teach --histogram to diff

Port JGit's HistogramDiff algorithm over to C. Rough numbers (TODO) show
that it is faster than its --patience cousin, as well as the default
Meyers algorithm.

The implementation has been reworked to use structs and pointers,
instead of bitmasks, thus doing away with JGit's 2^28 line limit.

We also use xdiff's default hash table implementation (xdl_hash_bits()
with XDL_HASHLONG()) for convenience.

Signed-off-by: Tay Ray Chuan <[email protected]>
Signed-off-by: Junio C Hamano <[email protected]>

Author: rctay

Makefile

@@ -1838,7 +1838,7 @@ ifndef NO_CURL
 	GIT_OBJS += http.o http-walker.o remote-curl.o
 endif
 XDIFF_OBJS = xdiff/xdiffi.o xdiff/xprepare.o xdiff/xutils.o xdiff/xemit.o \
-	xdiff/xmerge.o xdiff/xpatience.o
+	xdiff/xmerge.o xdiff/xpatience.o xdiff/xhistogram.o
 VCSSVN_OBJS = vcs-svn/string_pool.o vcs-svn/line_buffer.o \
 	vcs-svn/repo_tree.o vcs-svn/fast_export.o vcs-svn/svndump.o
 VCSSVN_TEST_OBJS = test-obj-pool.o test-string-pool.o \

diff.c

@@ -3369,6 +3369,8 @@ int diff_opt_parse(struct diff_options *options, const char **av, int ac)
 		DIFF_XDL_SET(options, IGNORE_WHITESPACE_AT_EOL);
 	else if (!strcmp(arg, "--patience"))
 		DIFF_XDL_SET(options, PATIENCE_DIFF);
+	else if (!strcmp(arg, "--histogram"))
+		DIFF_XDL_SET(options, HISTOGRAM_DIFF);
 
 	/* flags options */
 	else if (!strcmp(arg, "--binary")) {

merge-recursive.c

@@ -1759,6 +1759,8 @@ int parse_merge_opt(struct merge_options *o, const char *s)
 		o->subtree_shift = s + strlen("subtree=");
 	else if (!strcmp(s, "patience"))
 		o->xdl_opts |= XDF_PATIENCE_DIFF;
+	else if (!strcmp(s, "histogram"))
+		o->xdl_opts |= XDF_HISTOGRAM_DIFF;
 	else if (!strcmp(s, "ignore-space-change"))
 		o->xdl_opts |= XDF_IGNORE_WHITESPACE_CHANGE;
 	else if (!strcmp(s, "ignore-all-space"))

t/t4049-diff-histogram.sh

@@ -0,0 +1,12 @@
+#!/bin/sh
+
+test_description='histogram diff algorithm'
+
+. ./test-lib.sh
+. "$TEST_DIRECTORY"/lib-diff-alternative.sh
+
+test_diff_frobnitz "histogram"
+
+test_diff_unique "histogram"
+
+test_done

xdiff/xdiff.h

@@ -33,6 +33,7 @@ extern "C" {
 #define XDF_IGNORE_WHITESPACE_CHANGE (1 << 3)
 #define XDF_IGNORE_WHITESPACE_AT_EOL (1 << 4)
 #define XDF_PATIENCE_DIFF (1 << 5)
+#define XDF_HISTOGRAM_DIFF (1 << 6)
 #define XDF_WHITESPACE_FLAGS (XDF_IGNORE_WHITESPACE | XDF_IGNORE_WHITESPACE_CHANGE | XDF_IGNORE_WHITESPACE_AT_EOL)
 
 #define XDL_PATCH_NORMAL '-'

xdiff/xdiffi.c

@@ -331,6 +331,9 @@ int xdl_do_diff(mmfile_t *mf1, mmfile_t *mf2, xpparam_t const *xpp,
 	if (xpp->flags & XDF_PATIENCE_DIFF)
 		return xdl_do_patience_diff(mf1, mf2, xpp, xe);
 
+	if (xpp->flags & XDF_HISTOGRAM_DIFF)
+		return xdl_do_histogram_diff(mf1, mf2, xpp, xe);
+
 	if (xdl_prepare_env(mf1, mf2, xpp, xe) < 0) {
 
 		return -1;

xdiff/xdiffi.h

@@ -57,5 +57,7 @@ int xdl_emit_diff(xdfenv_t *xe, xdchange_t *xscr, xdemitcb_t *ecb,
 		  xdemitconf_t const *xecfg);
 int xdl_do_patience_diff(mmfile_t *mf1, mmfile_t *mf2, xpparam_t const *xpp,
 		xdfenv_t *env);
+int xdl_do_histogram_diff(mmfile_t *mf1, mmfile_t *mf2, xpparam_t const *xpp,
+		xdfenv_t *env);
 
 #endif /* #if !defined(XDIFFI_H) */