Improve performance of line-mode speedup

Reserve 2/3rds of space for text1, 1/3rds for text2.  Prevent overflow beyond hard limit.

Author: Neil Fraser

objectivec/DiffMatchPatch.h

@@ -157,7 +157,6 @@ typedef enum {
 - (NSMutableArray *)diff_computeFromOldString:(NSString *)text1 andNewString:(NSString *)text2 checkLines:(BOOL)checklines deadline:(NSTimeInterval)deadline;
 - (NSMutableArray *)diff_lineModeFromOldString:(NSString *)text1 andNewString:(NSString *)text2 deadline:(NSTimeInterval)deadline;
 - (NSArray *)diff_linesToCharsForFirstString:(NSString *)text1 andSecondString:(NSString *)text1;
-- (NSString *)diff_linesToCharsMungeOfText:(NSString *)text lineArray:(NSMutableArray *)lineArray lineHash:(NSMutableDictionary *)lineHash;
 - (void)diff_chars:(NSArray *)diffs toLines:(NSMutableArray *)lineArray;
 - (NSMutableArray *)diff_bisectOfOldString:(NSString *)text1 andNewString:(NSString *)text2 deadline:(NSTimeInterval)deadline;
 - (NSMutableArray *)diff_bisectSplitOfOldString:(NSString *)text1 andNewString:(NSString *)text2 x:(NSUInteger)x y:(NSUInteger)y deadline:(NSTimeInterval)deadline;

objectivec/DiffMatchPatch.m

@@ -623,23 +623,6 @@ - (NSMutableArray *)diff_lineModeFromOldString:(NSString *)text1
   return diffs;
 }
 
-/**
- * Split a text into a list of strings.  Reduce the texts to a string of
- * hashes where each Unicode character represents one line.
- * @param text NSString to encode.
- * @param lineArray NSMutableArray of unique strings.
- * @param lineHash Map of strings to indices.
- * @return Encoded string.
- */
-- (NSString *)diff_linesToCharsMungeOfText:(NSString *)text
-                                 lineArray:(NSMutableArray *)lineArray
-                                  lineHash:(NSMutableDictionary *)lineHash;
-{
-  return [((NSString *)diff_linesToCharsMungeCFStringCreate((CFStringRef)text,
-                                                            (CFMutableArrayRef)lineArray,
-                                                            (CFMutableDictionaryRef)lineHash)) autorelease];
-}
-
 /**
  * Find the 'middle snake' of a diff, split the problem in two
  * and return the recursively constructed diff.
@@ -859,12 +842,15 @@ - (NSArray *)diff_linesToCharsForFirstString:(NSString *)text1
   // So we'll insert a junk entry to avoid generating a nil character.
   [lineArray addObject:@""];
 
+  // Allocate 2/3rds of the space for text1, the rest for text2.
   NSString *chars1 = (NSString *)diff_linesToCharsMungeCFStringCreate((CFStringRef)text1,
                                                                       (CFMutableArrayRef)lineArray,
-                                                                      (CFMutableDictionaryRef)lineHash);
+                                                                      (CFMutableDictionaryRef)lineHash,
+                                                                      40000);
   NSString *chars2 = (NSString *)diff_linesToCharsMungeCFStringCreate((CFStringRef)text2,
                                                                       (CFMutableArrayRef)lineArray,
-                                                                      (CFMutableDictionaryRef)lineHash);
+                                                                      (CFMutableDictionaryRef)lineHash,
+                                                                      65535);
 
   NSArray *result = [NSArray arrayWithObjects:chars1, chars2, lineArray, nil];
 

objectivec/DiffMatchPatchCFUtilities.c

@@ -450,9 +450,10 @@ CFArrayRef diff_halfMatchICreate(CFStringRef longtext, CFStringRef shorttext, CF
  * @param text CFString to encode.
  * @param lineArray CFMutableArray of unique strings.
  * @param lineHash Map of strings to indices.
+ * @param maxLines Maximum length for lineArray.
  * @return Encoded CFStringRef.
  */
-CFStringRef diff_linesToCharsMungeCFStringCreate(CFStringRef text, CFMutableArrayRef lineArray, CFMutableDictionaryRef lineHash) {
+CFStringRef diff_linesToCharsMungeCFStringCreate(CFStringRef text, CFMutableArrayRef lineArray, CFMutableDictionaryRef lineHash, CFIndex maxLines) {
   #define lineStart lineStartRange.location
   #define lineEnd lineEndRange.location
 
@@ -487,7 +488,7 @@ CFStringRef diff_linesToCharsMungeCFStringCreate(CFStringRef text, CFMutableArra
       const UniChar hashChar = (UniChar)hash;
       CFStringAppendCharacters(chars, &hashChar, 1);
     } else {
-      if (CFArrayGetCount(lineArray) == 65535) {
+      if (CFArrayGetCount(lineArray) == maxLength) {
         // Bail out at 65535 because char 65536 == char 0.
         line = diff_CFStringCreateJavaSubstring(text, lineStart, textLength);
         lineEnd = textLength;

objectivec/DiffMatchPatchCFUtilities.h

@@ -31,7 +31,7 @@ CFIndex diff_commonOverlap(CFStringRef text1, CFStringRef text2);
 CFArrayRef diff_halfMatchCreate(CFStringRef text1, CFStringRef text2, const float diffTimeout);
 CFArrayRef diff_halfMatchICreate(CFStringRef longtext, CFStringRef shorttext, CFIndex i);
 
-CFStringRef diff_linesToCharsMungeCFStringCreate(CFStringRef text, CFMutableArrayRef lineArray, CFMutableDictionaryRef lineHash);
+CFStringRef diff_linesToCharsMungeCFStringCreate(CFStringRef text, CFMutableArrayRef lineArray, CFMutableDictionaryRef lineHash, CFIndex maxLines);
 
 CFIndex diff_cleanupSemanticScore(CFStringRef one, CFStringRef two);