#include "rsync.h"
-extern int csum_length;
-
extern int verbose;
extern int am_server;
-
-extern int remote_version;
+extern int do_progress;
typedef unsigned short tag;
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);
}
static OFF_T last_match;
+/**
+ * Transmit a literal and/or match token.
+ *
+ * This delightfully-named function is called either when we find a
+ * match and need to transmit all the unmatched data leading up to it,
+ * or when we get bored of accumulating literal data and just need to
+ * transmit it. As a result of this second case, it is called even if
+ * we have not matched at all!
+ *
+ * @param i If >0, the number of a matched token. If 0, indicates we
+ * have only literal data.
+ **/
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;
else
last_match = offset;
- if (buf)
- show_progress(last_match, buf->size);
+ if (buf && do_progress) {
+ show_progress(last_match, buf->file_size);
- if (i == -1) end_progress();
+ 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=%ld len=%.0f\n",
+ (long) s->blength, (double)len);
- k = MIN(len, s->n);
+ /* cast is to make s->blength signed; it should always be reasonably
+ * small */
+ k = MIN(len, (OFF_T) s->blength);
map = (schar *)map_ptr(buf,0,k);
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->blength=%ld len=%.0f count=%ld\n",
+ (long) s->blength, (double) len, (long) s->count);
do {
tag t = gettag2(s1,s2);
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;
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->blength,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;
}
- if (memcmp(sum2,s->sums[i].sum2,csum_length) != 0) {
+ if (memcmp(sum2,s->sums[i].sum2,s->s2length) != 0) {
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,s->s2length) != 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;
- k = MIN((len-offset), s->n);
+ k = MIN((len-offset), s->blength);
map = (schar *)map_ptr(buf,offset,k);
sum = get_checksum1((char *)map, k);
s1 = sum & 0xFFFF;
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->blength &&
(end-offset > CHUNK_SIZE)) {
- matched(f,s,buf,offset - s->n, -2);
+ matched(f,s,buf,offset - s->blength, -2);
}
} while (++offset < end);
}
-void match_sums(int f,struct sum_struct *s,struct map_struct *buf,OFF_T len)
+/**
+ * Scan through a origin file, looking for sections that match
+ * checksums from the generator, and transmit either literal or token
+ * data.
+ *
+ * Also calculates the MD4 checksum of the whole file, using the md
+ * accumulator. This is transmitted with the file as protection
+ * against corruption on the wire.
+ *
+ * @param s Checksums received from the generator. If <tt>s->count ==
+ * 0</tt>, then there are actually no checksums for this file.
+ *
+ * @param len Length of the file to send.
+ **/
+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;
sum_end(file_sum);
- if (remote_version >= 14) {
- if (verbose > 2)
- rprintf(FINFO,"sending file_sum\n");
- write_buf(f,file_sum,MD4_SUM_LENGTH);
- }
+ 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) {
free(targets);
Matt McCutchen's Web Site / rsync / rsync.git / blobdiff