LIB_FILE=libgit.a
LIB_H = \
- blob.h cache.h commit.h count-delta.h csum-file.h delta.h \
+ blob.h cache.h commit.h csum-file.h delta.h \
diff.h object.h pack.h pkt-line.h quote.h refs.h \
run-command.h strbuf.h tag.h tree.h git-compat-util.h revision.h
diffcore-delta.o
LIB_OBJS = \
- blob.o commit.o connect.o count-delta.o csum-file.o \
+ blob.o commit.o connect.o csum-file.o \
date.o diff-delta.o entry.o exec_cmd.o ident.o index.o \
object.o pack-check.o patch-delta.o path.o pkt-line.o \
quote.o read-cache.o refs.o run-command.o \
+++ /dev/null
-/*
- * Copyright (C) 2005 Junio C Hamano
- * The delta-parsing part is almost straight copy of patch-delta.c
- * which is (C) 2005 Nicolas Pitre <nico@cam.org>.
- */
-#include <stdlib.h>
-#include <string.h>
-#include <limits.h>
-#include "delta.h"
-#include "count-delta.h"
-
-/*
- * NOTE. We do not _interpret_ delta fully. As an approximation, we
- * just count the number of bytes that are copied from the source, and
- * the number of literal data bytes that are inserted.
- *
- * Number of bytes that are _not_ copied from the source is deletion,
- * and number of inserted literal bytes are addition, so sum of them
- * is the extent of damage.
- */
-int count_delta(void *delta_buf, unsigned long delta_size,
- unsigned long *src_copied, unsigned long *literal_added)
-{
- unsigned long copied_from_source, added_literal;
- const unsigned char *data, *top;
- unsigned char cmd;
- unsigned long src_size, dst_size, out;
-
- if (delta_size < DELTA_SIZE_MIN)
- return -1;
-
- data = delta_buf;
- top = delta_buf + delta_size;
-
- src_size = get_delta_hdr_size(&data);
- dst_size = get_delta_hdr_size(&data);
-
- added_literal = copied_from_source = out = 0;
- while (data < top) {
- cmd = *data++;
- if (cmd & 0x80) {
- unsigned long cp_off = 0, cp_size = 0;
- if (cmd & 0x01) cp_off = *data++;
- if (cmd & 0x02) cp_off |= (*data++ << 8);
- if (cmd & 0x04) cp_off |= (*data++ << 16);
- if (cmd & 0x08) cp_off |= (*data++ << 24);
- if (cmd & 0x10) cp_size = *data++;
- if (cmd & 0x20) cp_size |= (*data++ << 8);
- if (cmd & 0x40) cp_size |= (*data++ << 16);
- if (cp_size == 0) cp_size = 0x10000;
-
- copied_from_source += cp_size;
- out += cp_size;
- } else {
- /* write literal into dst */
- added_literal += cmd;
- out += cmd;
- data += cmd;
- }
- }
-
- /* sanity check */
- if (data != top || out != dst_size)
- return -1;
-
- /* delete size is what was _not_ copied from source.
- * edit size is that and literal additions.
- */
- *src_copied = copied_from_source;
- *literal_added = added_literal;
- return 0;
-}
+++ /dev/null
-/*
- * Copyright (C) 2005 Junio C Hamano
- */
-#ifndef COUNT_DELTA_H
-#define COUNT_DELTA_H
-
-int count_delta(void *, unsigned long,
- unsigned long *src_copied, unsigned long *literal_added);
-
-#endif
* The value we return is 1 if we want the pair to be broken,
* or 0 if we do not.
*/
- unsigned long delta_size, base_size, src_copied, literal_added;
- int to_break = 0;
+ unsigned long delta_size, base_size, src_copied, literal_added,
+ src_removed;
*merge_score_p = 0; /* assume no deletion --- "do not break"
* is the default.
&src_copied, &literal_added))
return 0;
+ /* sanity */
+ if (src->size < src_copied)
+ src_copied = src->size;
+ if (dst->size < literal_added + src_copied) {
+ if (src_copied < dst->size)
+ literal_added = dst->size - src_copied;
+ else
+ literal_added = 0;
+ }
+ src_removed = src->size - src_copied;
+
/* Compute merge-score, which is "how much is removed
* from the source material". The clean-up stage will
* merge the surviving pair together if the score is
* less than the minimum, after rename/copy runs.
*/
- if (src->size <= src_copied)
- ; /* all copied, nothing removed */
- else {
- delta_size = src->size - src_copied;
- *merge_score_p = delta_size * MAX_SCORE / src->size;
- }
-
+ *merge_score_p = src_removed * MAX_SCORE / src->size;
+
/* Extent of damage, which counts both inserts and
* deletes.
*/
- if (src->size + literal_added <= src_copied)
- delta_size = 0; /* avoid wrapping around */
- else
- delta_size = (src->size - src_copied) + literal_added;
-
- /* We break if the edit exceeds the minimum.
- * i.e. (break_score / MAX_SCORE < delta_size / base_size)
+ delta_size = src_removed + literal_added;
+ if (delta_size * MAX_SCORE / base_size < break_score)
+ return 0;
+
+ /* If you removed a lot without adding new material, that is
+ * not really a rewrite.
*/
- if (break_score * base_size < delta_size * MAX_SCORE)
- to_break = 1;
+ if ((src->size * break_score < src_removed * MAX_SCORE) &&
+ (literal_added * 20 < src_removed) &&
+ (literal_added * 20 < src_copied))
+ return 0;
- return to_break;
+ return 1;
}
void diffcore_break(int break_score)
#include "diff.h"
#include "diffcore.h"
-struct linehash {
- unsigned long bytes;
- unsigned long hash;
-};
+/*
+ * Idea here is very simple.
+ *
+ * We have total of (sz-N+1) N-byte overlapping sequences in buf whose
+ * size is sz. If the same N-byte sequence appears in both source and
+ * destination, we say the byte that starts that sequence is shared
+ * between them (i.e. copied from source to destination).
+ *
+ * For each possible N-byte sequence, if the source buffer has more
+ * instances of it than the destination buffer, that means the
+ * difference are the number of bytes not copied from source to
+ * destination. If the counts are the same, everything was copied
+ * from source to destination. If the destination has more,
+ * everything was copied, and destination added more.
+ *
+ * We are doing an approximation so we do not really have to waste
+ * memory by actually storing the sequence. We just hash them into
+ * somewhere around 2^16 hashbuckets and count the occurrences.
+ *
+ * The length of the sequence is arbitrarily set to 8 for now.
+ */
-static unsigned long hash_extended_line(const unsigned char **buf_p,
- unsigned long left)
-{
- /* An extended line is zero or more whitespace letters (including LF)
- * followed by one non whitespace letter followed by zero or more
- * non LF, and terminated with by a LF (or EOF).
- */
- const unsigned char *bol = *buf_p;
- const unsigned char *buf = bol;
- unsigned long hashval = 0;
- while (left) {
- unsigned c = *buf++;
- if (!c)
- goto binary;
- left--;
- if (' ' < c) {
- hashval = c;
- break;
- }
- }
- while (left) {
- unsigned c = *buf++;
- if (!c)
- goto binary;
- left--;
- if (c == '\n')
- break;
- if (' ' < c)
- hashval = hashval * 11 + c;
- }
- *buf_p = buf;
- return hashval;
-
- binary:
- *buf_p = NULL;
- return 0;
-}
-
-static int linehash_compare(const void *a_, const void *b_)
-{
- struct linehash *a = (struct linehash *) a_;
- struct linehash *b = (struct linehash *) b_;
- if (a->hash < b->hash) return -1;
- if (a->hash > b->hash) return 1;
- return 0;
-}
+#define HASHBASE 65537 /* next_prime(2^16) */
-static struct linehash *hash_lines(const unsigned char *buf,
- unsigned long size)
+static void hash_chars(unsigned char *buf, unsigned long sz, int *count)
{
- const unsigned char *eobuf = buf + size;
- struct linehash *line = NULL;
- int alloc = 0, used = 0;
+ unsigned int accum1, accum2, i;
- while (buf < eobuf) {
- const unsigned char *ptr = buf;
- unsigned long hash = hash_extended_line(&buf, eobuf-ptr);
- if (!buf) {
- free(line);
- return NULL;
- }
- if (alloc <= used) {
- alloc = alloc_nr(alloc);
- line = xrealloc(line, sizeof(*line) * alloc);
- }
- line[used].bytes = buf - ptr;
- line[used].hash = hash;
- used++;
+ /* an 8-byte shift register made of accum1 and accum2. New
+ * bytes come at LSB of accum2, and shifted up to accum1
+ */
+ for (i = accum1 = accum2 = 0; i < 7; i++, sz--) {
+ accum1 = (accum1 << 8) | (accum2 >> 24);
+ accum2 = (accum2 << 8) | *buf++;
+ }
+ while (sz) {
+ accum1 = (accum1 << 8) | (accum2 >> 24);
+ accum2 = (accum2 << 8) | *buf++;
+ /* We want something that hashes permuted byte
+ * sequences nicely; simpler hash like (accum1 ^
+ * accum2) does not perform as well.
+ */
+ i = (accum1 + accum2 * 0x61) % HASHBASE;
+ count[i]++;
+ sz--;
}
- qsort(line, used, sizeof(*line), linehash_compare);
-
- /* Terminate the list */
- if (alloc <= used)
- line = xrealloc(line, sizeof(*line) * (used+1));
- line[used].bytes = line[used].hash = 0;
- return line;
}
int diffcore_count_changes(void *src, unsigned long src_size,
unsigned long *src_copied,
unsigned long *literal_added)
{
- struct linehash *src_lines, *dst_lines;
+ int *src_count, *dst_count, i;
unsigned long sc, la;
- src_lines = hash_lines(src, src_size);
- if (!src_lines)
- return -1;
- dst_lines = hash_lines(dst, dst_size);
- if (!dst_lines) {
- free(src_lines);
+ if (src_size < 8 || dst_size < 8)
return -1;
- }
+
+ src_count = xcalloc(HASHBASE * 2, sizeof(int));
+ dst_count = src_count + HASHBASE;
+ hash_chars(src, src_size, src_count);
+ hash_chars(dst, dst_size, dst_count);
+
sc = la = 0;
- while (src_lines->bytes && dst_lines->bytes) {
- int cmp = linehash_compare(src_lines, dst_lines);
- if (!cmp) {
- sc += src_lines->bytes;
- src_lines++;
- dst_lines++;
- continue;
+ for (i = 0; i < HASHBASE; i++) {
+ if (src_count[i] < dst_count[i]) {
+ la += dst_count[i] - src_count[i];
+ sc += src_count[i];
}
- if (cmp < 0) {
- src_lines++;
- continue;
- }
- la += dst_lines->bytes;
- dst_lines++;
- }
- while (dst_lines->bytes) {
- la += dst_lines->bytes;
- dst_lines++;
+ else /* i.e. if (dst_count[i] <= src_count[i]) */
+ sc += dst_count[i];
}
*src_copied = sc;
*literal_added = la;
+ free(src_count);
return 0;
}
*/
#define MAX_SCORE 60000.0
#define DEFAULT_RENAME_SCORE 30000 /* rename/copy similarity minimum (50%) */
-#define DEFAULT_BREAK_SCORE 30000 /* minimum for break to happen (50%)*/
-#define DEFAULT_MERGE_SCORE 48000 /* maximum for break-merge to happen (80%)*/
+#define DEFAULT_BREAK_SCORE 30000 /* minimum for break to happen (50%) */
+#define DEFAULT_MERGE_SCORE 36000 /* maximum for break-merge to happen 60%) */
#define MINIMUM_BREAK_SIZE 400 /* do not break a file smaller than this */