--histogram::
Generate a diff using the "histogram diff" algorithm.
+--anchored=<text>::
+ Generate a diff using the "anchored diff" algorithm.
++
+This option may be specified more than once.
++
+If a line exists in both the source and destination, exists only once,
+and starts with this text, this algorithm attempts to prevent it from
+appearing as a deletion or addition in the output. It uses the "patience
+diff" algorithm internally.
+
--diff-algorithm={patience|minimal|histogram|myers}::
Choose a diff algorithm. The variants are as follows:
+
ecbdata.opt = o;
ecbdata.header = header.len ? &header : NULL;
xpp.flags = o->xdl_opts;
+ xpp.anchors = o->anchors;
+ xpp.anchors_nr = o->anchors_nr;
xecfg.ctxlen = o->context;
xecfg.interhunkctxlen = o->interhunkcontext;
xecfg.flags = XDL_EMIT_FUNCNAMES;
memset(&xpp, 0, sizeof(xpp));
memset(&xecfg, 0, sizeof(xecfg));
xpp.flags = o->xdl_opts;
+ xpp.anchors = o->anchors;
+ xpp.anchors_nr = o->anchors_nr;
xecfg.ctxlen = o->context;
xecfg.interhunkctxlen = o->interhunkcontext;
if (xdi_diff_outf(&mf1, &mf2, diffstat_consume, diffstat,
DIFF_XDL_SET(options, INDENT_HEURISTIC);
else if (!strcmp(arg, "--no-indent-heuristic"))
DIFF_XDL_CLR(options, INDENT_HEURISTIC);
- else if (!strcmp(arg, "--patience"))
+ else if (!strcmp(arg, "--patience")) {
+ int i;
options->xdl_opts = DIFF_WITH_ALG(options, PATIENCE_DIFF);
- else if (!strcmp(arg, "--histogram"))
+ /*
+ * Both --patience and --anchored use PATIENCE_DIFF
+ * internally, so remove any anchors previously
+ * specified.
+ */
+ for (i = 0; i < options->anchors_nr; i++)
+ free(options->anchors[i]);
+ options->anchors_nr = 0;
+ } else if (!strcmp(arg, "--histogram"))
options->xdl_opts = DIFF_WITH_ALG(options, HISTOGRAM_DIFF);
else if ((argcount = parse_long_opt("diff-algorithm", av, &optarg))) {
long value = parse_algorithm_value(optarg);
options->xdl_opts &= ~XDF_DIFF_ALGORITHM_MASK;
options->xdl_opts |= value;
return argcount;
+ } else if (skip_prefix(arg, "--anchored=", &arg)) {
+ options->xdl_opts = DIFF_WITH_ALG(options, PATIENCE_DIFF);
+ ALLOC_GROW(options->anchors, options->anchors_nr + 1,
+ options->anchors_alloc);
+ options->anchors[options->anchors_nr++] = xstrdup(arg);
}
/* flags options */
const char *stat_sep;
long xdl_opts;
+ /* see Documentation/diff-options.txt */
+ char **anchors;
+ size_t anchors_nr, anchors_alloc;
+
int stat_width;
int stat_name_width;
int stat_graph_width;
--- /dev/null
+#!/bin/sh
+
+test_description='anchored diff algorithm'
+
+. ./test-lib.sh
+
+test_expect_success '--anchored' '
+ printf "a\nb\nc\n" >pre &&
+ printf "c\na\nb\n" >post &&
+
+ # normally, c is moved to produce the smallest diff
+ test_expect_code 1 git diff --no-index pre post >diff &&
+ grep "^+c" diff &&
+
+ # with anchor, a is moved
+ test_expect_code 1 git diff --no-index --anchored=c pre post >diff &&
+ grep "^+a" diff
+'
+
+test_expect_success '--anchored multiple' '
+ printf "a\nb\nc\nd\ne\nf\n" >pre &&
+ printf "c\na\nb\nf\nd\ne\n" >post &&
+
+ # with 1 anchor, c is not moved, but f is moved
+ test_expect_code 1 git diff --no-index --anchored=c pre post >diff &&
+ grep "^+a" diff && # a is moved instead of c
+ grep "^+f" diff &&
+
+ # with 2 anchors, c and f are not moved
+ test_expect_code 1 git diff --no-index --anchored=c --anchored=f pre post >diff &&
+ grep "^+a" diff &&
+ grep "^+d" diff # d is moved instead of f
+'
+
+test_expect_success '--anchored with nonexistent line has no effect' '
+ printf "a\nb\nc\n" >pre &&
+ printf "c\na\nb\n" >post &&
+
+ test_expect_code 1 git diff --no-index --anchored=x pre post >diff &&
+ grep "^+c" diff
+'
+
+test_expect_success '--anchored with non-unique line has no effect' '
+ printf "a\nb\nc\nd\ne\nc\n" >pre &&
+ printf "c\na\nb\nc\nd\ne\n" >post &&
+
+ test_expect_code 1 git diff --no-index --anchored=c pre post >diff &&
+ grep "^+c" diff
+'
+
+test_expect_success 'diff still produced with impossible multiple --anchored' '
+ printf "a\nb\nc\n" >pre &&
+ printf "c\na\nb\n" >post &&
+
+ test_expect_code 1 git diff --no-index --anchored=a --anchored=c pre post >diff &&
+ mv post expected_post &&
+
+ # Ensure that the diff is correct by applying it and then
+ # comparing the result with the original
+ git apply diff &&
+ diff expected_post post
+'
+
+test_expect_success 'later algorithm arguments override earlier ones' '
+ printf "a\nb\nc\n" >pre &&
+ printf "c\na\nb\n" >post &&
+
+ test_expect_code 1 git diff --no-index --patience --anchored=c pre post >diff &&
+ grep "^+a" diff &&
+
+ test_expect_code 1 git diff --no-index --anchored=c --patience pre post >diff &&
+ grep "^+c" diff &&
+
+ test_expect_code 1 git diff --no-index --histogram --anchored=c pre post >diff &&
+ grep "^+a" diff &&
+
+ test_expect_code 1 git diff --no-index --anchored=c --histogram pre post >diff &&
+ grep "^+c" diff
+'
+
+test_expect_success '--anchored works with other commands like "git show"' '
+ printf "a\nb\nc\n" >file &&
+ git add file &&
+ git commit -m foo &&
+ printf "c\na\nb\n" >file &&
+ git add file &&
+ git commit -m foo &&
+
+ # with anchor, a is moved
+ git show --patience --anchored=c >diff &&
+ grep "^+a" diff
+'
+
+test_done
typedef struct s_xpparam {
unsigned long flags;
+
+ /* See Documentation/diff-options.txt. */
+ char **anchors;
+ size_t anchors_nr;
} xpparam_t;
typedef struct s_xdemitcb {
* initially, "next" reflects only the order in file1.
*/
struct entry *next, *previous;
+
+ /*
+ * If 1, this entry can serve as an anchor. See
+ * Documentation/diff-options.txt for more information.
+ */
+ unsigned anchor : 1;
} *entries, *first, *last;
/* were common records found? */
unsigned long has_matches;
xpparam_t const *xpp;
};
+static int is_anchor(xpparam_t const *xpp, const char *line)
+{
+ int i;
+ for (i = 0; i < xpp->anchors_nr; i++) {
+ if (!strncmp(line, xpp->anchors[i], strlen(xpp->anchors[i])))
+ return 1;
+ }
+ return 0;
+}
+
/* The argument "pass" is 1 for the first file, 2 for the second. */
-static void insert_record(int line, struct hashmap *map, int pass)
+static void insert_record(xpparam_t const *xpp, int line, struct hashmap *map,
+ int pass)
{
xrecord_t **records = pass == 1 ?
map->env->xdf1.recs : map->env->xdf2.recs;
return;
map->entries[index].line1 = line;
map->entries[index].hash = record->ha;
+ map->entries[index].anchor = is_anchor(xpp, map->env->xdf1.recs[line - 1]->ptr);
if (!map->first)
map->first = map->entries + index;
if (map->last) {
/* First, fill with entries from the first file */
while (count1--)
- insert_record(line1++, result, 1);
+ insert_record(xpp, line1++, result, 1);
/* Then search for matches in the second file */
while (count2--)
- insert_record(line2++, result, 2);
+ insert_record(xpp, line2++, result, 2);
return 0;
}
int longest = 0, i;
struct entry *entry;
+ /*
+ * If not -1, this entry in sequence must never be overridden.
+ * Therefore, overriding entries before this has no effect, so
+ * do not do that either.
+ */
+ int anchor_i = -1;
+
for (entry = map->first; entry; entry = entry->next) {
if (!entry->line2 || entry->line2 == NON_UNIQUE)
continue;
i = binary_search(sequence, longest, entry);
entry->previous = i < 0 ? NULL : sequence[i];
- sequence[++i] = entry;
- if (i == longest)
+ ++i;
+ if (i <= anchor_i)
+ continue;
+ sequence[i] = entry;
+ if (entry->anchor) {
+ anchor_i = i;
+ longest = anchor_i + 1;
+ } else if (i == longest) {
longest++;
+ }
}
/* No common unique lines were found */