X-Git-Url: https://mattmccutchen.net/rsync/rsync.git/blobdiff_plain/100e5241b067bdc54457f5b6b28d8a1bcdedcf05..fcb69e5cdcfdfd0bdcea2127e5f4c9133a2282b9:/match.c

diff --git a/match.c b/match.c
index 272025aa..ee8e565f 100644
--- a/match.c
+++ b/match.c
@@ -71,7 +71,7 @@ static void build_hash_table(struct sum_struct *s)
   if (!tag_table || !targets) 
     out_of_memory("build_hash_table");
 
-  for (i=0;i<s->count;i++) {
+  for (i=0;i<(int) s->count;i++) {
     targets[i].i = i;
     targets[i].t = gettag(s->sums[i].sum1);
   }
@@ -94,11 +94,11 @@ static void matched(int f,struct sum_struct *s,struct map_struct *buf,
 		    OFF_T offset,int i)
 {
 	OFF_T n = offset - last_match;
-	int j;
+	OFF_T j;
 
 	if (verbose > 2 && i >= 0)
-		rprintf(FINFO,"match at %d last_match=%d j=%d len=%d n=%d\n",
-			(int)offset,(int)last_match,i,(int)s->sums[i].len,(int)n);
+		rprintf(FINFO,"match at %.0f last_match=%.0f j=%d len=%d n=%.0f\n",
+			(double)offset,(double)last_match,i,s->sums[i].len,(double)n);
 
 	send_token(f,i,buf,last_match,n,i<0?0:s->sums[i].len);
 	data_transfer += n;
@@ -118,21 +118,30 @@ static void matched(int f,struct sum_struct *s,struct map_struct *buf,
 		last_match = offset + s->sums[i].len;
 	else
 		last_match = offset;
+
+	if (buf) {
+		show_progress(last_match, buf->file_size);
+
+		if (i == -1) end_progress(buf->file_size);
+	}
 }
 
 
 static void hash_search(int f,struct sum_struct *s,
 			struct map_struct *buf,OFF_T len)
 {
-	OFF_T offset;
-	int j,k;
-	int end;
+	OFF_T offset, end;
+	int j,k, last_i;
 	char sum2[SUM_LENGTH];
 	uint32 s1, s2, sum; 
 	schar *map;
 
+	/* last_i is used to encourage adjacent matches, allowing the RLL coding of the
+	   output to work more efficiently */
+	last_i = -1;
+
 	if (verbose > 2)
-		rprintf(FINFO,"hash search b=%d len=%d\n",s->n,(int)len);
+		rprintf(FINFO,"hash search b=%d len=%.0f\n",s->n,(double)len);
 
 	k = MIN(len, s->n);
 	
@@ -149,8 +158,8 @@ static void hash_search(int f,struct sum_struct *s,
 	end = len + 1 - s->sums[s->count-1].len;
 	
 	if (verbose > 3)
-		rprintf(FINFO,"hash search s->n=%d len=%d count=%d\n",
-			s->n,(int)len,s->count);
+		rprintf(FINFO,"hash search s->n=%d len=%.0f count=%d\n",
+			s->n,(double)len,s->count);
 	
 	do {
 		tag t = gettag2(s1,s2);
@@ -158,7 +167,7 @@ static void hash_search(int f,struct sum_struct *s,
 			
 		j = tag_table[t];
 		if (verbose > 4)
-			rprintf(FINFO,"offset=%d sum=%08x\n",(int)offset,sum);
+			rprintf(FINFO,"offset=%.0f sum=%08x\n",(double)offset,sum);
 		
 		if (j == NULL_TAG) {
 			goto null_tag;
@@ -166,17 +175,20 @@ static void hash_search(int f,struct sum_struct *s,
 
 		sum = (s1 & 0xffff) | (s2 << 16);
 		tag_hits++;
-		for (; j<s->count && targets[j].t == t; j++) {
-			int i = targets[j].i;
+		for (; j < (int) s->count && targets[j].t == t; j++) {
+			int l, i = targets[j].i;
 			
 			if (sum != s->sums[i].sum1) continue;
 			
+			/* also make sure the two blocks are the same length */
+			l = MIN(s->n,len-offset);
+			if (l != s->sums[i].len) continue;			
+
 			if (verbose > 3)
-				rprintf(FINFO,"potential match at %d target=%d %d sum=%08x\n",
-					(int)offset,j,i,sum);
+				rprintf(FINFO,"potential match at %.0f target=%d %d sum=%08x\n",
+					(double)offset,j,i,sum);
 			
 			if (!done_csum2) {
-				int l = MIN(s->n,len-offset);
 				map = (schar *)map_ptr(buf,offset,l);
 				get_checksum2((char *)map,l,sum2);
 				done_csum2 = 1;
@@ -186,6 +198,22 @@ static void hash_search(int f,struct sum_struct *s,
 				false_alarms++;
 				continue;
 			}
+
+			/* we've found a match, but now check to see
+                           if last_i can hint at a better match */
+			for (j++; j < (int) s->count && targets[j].t == t; j++) {
+				int i2 = targets[j].i;
+				if (i2 == last_i + 1) {
+					if (sum != s->sums[i2].sum1) break;
+					if (memcmp(sum2,s->sums[i2].sum2,csum_length) != 0) break;
+					/* we've found an adjacent match - the RLL coder 
+					   will be happy */
+					i = i2;
+					break;
+				}
+			}
+
+			last_i = i;
 			
 			matched(f,s,buf,offset,i);
 			offset += s->sums[i].len - 1;
@@ -218,7 +246,8 @@ static void hash_search(int f,struct sum_struct *s,
 		   match. The 3 reads are caused by the
 		   running match, the checksum update and the
 		   literal send. */
-		if (offset-last_match >= CHUNK_SIZE+s->n && 
+		if (offset > last_match &&
+		    offset-last_match >= CHUNK_SIZE+s->n && 
 		    (end-offset > CHUNK_SIZE)) {
 			matched(f,s,buf,offset - s->n, -2);
 		}
@@ -232,6 +261,7 @@ static void hash_search(int f,struct sum_struct *s,
 void match_sums(int f,struct sum_struct *s,struct map_struct *buf,OFF_T len)
 {
 	char file_sum[MD4_SUM_LENGTH];
+	extern int write_batch;  /*  dw */
 
 	last_match = 0;
 	false_alarms = 0;
@@ -252,6 +282,12 @@ void match_sums(int f,struct sum_struct *s,struct map_struct *buf,OFF_T len)
 		if (verbose > 2) 
 			rprintf(FINFO,"done hash search\n");
 	} else {
+		OFF_T j;
+		/* by doing this in pieces we avoid too many seeks */
+		for (j=0;j<(len-CHUNK_SIZE);j+=CHUNK_SIZE) {
+			int n1 = MIN(CHUNK_SIZE,(len-CHUNK_SIZE)-j);
+			matched(f,s,buf,j+n1,-2);
+		}
 		matched(f,s,buf,len,-1);
 	}
 
@@ -261,6 +297,8 @@ void match_sums(int f,struct sum_struct *s,struct map_struct *buf,OFF_T len)
 		if (verbose > 2)
 			rprintf(FINFO,"sending file_sum\n");
 		write_buf(f,file_sum,MD4_SUM_LENGTH);
+		if (write_batch) /* dw */
+		    write_batch_delta_file(file_sum, MD4_SUM_LENGTH);
 	}
 
 	if (targets) {