My modified version of Chris Shoemaker's improved batch-file handling.

[rsync/rsync.git] / match.c

diff --git a/match.c b/match.c
index 6909fca..d731aae 100644 (file)
--- a/match.c
+++ b/match.c
@@ -22,6 +22,7 @@
 extern int verbose;
 extern int am_server;
 extern int do_progress;
+extern int checksum_seed;
 
 typedef unsigned short tag;
 
@@ -141,14 +142,14 @@ static void hash_search(int f,struct sum_struct *s,
 {
        OFF_T offset, end;
        unsigned int k;
-       size_t last_i;
+       size_t want_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 = (size_t)-1;
+       /* want_i is used to encourage adjacent matches, allowing the RLL
+        * coding of the output to work more efficiently. */
+       want_i = 0;
 
        if (verbose > 2) {
                rprintf(FINFO,"hash search b=%u len=%.0f\n",
@@ -187,7 +188,7 @@ 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++) {
+               do {
                        unsigned int l;
                        size_t i = targets[j].i;
 
@@ -215,22 +216,15 @@ static void hash_search(int f,struct sum_struct *s,
                        }
 
                        /* we've found a match, but now check to see
-                        * if last_i can hint at a better match */
-                       for (j++; j < s->count && targets[j].t == t; j++) {
-                               size_t i2 = targets[j].i;
-                               if (i2 == last_i + 1) {
-                                       if (sum != s->sums[i2].sum1)
-                                               break;
-                                       if (memcmp(sum2,s->sums[i2].sum2,s->s2length) != 0)
-                                               break;
-                                       /* we've found an adjacent match - the RLL coder
-                                        * will be happy */
-                                       i = i2;
-                                       break;
-                               }
+                        * if want_i can hint at a better match. */
+                       if (i != want_i && want_i < s->count
+                           && sum == s->sums[want_i].sum1
+                           && memcmp(sum2, s->sums[want_i].sum2, s->s2length) == 0) {
+                               /* we've found an adjacent match - the RLL coder
+                                * will be happy */
+                               i = want_i;
                        }
-
-                       last_i = i;
+                       want_i = i + 1;
 
                        matched(f,s,buf,offset,i);
                        offset += s->sums[i].len - 1;
@@ -241,7 +235,7 @@ static void hash_search(int f,struct sum_struct *s,
                        s2 = sum >> 16;
                        matches++;
                        break;
-               }
+               } while (++j < s->count && targets[j].t == t);
 
        null_tag:
                /* Trim off the first byte from the checksum */
@@ -291,7 +285,6 @@ 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;
 
        last_match = 0;
        false_alarms = 0;
@@ -299,7 +292,7 @@ void match_sums(int f, struct sum_struct *s, struct map_struct *buf, OFF_T len)
        matches = 0;
        data_transfer = 0;
 
-       sum_init();
+       sum_init(checksum_seed);
 
        if (len > 0 && s->count>0) {
                build_hash_table(s);
@@ -329,8 +322,6 @@ 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)
-               write_batch_delta_file(file_sum, MD4_SUM_LENGTH);
 
        if (targets) {
                free(targets);