_ln += 1; } if (new_ln == 0) { new_ln += 1; } } else { if (line.length() > 0) { tag = line.substring(0, 1); rest = line.substring(1); if (tag.equals(" ") || tag.equals("+") || tag.equals("-")) { rawChunk.add(new Object[]{tag, rest}); } } else { rawChunk.add(new Object[]{" ", ""}); } } } // Process the lines in the last chunk if (rawChunk.size() != 0) { List oldChunkLines = new ArrayList(); List newChunkLines = new ArrayList(); for (Object[] raw_line : rawChunk) { tag = (String) raw_line[0]; rest = (String) raw_line[1]; if (tag.equals(" ") || tag.equals("-")) { oldChunkLines.add(rest); } if (tag.equals(" ") || tag.equals("+")) { newChunkLines.add(rest); } } patch.addDelta(new ChangeDelta(new Chunk(old_ln - 1, oldChunkLines), new Chunk( new_ln - 1, newChunkLines))); rawChunk.clear(); } return patch; } /** * generateUnifiedDiff takes a Patch and some other arguments, returning the Unified Diff format text representing the Patch. * * @param original - Filename of the original (unrevised file) * @param revised - Filename of the revised file * @param originalLines - Lines of the original file * @param patch - Patch created by the diff() function * @param contextSize - number of lines of context output around each difference in the file. * @return List of strings representing the Unified Diff representation of the Patch argument. * @author Bill James (tankerbay@gmail.com) */ public static List generateUnifiedDiff(String original, String revised, List originalLines, Patch patch, int contextSize) { if (!patch.getDeltas().isEmpty()) { List ret = new ArrayList(); ret.add("--- " + original); ret.add("+++ " + revised); List> patchDeltas = new ArrayList>(patch.getDeltas()); // code outside the if block also works for single-delta issues. List> deltas = new ArrayList>(); // current list of Delta's to process Delta delta = patchDeltas.get(0); deltas.add(delta); // add the first Delta to the current set // if there's more than 1 Delta, we may need to output them together if (patchDeltas.size() > 1) { for (int i = 1; i < patchDeltas.size(); i++) { int position = delta.getOriginal().getPosition(); // store the current position of // the first Delta // Check if the next Delta is too close to the current position. // And if it is, add it to the current set Delta nextDelta = patchDeltas.get(i); if ((position + delta.getOriginal().size() + contextSize) >= (nextDelta.getOriginal().getPosition() - contextSize)) { deltas.add(nextDelta); } else { // if it isn't, output the current set, // then create a new set and add the current Delta to it. List curBlock = processDeltas(originalLines, deltas, contextSize); ret.addAll(curBlock); deltas.clear(); deltas.add(nextDelta); } delta = nextDelta; } } // don't forget to process the last set of Deltas List curBlock = processDeltas(originalLines, deltas, contextSize); ret.addAll(curBlock); return ret; } return new ArrayList(); } /** * processDeltas takes a list of Deltas and outputs them together in a single block of Unified-Diff-format text. * * @param origLines - the lines of the original file * @param deltas - the Deltas to be output as a single block * @param contextSize - the number of lines of context to place around block * @return * @author Bill James (tankerbay@gmail.com) */ private static List processDeltas(List origLines, List> deltas, int contextSize) { List buffer = new ArrayList(); int origTotal = 0; // counter for total lines output from Original int revTotal = 0; // counter for total lines output from Original int line; Delta curDelta = deltas.get(0); // NOTE: +1 to overcome the 0-offset Position int origStart = curDelta.getOriginal().getPosition() + 1 - contextSize; if (origStart < 1) { origStart = 1; } int revStart = curDelta.getRevised().getPosition() + 1 - contextSize; if (revStart < 1) { revStart = 1; } // find the start of the wrapper context code int contextStart = curDelta.getOriginal().getPosition() - contextSize; if (contextStart < 0) { contextStart = 0; // clamp to the start of the file } // output the context before the first Delta for (line = contextStart; line < curDelta.getOriginal().getPosition(); line++) { // buffer.add(" " + origLines.get(line)); origTotal++; revTotal++; } // output the first Delta buffer.addAll(getDeltaText(curDelta)); origTotal += curDelta.getOriginal().getLines().size(); revTotal += curDelta.getRevised().getLines().size(); int deltaIndex = 1; while (deltaIndex < deltas.size()) { // for each of the other Deltas Delta nextDelta = deltas.get(deltaIndex); int intermediateStart = curDelta.getOriginal().getPosition() + curDelta.getOriginal().getLines().size(); for (line = intermediateStart; line < nextDelta.getOriginal().getPosition(); line++) { // output the code between the last Delta and this one buffer.add(" " + origLines.get(line)); origTotal++; revTotal++; } buffer.addAll(getDeltaText(nextDelta)); // output the Delta origTotal += nextDelta.getOriginal().getLines().size(); revTotal += nextDelta.getRevised().getLines().size(); curDelta = nextDelta; deltaIndex++; } // Now output the post-Delta context code, clamping the end of the file contextStart = curDelta.getOriginal().getPosition() + curDelta.getOriginal().getLines().size(); for (line = contextStart; (line < (contextStart + contextSize)) && (line < origLines.size()); line++) { buffer.add(" " + origLines.get(line)); origTotal++; revTotal++; } // Create and insert the block header, conforming to the Unified Diff standard StringBuffer header = new StringBuffer(); header.append("@@ -"); header.append(origStart); header.append(","); header.append(origTotal); header.append(" +"); header.append(revStart); header.append(","); header.append(revTotal); header.append(" @@"); buffer.add(0, header.toString()); return buffer; } /** * getDeltaText returns the lines to be added to the Unified Diff text from the Delta parameter * * @param delta - the Delta to output * @return list of String lines of code. * @author Bill James (tankerbay@gmail.com) */ private static List getDeltaText(Delta delta) { List buffer = new ArrayList(); for (Object line : delta.getOriginal().getLines()) { buffer.add("-" + line); } for (Object line : delta.getRevised().getLines()) { buffer.add("+" + line); } return buffer; } }