Re: [PATCH v3 03/13] add the basic data structure for line level history

@@ -496,6 +496,7 @@ LIB_H += grep.h
 LIB_H += hash.h
 LIB_H += help.h
 LIB_H += levenshtein.h
+LIB_H += line.h
 LIB_H += list-objects.h
 LIB_H += ll-merge.h
 LIB_H += log-tree.h

@@ -582,6 +583,7 @@ LIB_OBJS += help.o
 LIB_OBJS += hex.o
 LIB_OBJS += ident.o
 LIB_OBJS += levenshtein.o
+LIB_OBJS += line.o
 LIB_OBJS += list-objects.o
 LIB_OBJS += ll-merge.o
 LIB_OBJS += lockfile.o

@@ -23,6 +23,7 @@
 #define MINIMUM_BREAK_SIZE     400 /* do not break a file smaller than this */

 struct userdiff_driver;
+struct diff_options;

 struct diff_filespec {
 	unsigned char sha1[20];

@@ -0,0 +1,456 @@
+#include "line.h"
+#include "cache.h"
+#include "tag.h"
+#include "blob.h"
+#include "tree.h"
+#include "commit.h"
+#include "diff.h"
+#include "decorate.h"
+#include "revision.h"
+#include "xdiff-interface.h"
+#include "strbuf.h"
+#include "log-tree.h"
+
+static void cleanup(struct diff_line_range *r)
+{
+	while (r) {
+		struct diff_line_range *next = r->next;
+		DIFF_LINE_RANGE_CLEAR(r);
+		free(r);
+		r = next;
+	}
+}
+
+static struct object *verify_commit(struct rev_info *revs)
+{
+	struct object *commit = NULL;
+	const char *name = NULL;
+	int i;
+
+	for (i = 0; i < revs->pending.nr; i++) {
+		struct object *obj = revs->pending.objects[i].item;
+		if (obj->flags & UNINTERESTING)
+			continue;
+		while (obj->type == OBJ_TAG)
+			obj = deref_tag(obj, NULL, 0);
+		if (obj->type != OBJ_COMMIT)
+			die("Non commit %s?", revs->pending.objects[i].name);
+		if (commit)
+			die("More than one commit to dig from: %s and %s?",
+			    revs->pending.objects[i].name, name);
+		commit = obj;
+		name = revs->pending.objects[i].name;
+	}
+
+	if (commit == NULL)
+		die("No commit specified?");
+
+	return commit;
+}
+
+static void fill_blob_sha1(struct commit *commit, struct diff_line_range *r)
+{
+	unsigned mode;
+	unsigned char sha1[20];
+
+	while (r) {
+		if (get_tree_entry(commit->object.sha1, r->spec->path,
+			sha1, &mode))
+			goto error;
+		fill_filespec(r->spec, sha1, mode);
+		r = r->next;
+	}
+
+	return;
+error:
+	die("There is no path %s in the commit", r->spec->path);
+}
+
+static void fill_line_ends(struct diff_filespec *spec, long *lines,
+	unsigned long **line_ends)
+{
+	int num = 0, size = 50;
+	long cur = 0;
+	unsigned long *ends = NULL;
+	char *data = NULL;
+
+	if (diff_populate_filespec(spec, 0))
+		die("Cannot read blob %s", sha1_to_hex(spec->sha1));
+
+	ends = xmalloc(size * sizeof(*ends));
+	ends[cur++] = -1;
+	data = spec->data;
+	while (num < spec->size) {
+		if (data[num] == '\n' || num == spec->size - 1) {
+			ALLOC_GROW(ends, (cur + 1), size);
+			ends[cur++] = num;
+		}
+		num++;
+	}
+
+	/* shrink the array to fit the elements */
+	ends = xrealloc(ends, cur * sizeof(*ends));
+	*lines = cur;
+	*line_ends = ends;
+}
+
+static const char *nth_line(struct diff_filespec *spec, int line,
+		long lines, unsigned long *line_ends)
+{
+	assert(line < lines);
+	assert(spec && spec->data);
+
+	return (char *)spec->data + line_ends[line] + 1;
+}
+
+/*
+ * copied from blame.c, indeed, we can even to use this to test
+ * whether line log works. :)
+ */
+static const char *parse_loc(const char *spec, struct diff_filespec *file,
+			     long lines, unsigned long *line_ends,
+			     long begin, long *ret)
+{
+	char *term;
+	const char *line;
+	long num;
+	int reg_error;
+	regex_t regexp;
+	regmatch_t match[1];
+
+	/* Allow "-L <something>,+20" to mean starting at <something>
+	 * for 20 lines, or "-L <something>,-5" for 5 lines ending at
+	 * <something>.
+	 */
+	if (1 < begin && (spec[0] == '+' || spec[0] == '-')) {
+		num = strtol(spec + 1, &term, 10);
+		if (term != spec + 1) {
+			if (spec[0] == '-')
+				num = 0 - num;
+			if (0 < num)
+				*ret = begin + num - 2;
+			else if (!num)
+				*ret = begin;
+			else
+				*ret = begin + num;
+			return term;
+		}
+		return spec;
+	}
+	num = strtol(spec, &term, 10);
+	if (term != spec) {
+		*ret = num;
+		return term;
+	}
+	if (spec[0] != '/')
+		return spec;
+
+	/* it could be a regexp of form /.../ */
+	for (term = (char *) spec + 1; *term && *term != '/'; term++) {
+		if (*term == '\\')
+			term++;
+	}
+	if (*term != '/')
+		return spec;
+
+	/* try [spec+1 .. term-1] as regexp */
+	*term = 0;
+	begin--; /* input is in human terms */
+	line = nth_line(file, begin, lines, line_ends);
+
+	if (!(reg_error = regcomp(&regexp, spec + 1, REG_NEWLINE)) &&
+	    !(reg_error = regexec(&regexp, line, 1, match, 0))) {
+		const char *cp = line + match[0].rm_so;
+		const char *nline;
+
+		while (begin++ < lines) {
+			nline = nth_line(file, begin, lines, line_ends);
+			if (line <= cp && cp < nline)
+				break;
+			line = nline;
+		}
+		*ret = begin;
+		regfree(&regexp);
+		*term++ = '/';
+		return term;
+	}
+	else {
+		char errbuf[1024];
+		regerror(reg_error, &regexp, errbuf, 1024);
+		die("-L parameter '%s': %s", spec + 1, errbuf);
+	}
+}
+
+static void parse_range(long lines, unsigned long *line_ends,
+		struct range *r, struct diff_filespec *spec)
+{
+	const char *term;
+
+	term = parse_loc(r->arg, spec, lines, line_ends, 1, &r->start);
+	if (*term == ',') {
+		term = parse_loc(term + 1, spec, lines, line_ends,
+			r->start + 1, &r->end);
+		if (*term) {
+			die("-L parameter's argument should be <start>,<end>");
+		}
+	}
+
+	if (*term) {
+		die("-L parameter's argument should be <start>,<end>");
+	}
+
+	if (r->start > r->end) {
+		long tmp = r->start;
+		r->start = r->end;
+		r->end = tmp;
+	}
+
+	if (r->start < 1)
+		r->start = 1;
+	if (r->end >= lines)
+		r->end = lines - 1;
+}
+
+static void parse_lines(struct commit *commit, struct diff_line_range *r)
+{
+	int i;
+	struct range *old_range = NULL;
+	long lines = 0;
+	unsigned long *ends = NULL;
+
+	while (r) {
+		struct diff_filespec *spec = r->spec;
+		int num = r->nr;
+		assert(spec);
+		fill_blob_sha1(commit, r);
+		old_range = r->ranges;
+		r->ranges = NULL;
+		r->nr = r->alloc = 0;
+		fill_line_ends(spec, &lines, &ends);
+		for (i = 0; i < num; i++) {
+			parse_range(lines, ends, old_range + i, spec);
+			diff_line_range_insert(r, old_range[i].arg,
+				old_range[i].start, old_range[i].end);
+		}
+
+		free(ends);
+		ends = NULL;
+
+		r = r->next;
+		free(old_range);
+	}
+}
+
+/*
+ * Insert a new line range into a diff_line_range struct, and keep the
+ * r->ranges sorted by their starting line number.
+ */
+struct range *diff_line_range_insert(struct diff_line_range *r, const

@@ -0,0 +1,122 @@
+#ifndef LINE_H
+#define LINE_H
+
+#include "diffcore.h"
+
+struct rev_info;
+struct commit;
+struct diff_line_range;
+struct diff_options;
+
+struct print_range {
+	int start, end;
+	int pstart, pend;
+	int line_added : 1;
+};
+
+struct print_pair {
+	int alloc, nr;
+	struct print_range *ranges;
+};
+
+#define PRINT_RANGE_INIT(r) \
+	do { \
+		(r)->line_added = 0; \
+	} while(0);
+
+#define PRINT_PAIR_INIT(p) \
+	do { \
+		(p)->alloc = (p)->nr = 0; \
+		(p)->ranges = NULL; \
+	} while(0);
+
+#define PRINT_PAIR_GROW(p) \
+	do { \
+		(p)->nr ++; \
+		ALLOC_GROW((p)->ranges, (p)->nr, (p)->alloc); \
+	} while (0);
+
+#define PRINT_PAIR_CLEAR(p) \
+	do { \
+		(p)->alloc = (p)->nr = 0; \
+		if ((p)->ranges) \
+			free((p)->ranges); \
+		(p)->ranges = NULL; \
+	} while(0);
+
+struct range {
+	const char *arg;	//The argument to specify this line range
+	long start, end;	//The start line number, inclusive
+	long pstart, pend;	//The end line number, inclusive
+	struct print_pair pair;
+			//The changed lines inside this range
+	unsigned int diff:1;
+};
+
+struct diff_line_range {
+	struct diff_filespec *prev;
+	struct diff_filespec *spec;
+	char status;
+	int alloc;
+	int nr;
+	struct range *ranges;
+	unsigned int	touch:1,
+			diff:1;
+	struct diff_line_range *next;
+};
+
+#define RANGE_INIT(r) \
+	do { \
+		(r)->arg = NULL; \
+		(r)->start = (r)->end = 0; \
+		(r)->pstart = (r)->pend = 0; \
+		PRINT_PAIR_INIT(&((r)->pair)); \
+		(r)->diff = 0; \
+	} while (0);
+
+#define RANGE_CLEAR(r) \
+	do { \
+		(r)->arg = NULL; \
+		(r)->start = (r)->end = 0; \
+		(r)->pstart = (r)->pend = 0; \
+		PRINT_PAIR_CLEAR(&r->pair); \
+		(r)->diff = 0; \
+	} while(0);
+
+#define DIFF_LINE_RANGE_INIT(r) \
+	do { \
+		(r)->prev = (r)->spec = NULL; \
+		(r)->status = '\0'; \
+		(r)->alloc = (r)->nr = 0; \
+		(r)->ranges = NULL; \
+		(r)->next = NULL; \
+		(r)->touch = 0; \
+		(r)->diff = 0; \
+	} while(0);
+
+#define DIFF_LINE_RANGE_GROW(r) \
+	do { \
+		(r)->nr ++; \
+		ALLOC_GROW((r)->ranges, (r)->nr, (r)->alloc); \
+		RANGE_INIT(((r)->ranges + (r)->nr - 1)); \
+	} while(0);
+
+#define DIFF_LINE_RANGE_CLEAR(r) \
+	diff_line_range_clear((r));
+
+extern struct range *diff_line_range_insert(struct diff_line_range

@@ -1637,6 +1637,12 @@ int setup_revisions(int argc, const char

@@ -14,7 +14,8 @@
 #define CHILD_SHOWN	(1u<<6)
 #define ADDED		(1u<<7)	/* Parents already parsed and added? */
 #define SYMMETRIC_LEFT	(1u<<8)
-#define ALL_REV_FLAGS	((1u<<9)-1)
+#define RANGE_UPDATE	(1u<<9) /* for line level traverse */
+#define ALL_REV_FLAGS	((1u<<10)-1)

 #define DECORATE_SHORT_REFS	1
 #define DECORATE_FULL_REFS	2

@@ -68,7 +69,8 @@ struct rev_info {
 			cherry_pick:1,
 			bisect:1,
 			ancestry_path:1,
-			first_parent_only:1;
+			first_parent_only:1,
+			line:1;

 	/* Diff flags */
 	unsigned int	diff:1,

@@ -137,6 +139,8 @@ struct rev_info {
 	/* commit counts */
 	int count_left;
 	int count_right;
+	/* line-level intersting range */
+	struct decoration line_range;
 };

 #define REV_TREE_SAME		0