[PATCH v7 0/5] git log -L, all new and shiny

[PATCH v7 0/5] git log -L, all new and shiny

[Thomas Rast]

I too thought it would never happen -- but then again this is still
not ready, I'm just trying to give it some exposure.

My mail archive says we were at v6 last time, but this is a rewrite
rather than a reroll.  In efforts to make the code understandable, I
tried to compose it of little functions that do little operations on
little data structures, to at least some success.

If you just want to dive in, patch it into your git or pull from

  git://github.com/trast/git.git line-log-WIP

and then try something like

  git log -L '/^int main/,/^}/':git.c
  git log --graph -L '/^static void sha1_object/,/^}/':builtin/index-pack.c 681b07de11

The TODO list:

* I dropped the tests for now, so we need new ones.

* A good split of the main patch would be nice :-)

* Much of the code leaks memory like a sieve because I was lazy.  It's
  workable on the git repository, but eats a few hundred MB for simple
  tasks even there.

* Missing features:
  - detect/filter merges at the hunk level
    (and ideally show a combined diff for the nontrivial ones)
  - -M and -C support
  - ...?

There's also a longer-term wishlist hinted at in the commit message of
the main patch: the diff machinery currently makes no provisions for
chaining its various bells and whistles.  As the most obvious example,
it is not possible to run a word diff over the result of 'log -L'.
Changes in this direction would probably also make the implementation
of 'log -L' fit nicely in there somewhere.


Bo Yang (3):
  Refactor parse_loc
  Export three functions from diff.c
  Export rewrite_parents() for 'log -L'

Thomas Rast (2):
  blame: introduce $ as "end of file" in -L syntax
  Implement line-history search (git log -L)

 Documentation/blame-options.txt     |   19 +-
 Documentation/git-log.txt           |   22 +
 Documentation/line-range-format.txt |   24 +
 Makefile                            |    2 +
 builtin/blame.c                     |   99 +--
 builtin/log.c                       |   53 ++
 diff.c                              |    6 +-
 diff.h                              |   17 +
 line.c                              | 1245 +++++++++++++++++++++++++++++++++++
 line.h                              |   79 +++
 log-tree.c                          |    3 +
 revision.c                          |   22 +-
 revision.h                          |   16 +-
 t/t8003-blame-corner-cases.sh       |    6 +
 14 files changed, 1491 insertions(+), 122 deletions(-)
 create mode 100644 Documentation/line-range-format.txt
 create mode 100644 line.c
 create mode 100644 line.h

-- 
1.7.11.rc1.243.gbf4713c

[PATCH v7 1/5] Refactor parse_loc

[Thomas Rast]

From: Bo Yang <struggleyb.nku@gmail.com>

We want to use the same style of -L n,m argument for 'git log -L' as
for git-blame.  Refactor the argument parsing of the range arguments
from builtin/blame.c to the (new) file that will hold the 'git log -L'
logic.

To accommodate different data structures in blame and log -L, the file
contents are abstracted away; parse_range_arg takes a callback that it
uses to get the contents of a line of the (notional) file.

The new test is for a case that made me pause during debugging: the
'blame -L with invalid end' test was the only one that noticed an
outright failure to parse the end *at all*.  So make a more explicit
test for that.

Signed-off-by: Bo Yang <struggleyb.nku@gmail.com>
Signed-off-by: Thomas Rast <trast@student.ethz.ch>
---
 Documentation/blame-options.txt     |  19 +------
 Documentation/line-range-format.txt |  18 +++++++
 Makefile                            |   2 +
 builtin/blame.c                     |  99 +++--------------------------------
 line.c                              | 100 ++++++++++++++++++++++++++++++++++++
 line.h                              |  23 +++++++++
 t/t8003-blame-corner-cases.sh       |   6 +++
 7 files changed, 158 insertions(+), 109 deletions(-)
 create mode 100644 Documentation/line-range-format.txt
 create mode 100644 line.c
 create mode 100644 line.h

diff --git a/Documentation/blame-options.txt b/Documentation/blame-options.txt
index d4a51da..1d9305e 100644
--- a/Documentation/blame-options.txt
+++ b/Documentation/blame-options.txt
@@ -13,24 +13,7 @@
 	Annotate only the given line range.  <start> and <end> can take
 	one of these forms:
 
-	- number
-+
-If <start> or <end> is a number, it specifies an
-absolute line number (lines count from 1).
-+
-
-- /regex/
-+
-This form will use the first line matching the given
-POSIX regex.  If <end> is a regex, it will search
-starting at the line given by <start>.
-+
-
-- +offset or -offset
-+
-This is only valid for <end> and will specify a number
-of lines before or after the line given by <start>.
-+
+include::line-range-format.txt[]
 
 -l::
 	Show long rev (Default: off).
diff --git a/Documentation/line-range-format.txt b/Documentation/line-range-format.txt
new file mode 100644
index 0000000..265bc23
--- /dev/null
+++ b/Documentation/line-range-format.txt
@@ -0,0 +1,18 @@
+- number
++
+If <start> or <end> is a number, it specifies an
+absolute line number (lines count from 1).
++
+
+- /regex/
++
+This form will use the first line matching the given
+POSIX regex.  If <end> is a regex, it will search
+starting at the line given by <start>.
++
+
+- +offset or -offset
++
+This is only valid for <end> and will specify a number
+of lines before or after the line given by <start>.
++
diff --git a/Makefile b/Makefile
index 4592f1f..9166b86 100644
--- a/Makefile
+++ b/Makefile
@@ -626,6 +626,7 @@ LIB_H += hash.h
 LIB_H += help.h
 LIB_H += kwset.h
 LIB_H += levenshtein.h
+LIB_H += line.h
 LIB_H += list-objects.h
 LIB_H += ll-merge.h
 LIB_H += log-tree.h
@@ -731,6 +732,7 @@ LIB_OBJS += hex.o
 LIB_OBJS += ident.o
 LIB_OBJS += kwset.o
 LIB_OBJS += levenshtein.o
+LIB_OBJS += line.o
 LIB_OBJS += list-objects.o
 LIB_OBJS += ll-merge.o
 LIB_OBJS += lockfile.o
diff --git a/builtin/blame.c b/builtin/blame.c
index 24d3dd5..c3b379b 100644
--- a/builtin/blame.c
+++ b/builtin/blame.c
@@ -21,6 +21,7 @@
 #include "parse-options.h"
 #include "utf8.h"
 #include "userdiff.h"
+#include "line.h"
 
 static char blame_usage[] = "git blame [options] [rev-opts] [rev] [--] file";
 
@@ -566,11 +567,16 @@ static void dup_entry(struct blame_entry *dst, struct blame_entry *src)
 	dst->score = 0;
 }
 
-static const char *nth_line(struct scoreboard *sb, int lno)
+static const char *nth_line(struct scoreboard *sb, long lno)
 {
 	return sb->final_buf + sb->lineno[lno];
 }
 
+static const char *nth_line_cb(void *data, long lno)
+{
+	return nth_line((struct scoreboard *)data, lno);
+}
+
 /*
  * It is known that lines between tlno to same came from parent, and e
  * has an overlap with that range.  it also is known that parent's
@@ -1934,83 +1940,6 @@ static const char *add_prefix(const char *prefix, const char *path)
 }
 
 /*
- * Parsing of (comma separated) one item in the -L option
- */
-static const char *parse_loc(const char *spec,
-			     struct scoreboard *sb, long lno,
-			     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(sb, begin);
-
-	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++ < lno) {
-			nline = nth_line(sb, begin);
-			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);
-	}
-}
-
-/*
  * Parsing of -L option
  */
 static void prepare_blame_range(struct scoreboard *sb,
@@ -2018,15 +1947,7 @@ static void prepare_blame_range(struct scoreboard *sb,
 				long lno,
 				long *bottom, long *top)
 {
-	const char *term;
-
-	term = parse_loc(bottomtop, sb, lno, 1, bottom);
-	if (*term == ',') {
-		term = parse_loc(term + 1, sb, lno, *bottom + 1, top);
-		if (*term)
-			usage(blame_usage);
-	}
-	if (*term)
+	if (parse_range_arg(bottomtop, nth_line_cb, sb, lno, bottom, top))
 		usage(blame_usage);
 }
 
@@ -2516,10 +2437,6 @@ int cmd_blame(int argc, const char **argv, const char *prefix)
 	bottom = top = 0;
 	if (bottomtop)
 		prepare_blame_range(&sb, bottomtop, lno, &bottom, &top);
-	if (bottom && top && top < bottom) {
-		long tmp;
-		tmp = top; top = bottom; bottom = tmp;
-	}
 	if (bottom < 1)
 		bottom = 1;
 	if (top < 1)
diff --git a/line.c b/line.c
new file mode 100644
index 0000000..afd2e3b
--- /dev/null
+++ b/line.c
@@ -0,0 +1,100 @@
+#include "git-compat-util.h"
+#include "line.h"
+
+/*
+ * Parse one item in the -L option
+ */
+const char *parse_loc(const char *spec, nth_line_fn_t nth_line,
+		void *data, long lines, 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 (begin != -1 && (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;
+	if (begin == -1)
+		begin = 1;
+	begin--; /* input is in human terms */
+	line = nth_line(data, begin);
+
+	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(data, begin);
+			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);
+	}
+}
+
+int parse_range_arg(const char *arg, nth_line_fn_t nth_line_cb,
+		void *cb_data, long lines, long *begin, long *end)
+{
+	arg = parse_loc(arg, nth_line_cb, cb_data, lines, -1, begin);
+
+	if (*arg == ',') {
+		arg = parse_loc(arg+1, nth_line_cb, cb_data, lines, *begin+1, end);
+		if (*begin > *end) {
+			long tmp = *begin;
+			*begin = *end;
+			*end = tmp;
+		}
+	}
+
+	if (*arg)
+		return -1;
+
+	return 0;
+}
diff --git a/line.h b/line.h
new file mode 100644
index 0000000..5878c94
--- /dev/null
+++ b/line.h
@@ -0,0 +1,23 @@
+#ifndef LINE_H
+#define LINE_H
+
+/*
+ * Parse one item in an -L begin,end option w.r.t. the notional file
+ * object 'cb_data' consisting of 'lines' lines.
+ *
+ * The 'nth_line_cb' callback is used to determine the start of the
+ * line 'lno' inside the 'cb_data'.  The caller is expected to already
+ * have a suitable map at hand to make this a constant-time lookup.
+ *
+ * Returns 0 in case of success and -1 if there was an error.  The
+ * caller should print a usage message in the latter case.
+ */
+
+typedef const char *(*nth_line_fn_t)(void *data, long lno);
+
+extern int parse_range_arg(const char *arg,
+			   nth_line_fn_t nth_line_cb,
+			   void *cb_data, long lines,
+			   long *begin, long *end);
+
+#endif /* LINE_H */
diff --git a/t/t8003-blame-corner-cases.sh b/t/t8003-blame-corner-cases.sh
index 230143c..e7cac1d 100755
--- a/t/t8003-blame-corner-cases.sh
+++ b/t/t8003-blame-corner-cases.sh
@@ -175,6 +175,12 @@ test_expect_success 'blame -L with invalid end' '
 	grep "has only 2 lines" errors
 '
 
+test_expect_success 'blame parses <end> part of -L' '
+	git blame -L1,1 tres >out &&
+	cat out &&
+	test $(wc -l < out) -eq 1
+'
+
 test_expect_success 'indent of line numbers, nine lines' '
 	git blame nine_lines >actual &&
 	test $(grep -c "  " actual) = 0
-- 
1.7.11.rc1.243.gbf4713c

[PATCH v7 2/5] blame: introduce $ as "end of file" in -L syntax

[Thomas Rast]

To save the user a lookup of the last line number, introduce $ as a
shorthand for the last line.  This is mostly useful to spell "until
the end of the file" as '-L<begin>,$'.

Signed-off-by: Thomas Rast <trast@student.ethz.ch>
---
 Documentation/line-range-format.txt | 6 ++++++
 line.c                              | 8 ++++++++
 2 files changed, 14 insertions(+)

diff --git a/Documentation/line-range-format.txt b/Documentation/line-range-format.txt
index 265bc23..9ce0688 100644
--- a/Documentation/line-range-format.txt
+++ b/Documentation/line-range-format.txt
@@ -16,3 +16,9 @@ starting at the line given by <start>.
 This is only valid for <end> and will specify a number
 of lines before or after the line given by <start>.
 +
+
+- `$`
++
+A literal dollar sign can be used as a shorthand for the last line in
+the file.
++
diff --git a/line.c b/line.c
index afd2e3b..a7f33ed 100644
--- a/line.c
+++ b/line.c
@@ -15,6 +15,14 @@ const char *parse_loc(const char *spec, nth_line_fn_t nth_line,
 	regmatch_t match[1];
 
 	/*
+	 * $ is a synonym for "the end of the file".
+	 */
+	if (spec[0] == '$') {
+		*ret = lines;
+		return spec + 1;
+	}
+
+	/*
 	 * Allow "-L <something>,+20" to mean starting at <something>
 	 * for 20 lines, or "-L <something>,-5" for 5 lines ending at
 	 * <something>.
-- 
1.7.11.rc1.243.gbf4713c

[PATCH v7 3/5] Export three functions from diff.c

[Thomas Rast]

From: Bo Yang <struggleyb.nku@gmail.com>

Use fill_metainfo to fill the line level diff meta data,
emit_line to print out a line and quote_two to quote
paths.

Signed-off-by: Bo Yang <struggleyb.nku@gmail.com>
Signed-off-by: Thomas Rast <trast@student.ethz.ch>
---
 diff.c |  6 +++---
 diff.h | 17 +++++++++++++++++
 2 files changed, 20 insertions(+), 3 deletions(-)

diff --git a/diff.c b/diff.c
index 77edd50..f9673f3 100644
--- a/diff.c
+++ b/diff.c
@@ -220,7 +220,7 @@ int git_diff_basic_config(const char *var, const char *value, void *cb)
 	return git_default_config(var, value, cb);
 }
 
-static char *quote_two(const char *one, const char *two)
+char *quote_two(const char *one, const char *two)
 {
 	int need_one = quote_c_style(one, NULL, NULL, 1);
 	int need_two = quote_c_style(two, NULL, NULL, 1);
@@ -410,7 +410,7 @@ static void emit_line_0(struct diff_options *o, const char *set, const char *res
 		fputc('\n', file);
 }
 
-static void emit_line(struct diff_options *o, const char *set, const char *reset,
+void emit_line(struct diff_options *o, const char *set, const char *reset,
 		      const char *line, int len)
 {
 	emit_line_0(o, set, reset, line[0], line+1, len-1);
@@ -2925,7 +2925,7 @@ static int similarity_index(struct diff_filepair *p)
 	return p->score * 100 / MAX_SCORE;
 }
 
-static void fill_metainfo(struct strbuf *msg,
+void fill_metainfo(struct strbuf *msg,
 			  const char *name,
 			  const char *other,
 			  struct diff_filespec *one,
diff --git a/diff.h b/diff.h
index e027650..4a7d085 100644
--- a/diff.h
+++ b/diff.h
@@ -14,6 +14,7 @@
 struct userdiff_driver;
 struct sha1_array;
 struct commit;
+struct diff_filepair;
 
 typedef void (*change_fn_t)(struct diff_options *options,
 		 unsigned old_mode, unsigned new_mode,
@@ -329,6 +330,22 @@ extern size_t fill_textconv(struct userdiff_driver *driver,
 
 extern int parse_rename_score(const char **cp_p);
 
+/* some output functions line.c need */
+extern void fill_metainfo(struct strbuf *msg,
+			  const char *name,
+			  const char *other,
+			  struct diff_filespec *one,
+			  struct diff_filespec *two,
+			  struct diff_options *o,
+			  struct diff_filepair *p,
+			  int *must_show_header,
+			  int use_color);
+
+extern void emit_line(struct diff_options *o, const char *set, const char *reset,
+		      const char *line, int len);
+
+extern char *quote_two(const char *one, const char *two);
+
 extern int print_stat_summary(FILE *fp, int files,
 			      int insertions, int deletions);
 
-- 
1.7.11.rc1.243.gbf4713c

[PATCH v7 4/5] Export rewrite_parents() for 'log -L'

[Thomas Rast]

From: Bo Yang <struggleyb.nku@gmail.com>

The function rewrite_one is used to rewrite a single
parent of the current commit, and is used by rewrite_parents
to rewrite all the parents.

Decouple the dependence between them by making rewrite_one
a callback function that is passed to rewrite_parents. Then
export rewrite_parents for reuse by the line history browser.

We will use this function in line.c.

Signed-off-by: Bo Yang <struggleyb.nku@gmail.com>
Signed-off-by: Thomas Rast <trast@student.ethz.ch>
---
 revision.c | 13 ++++---------
 revision.h | 10 ++++++++++
 2 files changed, 14 insertions(+), 9 deletions(-)

diff --git a/revision.c b/revision.c
index 935e7a7..183ca58 100644
--- a/revision.c
+++ b/revision.c
@@ -2109,12 +2109,6 @@ int prepare_revision_walk(struct rev_info *revs)
 	return 0;
 }
 
-enum rewrite_result {
-	rewrite_one_ok,
-	rewrite_one_noparents,
-	rewrite_one_error
-};
-
 static enum rewrite_result rewrite_one(struct rev_info *revs, struct commit **pp)
 {
 	struct commit_list *cache = NULL;
@@ -2136,12 +2130,13 @@ static enum rewrite_result rewrite_one(struct rev_info *revs, struct commit **pp
 	}
 }
 
-static int rewrite_parents(struct rev_info *revs, struct commit *commit)
+int rewrite_parents(struct rev_info *revs, struct commit *commit,
+	rewrite_parent_fn_t rewrite_parent)
 {
 	struct commit_list **pp = &commit->parents;
 	while (*pp) {
 		struct commit_list *parent = *pp;
-		switch (rewrite_one(revs, &parent->item)) {
+		switch (rewrite_parent(revs, &parent->item)) {
 		case rewrite_one_ok:
 			break;
 		case rewrite_one_noparents:
@@ -2213,7 +2208,7 @@ enum commit_action simplify_commit(struct rev_info *revs, struct commit *commit)
 	if (action == commit_show &&
 	    !revs->show_all &&
 	    revs->prune && revs->dense && want_ancestry(revs)) {
-		if (rewrite_parents(revs, commit) < 0)
+		if (rewrite_parents(revs, commit, rewrite_one) < 0)
 			return commit_error;
 	}
 	return action;
diff --git a/revision.h b/revision.h
index 863f4f6..6d09550 100644
--- a/revision.h
+++ b/revision.h
@@ -231,4 +231,14 @@ enum commit_action {
 extern enum commit_action get_commit_action(struct rev_info *revs, struct commit *commit);
 extern enum commit_action simplify_commit(struct rev_info *revs, struct commit *commit);
 
+enum rewrite_result {
+	rewrite_one_ok,
+	rewrite_one_noparents,
+	rewrite_one_error
+};
+
+typedef enum rewrite_result (*rewrite_parent_fn_t)(struct rev_info *revs, struct commit **pp);
+
+extern int rewrite_parents(struct rev_info *revs, struct commit *commit,
+	rewrite_parent_fn_t rewrite_parent);
 #endif
-- 
1.7.11.rc1.243.gbf4713c

[PATCH v7 5/5] Implement line-history search (git log -L)

[Thomas Rast]

This is a rewrite of much of Bo's work, mainly in an effort to split
it into smaller, easier to understand routines.

The algorithm is built around the struct range_set, which encodes a
series of line ranges as intervals [a,b).  This is used in two
contexts:

* A set of lines we are tracking (which will change as we dig through
  history).
* To encode diffs, as pairs of ranges.

The main routine is range_set_map_across_diff().  It processes the
diff between a commit C and some parent P.  It determines which diff
hunks are relevant to the ranges tracked in C, and computes the new
ranges for P.

The algorithm is then simply to process history in topological order
from newest to oldest, computing ranges and (partial) diffs.  At
branch points, we need to merge the ranges we are watching.  We will
find that many commits do not affect the chosen ranges, and mark them
TREESAME (in addition to those already filtered by pathspec limiting).
Another pass of history simplification then gets rid of such commits.

This is wired as an extra filtering pass in the log machinery.  This
currently only reduces code duplication, but should allow for other
simplifications and options to be used.

Finally, we hook a diff printer into the output chain.  Ideally we
would wire directly into the diff logic, to optionally use features
like word diff.  However, that will require some major reworking of
the diff chain, so we completely replace the output with our own diff
for now.

Signed-off-by: Bo Yang <struggleyb.nku@gmail.com>
Signed-off-by: Thomas Rast <trast@student.ethz.ch>
---
 Documentation/git-log.txt |   22 +
 builtin/log.c             |   53 +++
 line.c                    | 1139 ++++++++++++++++++++++++++++++++++++++++++++-
 line.h                    |   56 +++
 log-tree.c                |    3 +
 revision.c                |    9 +
 revision.h                |    6 +-
 7 files changed, 1286 insertions(+), 2 deletions(-)

diff --git a/Documentation/git-log.txt b/Documentation/git-log.txt
index 1f90620..9104a6d 100644
--- a/Documentation/git-log.txt
+++ b/Documentation/git-log.txt
@@ -68,6 +68,23 @@ produced by --stat etc.
 	Note that only message is considered, if also a diff is shown
 	its size is not included.
 
+-L <start>,<end>:<file>::
+	Trace the evolution of the line range given by "<start>,<end>"
+	within the <file>.  You may not give any pathspec limiters.
+	This is currently limited to a walk starting from a single
+	revision, i.e., you may only give zero or one positive
+	revision arguments.
+
+<start> and <end> can take one of these forms:
+
+include::line-range-format.txt[]
+You can specify this option more than once.
+
+
+--full-line-diff::
+	Always print the interesting range even if the current commit
+	does not change any line of the range.
+
 [\--] <path>...::
 	Show only commits that are enough to explain how the files
 	that match the specified paths came to be.  See "History
@@ -137,6 +154,11 @@ Examples
 	This makes sense only when following a strict policy of merging all
 	topic branches when staying on a single integration branch.
 
+git log -L '/int main/',/^}/:main.c::
+
+	Shows how the function `main()` in the file 'main.c' evolved
+	over time.
+
 
 Discussion
 ----------
diff --git a/builtin/log.c b/builtin/log.c
index 906dca4..4a0d5da 100644
--- a/builtin/log.c
+++ b/builtin/log.c
@@ -19,6 +19,7 @@
 #include "remote.h"
 #include "string-list.h"
 #include "parse-options.h"
+#include "line.h"
 #include "branch.h"
 #include "streaming.h"
 
@@ -38,6 +39,12 @@
 	NULL
 };
 
+struct line_opt_callback_data {
+	struct rev_info *rev;
+	const char *prefix;
+	struct line_log_data *ranges, *cur_range;
+};
+
 static int parse_decoration_style(const char *var, const char *value)
 {
 	switch (git_config_maybe_bool(var, value)) {
@@ -72,6 +79,40 @@ static int decorate_callback(const struct option *opt, const char *arg, int unse
 	return 0;
 }
 
+static int log_line_range_callback(const struct option *option, const char *arg, int unset)
+{
+	struct line_opt_callback_data *data = option->value;
+	struct line_log_data *r;
+	const char *name_start, *range_arg, *full_path;
+	const char *prefix = data->prefix;
+
+	if (!arg)
+		return -1;
+
+	name_start = skip_range_arg(arg);
+	if (!name_start || *name_start != ':')
+		die("-L argument '%s' not of the form start,end:file", arg);
+
+	range_arg = xstrndup(arg, name_start-arg);
+	name_start++;
+
+	full_path = prefix_path(prefix, prefix ? strlen(prefix) : 0,
+				name_start);
+
+	r = xmalloc(sizeof(struct line_log_data));
+	line_log_data_init(r);
+	if (data->cur_range)
+		data->cur_range->next = r;
+	else
+		data->ranges = r;
+	data->cur_range = r;
+	r->spec = alloc_filespec(full_path);
+	ALLOC_GROW(r->args, r->arg_nr+1, r->arg_alloc);
+	r->args[r->arg_nr++] = range_arg;
+	data->rev->line_level_traverse = 1;
+	return 0;
+}
+
 static void cmd_log_init_defaults(struct rev_info *rev)
 {
 	if (fmt_pretty)
@@ -94,15 +135,23 @@ static void cmd_log_init_finish(int argc, const char **argv, const char *prefix,
 {
 	struct userformat_want w;
 	int quiet = 0, source = 0;
+	static struct line_opt_callback_data line_cb = {0};
+	static int full_line_diff;
 
 	const struct option builtin_log_options[] = {
 		OPT_BOOLEAN(0, "quiet", &quiet, "suppress diff output"),
 		OPT_BOOLEAN(0, "source", &source, "show source"),
 		{ OPTION_CALLBACK, 0, "decorate", NULL, NULL, "decorate options",
 		  PARSE_OPT_OPTARG, decorate_callback},
+		OPT_CALLBACK('L', NULL, &line_cb, "n,m:file",
+			     "Process line range n,m in file, counting from 1",
+			     log_line_range_callback),
 		OPT_END()
 	};
 
+	line_cb.rev = rev;
+	line_cb.prefix = prefix;
+
 	argc = parse_options(argc, argv, prefix,
 			     builtin_log_options, builtin_log_usage,
 			     PARSE_OPT_KEEP_ARGV0 | PARSE_OPT_KEEP_UNKNOWN |
@@ -150,6 +199,10 @@ static void cmd_log_init_finish(int argc, const char **argv, const char *prefix,
 		rev->show_decorations = 1;
 		load_ref_decorations(decoration_style);
 	}
+
+	if (rev->line_level_traverse)
+		line_log_init(rev, line_cb.ranges);
+
 	setup_pager();
 }
 
diff --git a/line.c b/line.c
index a7f33ed..d837bb3 100644
--- a/line.c
+++ b/line.c
@@ -1,6 +1,459 @@
 #include "git-compat-util.h"
+#include "cache.h"
+#include "tag.h"
+#include "blob.h"
+#include "tree.h"
+#include "diff.h"
+#include "commit.h"
+#include "decorate.h"
+#include "revision.h"
+#include "xdiff-interface.h"
+#include "strbuf.h"
+#include "log-tree.h"
+#include "graph.h"
 #include "line.h"
 
+void range_set_grow (struct range_set *rs, size_t extra)
+{
+	ALLOC_GROW(rs->ranges, rs->nr + extra, rs->alloc);
+}
+
+/* Either initialization would be fine */
+#define RANGE_SET_INIT {0}
+
+void range_set_init (struct range_set *rs, size_t prealloc)
+{
+	rs->alloc = rs->nr = 0;
+	rs->ranges = NULL;
+	if (prealloc)
+		range_set_grow(rs, prealloc);
+}
+
+void range_set_release (struct range_set *rs)
+{
+	free(rs->ranges);
+	rs->alloc = rs->nr = 0;
+	rs->ranges = NULL;
+}
+
+/* dst must be uninitialized! */
+void range_set_copy (struct range_set *dst, struct range_set *src)
+{
+	range_set_init(dst, src->nr);
+	memcpy(dst->ranges, src->ranges, src->nr*sizeof(struct range_set));
+	dst->nr = src->nr;
+}
+void range_set_move (struct range_set *dst, struct range_set *src)
+{
+	range_set_release(dst);
+	dst->ranges = src->ranges;
+	dst->nr = src->nr;
+	dst->alloc = src->alloc;
+	src->ranges = NULL;
+	src->alloc = src->nr = 0;
+}
+
+/* tack on a _new_ range _at the end_ */
+void range_set_append (struct range_set *rs, long a, long b)
+{
+	assert(a <= b);
+	assert(rs->nr == 0 || rs->ranges[rs->nr-1].end <= a);
+	range_set_grow(rs, 1);
+	rs->ranges[rs->nr].start = a;
+	rs->ranges[rs->nr].end = b;
+	rs->nr++;
+}
+
+static int range_cmp (const void *_r, const void *_s)
+{
+	const struct range *r = _r;
+	const struct range *s = _s;
+
+	/* this could be simply 'return r.start-s.start', but for the types */
+	if (r->start == s->start)
+		return 0;
+	if (r->start < s->start)
+		return -1;
+	return 1;
+}
+
+/*
+ * Helper: In-place pass of sorting and merging the ranges in the
+ * range set, to re-establish the invariants after another operation
+ *
+ * NEEDSWORK currently not needed
+ */
+static void sort_and_merge_range_set (struct range_set *rs)
+{
+	int i;
+	int o = 1; /* output cursor */
+
+	qsort(rs->ranges, rs->nr, sizeof(struct range), range_cmp);
+
+	for (i = 1; i < rs->nr; i++) {
+		if (rs->ranges[i].start <= rs->ranges[o-1].end) {
+			rs->ranges[o-1].end = rs->ranges[i].end;
+		} else {
+			rs->ranges[o].start = rs->ranges[i].start;
+			rs->ranges[o].end = rs->ranges[i].end;
+			o++;
+		}
+	}
+	assert(o <= rs->nr);
+	rs->nr = o;
+}
+
+/*
+ * Union of range sets (i.e., sets of line numbers).  Used to merge
+ * them when searches meet at a common ancestor.
+ */
+static void range_set_union (struct range_set *out,
+			     struct range_set *a, struct range_set *b)
+{
+	int i = 0, j = 0, o = 0;
+	struct range *ra = a->ranges;
+	struct range *rb = b->ranges;
+	/* cannot make an alias of out->ranges: it may change during grow */
+
+	assert(out->nr == 0);
+	while (i < a->nr || j < b->nr) {
+		struct range *new;
+		if (i < a->nr && j < b->nr) {
+			if (ra[i].start < rb[j].start)
+				new = &ra[i++];
+			else if (ra[i].start > rb[j].start)
+				new = &rb[j++];
+			else if (ra[i].end < rb[j].end)
+				new = &ra[i++];
+			else
+				new = &rb[j++];
+		} else if (i < a->nr)        /* b exhausted */
+			new = &ra[i++];
+		else                       /* a exhausted */
+			new = &rb[j++];
+		if (!o || out->ranges[o-1].end < new->start) {
+			range_set_grow(out, 1);
+			out->ranges[o].start = new->start;
+			out->ranges[o].end = new->end;
+			o++;
+		} else if (out->ranges[o-1].end < new->end) {
+			out->ranges[o-1].end = new->end;
+		}
+	}
+	out->nr = o;
+}
+
+/*
+ * Difference of range sets (out = a \ b).  Pass the "interesting"
+ * ranges as 'a' and the target side of the diff as 'b': it removes
+ * the ranges for which the commit is responsible.
+ */
+static void range_set_difference (struct range_set *out,
+				  struct range_set *a, struct range_set *b)
+{
+	int i, j =  0;
+	for (i = 0; i < a->nr; i++) {
+		long start = a->ranges[i].start;
+		long end = a->ranges[i].end;
+		while (start < end) {
+			while (j < b->nr && start >= b->ranges[j].end)
+				/*
+				 * a:         |-------
+				 * b: ------|
+				 */
+				j++;
+			if (j >= b->nr || end < b->ranges[j].start) {
+				/*
+				 * b exhausted, or
+				 * a:  ----|
+				 * b:         |----
+				 */
+				range_set_append(out, start, end);
+				break;
+			}
+			if (start >= b->ranges[j].start) {
+				/*
+				 * a:     |--????
+				 * b: |------|
+				 */
+				start = b->ranges[j].end;
+			} else if (end > b->ranges[j].start) {
+				/*
+				 * a: |-----|
+				 * b:    |--?????
+				 */
+				if (start < b->ranges[j].start)
+					range_set_append(out, start, b->ranges[j].start);
+				start = b->ranges[j].end;
+			}
+		}
+	}
+}
+
+static void diff_ranges_init (struct diff_ranges *diff)
+{
+	range_set_init(&diff->parent, 0);
+	range_set_init(&diff->target, 0);
+}
+
+static void diff_ranges_release (struct diff_ranges *diff)
+{
+	range_set_release(&diff->parent);
+	range_set_release(&diff->target);
+}
+
+void line_log_data_init(struct line_log_data *r)
+{
+	memset(r, 0, sizeof(struct line_log_data));
+	range_set_init(&r->ranges, 0);
+}
+
+static void line_log_data_clear(struct line_log_data *r)
+{
+	range_set_release(&r->ranges);
+}
+
+static void free_line_log_datas(struct line_log_data *r)
+{
+	while (r) {
+		struct line_log_data *next = r->next;
+		line_log_data_clear(r);
+		free(r);
+		r = next;
+	}
+}
+
+static struct line_log_data *
+search_line_log_data_1(struct line_log_data *list, const char *path,
+			 struct line_log_data **insertion_point)
+{
+	struct line_log_data *p = list;
+	while (p) {
+		int cmp = strcmp(p->spec->path, path);
+		if (!cmp)
+			return p;
+		if (insertion_point && cmp < 0)
+			*insertion_point = p;
+		p = p->next;
+	}
+	return NULL;
+}
+
+static struct line_log_data *
+search_line_log_data(struct line_log_data *list, const char *path)
+{
+	return search_line_log_data_1(list, path, NULL);
+}
+
+static void line_log_data_extend (struct line_log_data **list,
+				    const char *path, struct range_set *rs)
+{
+	struct line_log_data *ip;
+	struct line_log_data *p = search_line_log_data_1(*list, path, &ip);
+
+	if (p) {
+		int i;
+		for (i = 0; i < rs->nr; i++)
+			range_set_append(&p->ranges, rs->ranges[i].start,
+					 rs->ranges[i].end);
+		return;
+	}
+
+	p = xcalloc(1, sizeof(struct line_log_data));
+	range_set_copy(&p->ranges, rs);
+	if (ip) {
+		p->next = ip->next;
+		ip->next = p;
+	} else {
+		p->next = *list;
+		*list = p;
+	}
+}
+
+static void line_log_data_insert (struct line_log_data **list,
+				    const char *path, struct range rg)
+{
+	/* we fake a 1-element range_set without any allocations */
+	struct range_set rs = { 1, 1, &rg };
+	line_log_data_extend(list, path, &rs);
+}
+
+static void line_log_data_sort (struct line_log_data *list)
+{
+	struct line_log_data *p = list;
+	while (p) {
+		sort_and_merge_range_set(&p->ranges);
+		p = p->next;
+	}
+}
+
+/* In the diff handling, p=parent and t=target. */
+
+struct collect_diff_cbdata {
+	long plno, tlno;
+	struct diff_ranges *diff;
+};
+
+/*
+ * This callback being called means:
+ * - lines [tlno,same] are from parent
+ * - line tlno in target corresponds to plno in parent
+ */
+static int collect_diff_cb (long start_a, long count_a,
+			    long start_b, long count_b,
+			    void *data)
+{
+	struct collect_diff_cbdata *d = data;
+
+	if (count_a >= 0)
+		range_set_append(&d->diff->parent, start_a, start_a + count_a);
+	if (count_b >= 0)
+		range_set_append(&d->diff->target, start_b, start_b + count_b);
+
+	d->plno = start_a + count_a;
+	d->tlno = start_b + count_b;
+
+	return 0;
+}
+
+static void collect_diff (mmfile_t *parent, mmfile_t *target, struct diff_ranges *out)
+{
+	struct collect_diff_cbdata cbdata = {0};
+	xpparam_t xpp;
+	xdemitconf_t xecfg;
+	xdemitcb_t ecb;
+
+	memset(&xpp, 0, sizeof(xpp));
+	memset(&xecfg, 0, sizeof(xecfg));
+	xecfg.ctxlen = xecfg.interhunkctxlen = 0;
+
+	cbdata.diff = out;
+	xecfg.hunk_func = collect_diff_cb;
+	memset(&ecb, 0, sizeof(ecb));
+	ecb.priv = &cbdata;
+	xdi_diff(parent, target, &xpp, &xecfg, &ecb);
+}
+
+static void dump_range_set (struct range_set *rs, const char *desc)
+{
+	int i;
+	printf("range set %s (%d items):\n", desc, rs->nr);
+	for (i = 0; i < rs->nr; i++)
+		printf("\t[%ld,%ld]\n", rs->ranges[i].start, rs->ranges[i].end);
+}
+
+static void dump_line_log_datas (struct line_log_data *r)
+{
+	char buf[4096];
+	while (r) {
+		snprintf(buf, 4096, "file %s\n", r->spec->path);
+		dump_range_set(&r->ranges, buf);
+		r = r->next;
+	}
+}
+
+static void dump_diff_ranges (struct diff_ranges *diff, const char *desc)
+{
+	int i;
+	assert(diff->parent.nr == diff->target.nr);
+	printf("diff ranges %s (%d items):\n", desc, diff->parent.nr);
+	printf("\tparent\ttarget\n");
+	for (i = 0; i < diff->parent.nr; i++) {
+		printf("\t[%ld,%ld]\t[%ld,%ld]\n",
+		       diff->parent.ranges[i].start,
+		       diff->parent.ranges[i].end,
+		       diff->target.ranges[i].start,
+		       diff->target.ranges[i].end);
+	}
+}
+
+
+static int ranges_overlap (struct range *a, struct range *b)
+{
+	return !(a->end <= b->start || b->end <= a->start);
+}
+
+/*
+ * Given a diff and the set of interesting ranges, determine all hunks
+ * of the diff which touch (overlap) at least one of the interesting
+ * ranges in the target.
+ */
+static void diff_ranges_filter_touched (struct diff_ranges *out,
+					struct diff_ranges *diff,
+					struct range_set *rs)
+{
+	int i, j = 0;
+
+	assert(out->target.nr == 0);
+
+	for (i = 0; i < diff->target.nr; i++) {
+		while (diff->target.ranges[i].start > rs->ranges[j].end) {
+			j++;
+			if (j == rs->nr)
+				return;
+		}
+		if (ranges_overlap(&diff->target.ranges[i], &rs->ranges[j])) {
+			range_set_append(&out->parent,
+					 diff->parent.ranges[i].start,
+					 diff->parent.ranges[i].end);
+			range_set_append(&out->target,
+					 diff->target.ranges[i].start,
+					 diff->target.ranges[i].end);
+		}
+	}
+}
+
+/*
+ * Adjust the line counts in 'rs' to account for the lines
+ * added/removed in the diff.
+ */
+static void range_set_shift_diff (struct range_set *out,
+				  struct range_set *rs,
+				  struct diff_ranges *diff)
+{
+	int i, j = 0;
+	long offset = 0;
+	struct range *src = rs->ranges;
+	struct range *target = diff->target.ranges;
+	struct range *parent = diff->parent.ranges;
+
+	for (i = 0; i < rs->nr; i++) {
+		while (j < diff->target.nr && src[i].start >= target[j].start) {
+			offset += (parent[j].end-parent[j].start)
+				- (target[j].end-target[j].start);
+			j++;
+		}
+		range_set_append(out, src[i].start+offset, src[i].end+offset);
+	}
+}
+
+/*
+ * Given a diff and the set of interesting ranges, map the ranges
+ * across the diff.  That is: observe that the target commit takes
+ * blame for all the + (target-side) ranges.  So for every pair of
+ * ranges in the diff that was touched, we remove the latter and add
+ * its parent side.
+ */
+static void range_set_map_across_diff (struct range_set *out,
+				       struct range_set *rs,
+				       struct diff_ranges *diff,
+				       struct diff_ranges **touched_out)
+{
+	struct diff_ranges *touched = xmalloc(sizeof(*touched));
+	struct range_set tmp1 = RANGE_SET_INIT;
+	struct range_set tmp2 = RANGE_SET_INIT;
+
+	diff_ranges_init(touched);
+	diff_ranges_filter_touched(touched, diff, rs);
+	range_set_difference(&tmp1, rs, &touched->target);
+	range_set_shift_diff(&tmp2, &tmp1, diff);
+	range_set_union(out, &tmp2, &touched->parent);
+	range_set_release(&tmp1);
+	range_set_release(&tmp2);
+
+	*touched_out = touched;
+}
+
 /*
  * Parse one item in the -L option
  */
@@ -30,6 +483,8 @@ const char *parse_loc(const char *spec, nth_line_fn_t nth_line,
 	if (begin != -1 && (spec[0] == '+' || spec[0] == '-')) {
 		num = strtol(spec + 1, &term, 10);
 		if (term != spec + 1) {
+			if (!ret)
+				return term;
 			if (spec[0] == '-')
 				num = 0 - num;
 			if (0 < num)
@@ -44,7 +499,8 @@ const char *parse_loc(const char *spec, nth_line_fn_t nth_line,
 	}
 	num = strtol(spec, &term, 10);
 	if (term != spec) {
-		*ret = num;
+		if (ret)
+			*ret = num;
 		return term;
 	}
 	if (spec[0] != '/')
@@ -58,6 +514,10 @@ const char *parse_loc(const char *spec, nth_line_fn_t nth_line,
 	if (*term != '/')
 		return spec;
 
+	/* in the scan-only case we are not interested in the regex */
+	if (!ret)
+		return term+1;
+
 	/* try [spec+1 .. term-1] as regexp */
 	*term = 0;
 	if (begin == -1)
@@ -106,3 +566,680 @@ int parse_range_arg(const char *arg, nth_line_fn_t nth_line_cb,
 
 	return 0;
 }
+
+const char *skip_range_arg(const char *arg)
+{
+	arg = parse_loc(arg, NULL, NULL, 0, -1, 0);
+
+	if (*arg == ',')
+		arg = parse_loc(arg+1, NULL, NULL, 0, 0, 0);
+
+	return arg;
+}
+
+static struct commit *check_single_commit(struct rev_info *revs)
+{
+	struct object *commit = NULL;
+	int found = -1;
+	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,
+				revs->pending.objects[found].name);
+		commit = obj;
+		found = i;
+	}
+
+	if (!commit)
+		die("No commit specified?");
+
+	return (struct commit *) commit;
+}
+
+static void fill_blob_sha1(struct commit *commit, struct line_log_data *r)
+{
+	unsigned mode;
+	unsigned char sha1[20];
+
+	while (r) {
+		if (get_tree_entry(commit->object.sha1, r->spec->path,
+			sha1, &mode))
+			die("There is no path %s in the commit", r->spec->path);
+		fill_filespec(r->spec, sha1, mode);
+		r = r->next;
+	}
+
+	return;
+}
+
+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++] = 0;
+	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;
+}
+
+struct nth_line_cb {
+	struct diff_filespec *spec;
+	long lines;
+	unsigned long *line_ends;
+};
+
+static const char *nth_line(void *data, long line)
+{
+	struct nth_line_cb *d = data;
+	assert(d && line < d->lines);
+	assert(d->spec && d->spec->data);
+
+	if (line == 0)
+		return (char *)d->spec->data;
+	else
+		return (char *)d->spec->data + d->line_ends[line] + 1;
+}
+
+static void parse_lines(struct commit *commit, struct line_log_data *r)
+{
+	int i;
+	long lines = 0;
+	unsigned long *ends = NULL;
+	struct nth_line_cb cb_data;
+
+	while (r) {
+		struct diff_filespec *spec = r->spec;
+		int num = r->arg_nr;
+		assert(spec);
+		fill_blob_sha1(commit, r);
+		fill_line_ends(spec, &lines, &ends);
+		cb_data.spec = spec;
+		cb_data.lines = lines;
+		cb_data.line_ends = ends;
+		for (i = 0; i < num; i++) {
+			long begin, end;
+			struct range rg;
+			if (parse_range_arg(r->args[i], nth_line, &cb_data,
+					    lines-1, &begin, &end))
+				die("malformed -L argument '%s'", r->args[i]);
+			rg.start = begin-1;
+			rg.end = end;
+			line_log_data_insert(&r, r->spec->path, rg);
+		}
+
+		free(ends);
+		ends = NULL;
+
+		r = r->next;
+	}
+}
+
+static struct line_log_data *line_log_data_copy_one(struct line_log_data *r)
+{
+	struct line_log_data *ret = xmalloc(sizeof(*ret));
+
+	assert(r);
+	line_log_data_init(ret);
+	range_set_copy(&ret->ranges, &r->ranges);
+
+	ret->spec = r->spec;
+	assert(ret->spec);
+	ret->spec->count++;
+
+	return ret;
+}
+
+static struct line_log_data *
+line_log_data_copy(struct line_log_data *r)
+{
+	struct line_log_data *ret = NULL;
+	struct line_log_data *tmp = NULL, *prev = NULL;
+
+	assert(r);
+	ret = tmp = prev = line_log_data_copy_one(r);
+	r = r->next;
+	while (r) {
+		tmp = line_log_data_copy_one(r);
+		prev->next = tmp;
+		prev = tmp;
+		r = r->next;
+	}
+
+	return ret;
+}
+
+/* merge two range sets across files */
+static struct line_log_data *line_log_data_merge(struct line_log_data *a,
+		struct line_log_data *b)
+{
+	struct line_log_data *o, *prev = NULL, *head;
+
+	if (!a && !b)
+		return NULL;
+
+	while (a || b) {
+		struct line_log_data *src;
+		struct line_log_data *src2 = NULL;
+		int cmp;
+		if (!a)
+			cmp = 1;
+		else if (!b)
+			cmp = -1;
+		else
+			cmp = strcmp(a->spec->path, b->spec->path);
+		if (cmp < 0) {
+			src = a;
+			a = a->next;
+		} else if (cmp == 0) {
+			src = a;
+			a = a->next;
+			src2 = b;
+			b = b->next;
+		} else {
+			src = b;
+			b = b->next;
+		}
+		o = xmalloc(sizeof(struct line_log_data));
+		line_log_data_init(o);
+		o->spec = src->spec;
+		o->spec->count++;
+		if (prev)
+			prev->next = o;
+		else
+			head = o;
+		prev = o;
+		if (src2)
+			range_set_union(&o->ranges, &src->ranges, &src2->ranges);
+		else
+			range_set_copy(&o->ranges, &src->ranges);
+	}
+
+	return head;
+}
+
+static void add_line_range(struct rev_info *revs, struct commit *commit,
+		struct line_log_data *range)
+{
+	struct line_log_data *old = NULL;
+	struct line_log_data *new = NULL;
+
+	old = lookup_decoration(&revs->line_log_data, &commit->object);
+	if (old && range) {
+		new = line_log_data_merge(old, range);
+		free_line_log_datas(old);
+	} else if (range)
+		new = line_log_data_copy(range);
+
+	if (new)
+		add_decoration(&revs->line_log_data, &commit->object, new);
+}
+
+static void clear_commit_line_range(struct rev_info *revs, struct commit *commit)
+{
+	struct line_log_data *r;
+	r = lookup_decoration(&revs->line_log_data, &commit->object);
+	if (!r)
+		return;
+	free_line_log_datas(r);
+	add_decoration(&revs->line_log_data, &commit->object, NULL);
+}
+
+static struct line_log_data *lookup_line_range(struct rev_info *revs,
+		struct commit *commit)
+{
+	struct line_log_data *ret = NULL;
+
+	ret = lookup_decoration(&revs->line_log_data, &commit->object);
+	return ret;
+}
+
+void line_log_init(struct rev_info *rev, struct line_log_data *r)
+{
+	struct commit *commit = NULL;
+
+	commit = check_single_commit(rev);
+	parse_lines(commit, r);
+
+	add_line_range(rev, commit, r);
+}
+
+static void load_tree_desc(struct tree_desc *desc, void **tree,
+		const unsigned char *sha1)
+{
+	unsigned long size;
+	*tree = read_object_with_reference(sha1, tree_type, &size, NULL);
+	if (!tree)
+		die("Unable to read tree (%s)", sha1_to_hex(sha1));
+	init_tree_desc(desc, *tree, size);
+}
+
+static int count_parents(struct commit *commit)
+{
+	struct commit_list *parents = commit->parents;
+	int count = 0;
+	while (parents) {
+		count++;
+		parents = parents->next;
+	}
+	return count;
+}
+
+static void move_diff_queue(struct diff_queue_struct *dst,
+			    struct diff_queue_struct *src)
+{
+	assert(src != dst);
+	memcpy(dst, src, sizeof(struct diff_queue_struct));
+	DIFF_QUEUE_CLEAR(src);
+}
+
+static void queue_diffs(struct diff_options *opt,
+			struct diff_queue_struct *queue,
+			struct commit *commit, struct commit *parent)
+{
+	void *tree1 = NULL, *tree2 = NULL;
+	struct tree_desc desc1, desc2;
+
+	/*
+	 * Compose up two trees, for root commit, we make up a empty tree.
+	 */
+	assert(commit);
+	load_tree_desc(&desc2, &tree2, commit->tree->object.sha1);
+	if (parent) {
+		load_tree_desc(&desc1, &tree1, parent->tree->object.sha1);
+	} else {
+		init_tree_desc(&desc1, "", 0);
+	}
+
+	DIFF_QUEUE_CLEAR(&diff_queued_diff);
+	diff_tree(&desc1, &desc2, "", opt);
+	diffcore_std(opt);
+	move_diff_queue(queue, &diff_queued_diff);
+
+	if (tree1)
+		free(tree1);
+	if (tree2)
+		free(tree2);
+}
+
+static char *get_nth_line(long line, unsigned long *ends, void *data)
+{
+	if (line == 0)
+		return (char *)data;
+	else
+		return (char *)data + ends[line] + 1;
+}
+
+static void print_line(const char *prefix, char first,
+		       long line, unsigned long *ends, void *data,
+		       const char *color, const char *reset)
+{
+	char *begin = get_nth_line(line, ends, data);
+	char *end = get_nth_line(line+1, ends, data);
+	if (end > begin && end[-1] == '\n')
+		end--;
+
+	fputs(prefix, stdout);
+	fputs(color, stdout);
+	putchar(first);
+	fwrite(begin, 1, end-begin, stdout);
+	fputs(reset, stdout);
+	putchar('\n');
+}
+
+static char *output_prefix(struct diff_options *opt)
+{
+	char *prefix = "";
+
+	if (opt->output_prefix) {
+		struct strbuf *sb = opt->output_prefix(opt, opt->output_prefix_data);
+		prefix = sb->buf;
+	}
+
+	return prefix;
+}
+
+static void dump_diff_hacky_one(struct rev_info *rev, struct line_log_data *range)
+{
+	int i, j = 0;
+	long p_lines, t_lines;
+	unsigned long *p_ends = NULL, *t_ends = NULL;
+	struct diff_filepair *pair = range->pair;
+	struct diff_ranges *diff = &range->diff;
+
+	struct diff_options *opt = &rev->diffopt;
+	char *prefix = output_prefix(opt);
+	const char *c_reset = diff_get_color(opt->use_color, DIFF_RESET);
+	const char *c_frag = diff_get_color(opt->use_color, DIFF_FRAGINFO);
+	const char *c_meta = diff_get_color(opt->use_color, DIFF_METAINFO);
+	const char *c_old = diff_get_color(opt->use_color, DIFF_FILE_OLD);
+	const char *c_new = diff_get_color(opt->use_color, DIFF_FILE_NEW);
+	const char *c_plain = diff_get_color(opt->use_color, DIFF_PLAIN);
+
+	if (!pair || !diff)
+		return;
+
+	if (pair->one->sha1_valid)
+		fill_line_ends(pair->one, &p_lines, &p_ends);
+	fill_line_ends(pair->two, &t_lines, &t_ends);
+
+	printf("%s%sdiff --git a/%s b/%s%s\n", prefix, c_meta, pair->one->path, pair->two->path, c_reset);
+	printf("%s%s--- %s%s%s\n", prefix, c_meta,
+	       pair->one->sha1_valid ? "a/" : "",
+	       pair->one->sha1_valid ? pair->one->path : "/dev/null",
+	       c_reset);
+	printf("%s%s+++ b/%s%s\n", prefix, c_meta, pair->two->path, c_reset);
+	for (i = 0; i < range->ranges.nr; i++) {
+		long p_start, p_end;
+		long t_start = range->ranges.ranges[i].start;
+		long t_end = range->ranges.ranges[i].end;
+		long t_cur = t_start;
+		int j_last;
+
+		while (j < diff->target.nr && diff->target.ranges[j].end < t_start)
+			j++;
+		if (j == diff->target.nr || diff->target.ranges[j].start > t_end)
+			continue;
+
+		/* Scan ahead to determine the last diff that falls in this range */
+		j_last = j;
+		while (j_last < diff->target.nr && diff->target.ranges[j_last].start < t_end)
+			j_last++;
+		if (j_last > j)
+			j_last--;
+
+		/*
+		 * Compute parent hunk headers: we know that the diff
+		 * has the correct line numbers (but not all hunks).
+		 * So it suffices to shift the start/end according to
+		 * the line numbers of the first/last hunk(s) that
+		 * fall in this range.
+		 */
+		p_start = diff->parent.ranges[j].start - (diff->target.ranges[j].start-t_start);
+		p_end = diff->parent.ranges[j_last].end + (t_end-diff->target.ranges[j_last].end);
+
+		/* Now output a diff hunk for this range */
+		printf("%s%s@@ -%ld,%ld +%ld,%ld @@%s\n",
+		       prefix, c_frag,
+		       p_start+1, p_end-p_start, t_start+1, t_end-t_start,
+		       c_reset);
+		while (j < diff->target.nr && diff->target.ranges[j].start < t_end) {
+			int k;
+			for (; t_cur < diff->target.ranges[j].start; t_cur++)
+				print_line(prefix, ' ', t_cur, t_ends, pair->two->data,
+					   c_plain, c_reset);
+			for (k = diff->parent.ranges[j].start; k < diff->parent.ranges[j].end; k++)
+				print_line(prefix, '-', k, p_ends, pair->one->data,
+					   c_old, c_reset);
+			for (; t_cur < diff->target.ranges[j].end; t_cur++)
+				print_line(prefix, '+', t_cur, t_ends, pair->two->data,
+					   c_new, c_reset);
+			j++;
+		}
+		for (; t_cur < t_end; t_cur++)
+			print_line(prefix, ' ', t_cur, t_ends, pair->two->data,
+				   c_plain, c_reset);
+	}
+}
+
+static void dump_diff_hacky(struct rev_info *rev, struct line_log_data *range)
+{
+	puts(output_prefix(&rev->diffopt));
+	while (range) {
+		dump_diff_hacky_one(rev, range);
+		range = range->next;
+	}
+}
+
+/*
+ * Unlike most other functions, this destructively operates on
+ * 'range'.
+ */
+static int process_diff_filepair(struct rev_info *rev,
+				 struct diff_filepair *pair,
+				 struct line_log_data *range,
+				 struct diff_ranges **diff_out)
+{
+	struct line_log_data *rg = range;
+	struct range_set tmp;
+	struct diff_ranges diff;
+	mmfile_t file_parent, file_target;
+
+	assert(pair->two->path);
+	while (rg) {
+		assert(rg->spec->path);
+		if (!strcmp(rg->spec->path, pair->two->path))
+			break;
+		rg = rg->next;
+	}
+
+	if (!rg)
+		return 0;
+	if (rg->ranges.nr == 0)
+		return 0;
+
+	assert(pair->two->sha1_valid);
+	diff_populate_filespec(pair->two, 0);
+	file_target.ptr = pair->two->data;
+	file_target.size = pair->two->size;
+
+	if (pair->one->sha1_valid) {
+		diff_populate_filespec(pair->one, 0);
+		file_parent.ptr = pair->one->data;
+		file_parent.size = pair->one->size;
+	} else {
+		file_parent.ptr = "";
+		file_parent.size = 0;
+	}
+
+	diff_ranges_init(&diff);
+	collect_diff(&file_parent, &file_target, &diff);
+
+	range_set_init(&tmp, 0);
+	range_set_map_across_diff(&tmp, &rg->ranges, &diff, diff_out);
+	range_set_release(&rg->ranges);
+	range_set_move(&rg->ranges, &tmp);
+
+	return ((*diff_out)->parent.nr > 0);
+}
+
+static int process_all_files(struct line_log_data **range_out,
+			     struct rev_info *rev,
+			     struct diff_queue_struct *queue,
+			     struct line_log_data *range)
+{
+	int i, changed = 0;
+
+	*range_out = line_log_data_copy(range);
+
+	for (i = 0; i < queue->nr; i++) {
+		struct diff_ranges *pairdiff = NULL;
+		if (process_diff_filepair(rev, queue->queue[i], *range_out, &pairdiff)) {
+			struct line_log_data *rg = range;
+			changed++;
+			/* NEEDSWORK tramples over data structures not owned here */
+			while (rg && strcmp(rg->spec->path, queue->queue[i]->two->path))
+				rg = rg->next;
+			assert(rg);
+			rg->pair = queue->queue[i];
+			memcpy(&rg->diff, pairdiff, sizeof(struct diff_ranges));
+		}
+	}
+
+	return changed;
+}
+
+int line_log_print(struct rev_info *rev, struct commit *commit)
+{
+       struct line_log_data *range = lookup_line_range(rev, commit);
+
+       show_log(rev);
+       dump_diff_hacky(rev, range);
+       return 1;
+}
+
+static int process_ranges_ordinary_commit(struct rev_info *rev, struct commit *commit)
+{
+	struct commit *parent = NULL;
+	struct diff_queue_struct queue;
+	struct line_log_data *range = lookup_line_range(rev, commit);
+	struct line_log_data *parent_range;
+	int changed;
+
+	if (commit->parents)
+		parent = commit->parents->item;
+
+	queue_diffs(&rev->diffopt, &queue, commit, parent);
+	changed = process_all_files(&parent_range, rev, &queue, range);
+	if (parent)
+		add_line_range(rev, parent, parent_range);
+	return changed;
+}
+
+static int process_ranges_merge_commit(struct rev_info *rev, struct commit *commit)
+{
+	struct line_log_data *range = lookup_line_range(rev, commit);
+	struct diff_queue_struct *diffqueues;
+	struct line_log_data **cand;
+	struct commit **parents;
+	struct commit_list *p;
+	int i;
+	int nparents = count_parents(commit);
+
+	diffqueues = xmalloc(nparents * sizeof(*diffqueues));
+	cand = xmalloc(nparents * sizeof(*cand));
+	parents = xmalloc(nparents * sizeof(*parents));
+
+	p = commit->parents;
+	for (i = 0; i < nparents; i++) {
+		parents[i] = p->item;
+		p = p->next;
+		queue_diffs(&rev->diffopt, &diffqueues[i], commit, parents[i]);
+	}
+
+	for (i = 0; i < nparents; i++) {
+		int changed;
+		cand[i] = NULL;
+		changed = process_all_files(&cand[i], rev, &diffqueues[i], range);
+		if (!changed) {
+			/*
+			 * This parent can take all the blame, so we
+			 * don't follow any other path in history
+			 */
+			add_line_range(rev, parents[i], cand[i]);
+			clear_commit_line_range(rev, commit);
+			commit->parents = xmalloc(sizeof(struct commit_list));
+			commit->parents->item = parents[i];
+			commit->parents->next = NULL;
+			/* NEEDSWORK leaking like a sieve */
+			return 0;
+		}
+	}
+
+	/*
+	 * No single parent took the blame.  We add the candidates
+	 * from the above loop to the parents.
+	 */
+	for (i = 0; i < nparents; i++) {
+		add_line_range(rev, parents[i], cand[i]);
+	}
+
+	clear_commit_line_range(rev, commit);
+	return 1;
+
+	/* NEEDSWORK evil merge detection stuff */
+	/* NEEDSWORK leaking like a sieve */
+}
+
+static int process_ranges_arbitrary_commit(struct rev_info *rev, struct commit *commit)
+{
+	int changed = 0;
+
+	if (lookup_line_range(rev, commit)) {
+		if (!commit->parents || !commit->parents->next)
+			changed = process_ranges_ordinary_commit(rev, commit);
+		else
+			changed = process_ranges_merge_commit(rev, commit);
+	}
+
+	if (!changed)
+		commit->object.flags |= TREESAME;
+
+	return changed;
+}
+
+static enum rewrite_result line_log_rewrite_one(struct rev_info *rev, struct commit **pp)
+{
+	for (;;) {
+		struct commit *p = *pp;
+		if (p->parents && p->parents->next)
+			return rewrite_one_ok;
+		if (p->object.flags & UNINTERESTING)
+			return rewrite_one_ok;
+		if (!(p->object.flags & TREESAME))
+			return rewrite_one_ok;
+		if (!p->parents)
+			return rewrite_one_noparents;
+		*pp = p->parents->item;
+	}
+}
+
+int line_log_filter(struct rev_info *rev)
+{
+	struct commit *commit;
+	struct commit_list *list = rev->commits;
+	struct commit_list *out = NULL, *cur = NULL;
+
+	list = rev->commits;
+	while (list) {
+		struct commit_list *to_free = NULL;
+		commit = list->item;
+		if (process_ranges_arbitrary_commit(rev, commit)) {
+			if (cur)
+				cur->next = list;
+			else
+				out = list;
+			cur = list;
+		} else
+			to_free = list;
+		list = list->next;
+		free(to_free);
+	}
+	cur->next = NULL;
+
+	list = out;
+	while (list) {
+		rewrite_parents(rev, list->item, line_log_rewrite_one);
+		list = list->next;
+	}
+
+	rev->commits = out;
+
+	return 0;
+}
diff --git a/line.h b/line.h
index 5878c94..9cbff56 100644
--- a/line.h
+++ b/line.h
@@ -1,6 +1,8 @@
 #ifndef LINE_H
 #define LINE_H
 
+#include "diffcore.h"
+
 /*
  * Parse one item in an -L begin,end option w.r.t. the notional file
  * object 'cb_data' consisting of 'lines' lines.
@@ -20,4 +22,58 @@ extern int parse_range_arg(const char *arg,
 			   void *cb_data, long lines,
 			   long *begin, long *end);
 
+/*
+ * Scan past a range argument that could be parsed by
+ * 'parse_range_arg', to help the caller determine the start of the
+ * filename in '-L n,m:file' syntax.
+ *
+ * Returns a pointer to the first character after the 'n,m' part, or
+ * NULL in case the argument is obviously malformed.
+ */
+
+extern const char *skip_range_arg(const char *arg);
+
+struct rev_info;
+struct commit;
+
+/* A range [start,end].  Lines are numbered starting at 0, and the
+ * ranges include start but exclude end. */
+struct range {
+	long start, end;
+};
+
+/* A set of ranges.  The ranges must always be disjoint and sorted. */
+struct range_set {
+	int alloc, nr;
+	struct range *ranges;
+};
+
+/* A diff, encoded as the set of pre- and post-image ranges where the
+ * files differ. A pair of ranges corresponds to a hunk. */
+struct diff_ranges {
+	struct range_set parent;
+	struct range_set target;
+};
+
+/* Linked list of interesting files and their associated ranges.  The
+ * list must be kept sorted by spec->path */
+struct line_log_data {
+	struct line_log_data *next;
+	struct diff_filespec *spec;
+	char status;
+	struct range_set ranges;
+	int arg_alloc, arg_nr;
+	char **args;
+	struct diff_filepair *pair;
+	struct diff_ranges diff;
+};
+
+extern void line_log_data_init(struct line_log_data *r);
+
+extern void line_log_init(struct rev_info *rev, struct line_log_data *r);
+
+extern int line_log_filter(struct rev_info *rev);
+
+extern int line_log_print(struct rev_info *rev, struct commit *commit);
+
 #endif /* LINE_H */
diff --git a/log-tree.c b/log-tree.c
index c894930..3fb025d 100644
--- a/log-tree.c
+++ b/log-tree.c
@@ -809,6 +809,9 @@ int log_tree_commit(struct rev_info *opt, struct commit *commit)
 	log.parent = NULL;
 	opt->loginfo = &log;
 
+	if (opt->line_level_traverse)
+		return line_log_print(opt, commit);
+
 	shown = log_tree_diff(opt, commit, &log);
 	if (!shown && opt->loginfo && opt->always_show_header) {
 		log.parent = NULL;
diff --git a/revision.c b/revision.c
index 183ca58..a92e359 100644
--- a/revision.c
+++ b/revision.c
@@ -13,6 +13,7 @@
 #include "decorate.h"
 #include "log-tree.h"
 #include "string-list.h"
+#include "line.h"
 
 volatile show_early_output_fn_t show_early_output;
 
@@ -1851,6 +1852,12 @@ int setup_revisions(int argc, const char **argv, struct rev_info *revs, struct s
 	if (revs->combine_merges)
 		revs->ignore_merges = 0;
 	revs->diffopt.abbrev = revs->abbrev;
+
+	if (revs->line_level_traverse) {
+		revs->limited = 1;
+		revs->topo_order = 1;
+	}
+
 	if (diff_setup_done(&revs->diffopt) < 0)
 		die("diff_setup_done failed");
 
@@ -2102,6 +2109,8 @@ int prepare_revision_walk(struct rev_info *revs)
 			return -1;
 	if (revs->topo_order)
 		sort_in_topological_order(&revs->commits, revs->lifo);
+	if (revs->line_level_traverse)
+		line_log_filter(revs);
 	if (revs->simplify_merges)
 		simplify_merges(revs);
 	if (revs->children.name)
diff --git a/revision.h b/revision.h
index 6d09550..01902cf 100644
--- a/revision.h
+++ b/revision.h
@@ -92,7 +92,8 @@ struct rev_info {
 			cherry_mark:1,
 			bisect:1,
 			ancestry_path:1,
-			first_parent_only:1;
+			first_parent_only:1,
+			line_level_traverse:1;
 
 	/* Diff flags */
 	unsigned int	diff:1,
@@ -168,6 +169,9 @@ struct rev_info {
 	int count_left;
 	int count_right;
 	int count_same;
+
+	/* line level range that we are chasing */
+	struct decoration line_log_data;
 };
 
 #define REV_TREE_SAME		0
-- 
1.7.11.rc1.243.gbf4713c

Re: [PATCH v7 2/5] blame: introduce $ as "end of file" in -L syntax

[Junio C Hamano]

Thomas Rast <trast@student.ethz.ch> writes:

> To save the user a lookup of the last line number, introduce $ as a
> shorthand for the last line.  This is mostly useful to spell "until
> the end of the file" as '-L<begin>,$'.

Hmph.  This is mostly useful not to error out when user (perhaps
mistakenly) expects that the end of file is spelled as "$"; both
"git blame -L<begin>" and "git blame L<begin>," have always meant
"til the end", IIRC.

Because I do not offhand think of other & better uses of "$" as a
special case that conflicts with "the end of file", I do not think
it is a bad patch that needs to be rejected, though.

Re: [PATCH v7 5/5] Implement line-history search (git log -L)

[Junio C Hamano]

Thomas Rast <trast@student.ethz.ch> writes:

> This is a rewrite of much of Bo's work, mainly in an effort to split
> it into smaller, easier to understand routines.

You mentioned "splitting" in the cover letter, but it does seem to
need a bit more work.

> diff --git a/builtin/log.c b/builtin/log.c
> index 906dca4..4a0d5da 100644
> --- a/builtin/log.c
> +++ b/builtin/log.c
> @@ -38,6 +39,12 @@
> ...
> +static int log_line_range_callback(const struct option *option, const char *arg, int unset)
> +{
> +	struct line_opt_callback_data *data = option->value;
> +	struct line_log_data *r;
> +	const char *name_start, *range_arg, *full_path;
> ...
> +	ALLOC_GROW(r->args, r->arg_nr+1, r->arg_alloc);
> +	r->args[r->arg_nr++] = range_arg;

Assignment discards qualifiers from pointer target type; this
pointer does not have to be "const char *" perhaps?

> @@ -94,15 +135,23 @@ static void cmd_log_init_finish(int argc, const char **argv, const char *prefix,
>  {
>  	struct userformat_want w;
>  	int quiet = 0, source = 0;
> +	static struct line_opt_callback_data line_cb = {0};
> +	static int full_line_diff;

Variable unused.

> diff --git a/line.c b/line.c
> index a7f33ed..d837bb3 100644
> --- a/line.c
> +++ b/line.c
> @@ -1,6 +1,459 @@
> ...
> +static void diff_ranges_release (struct diff_ranges *diff)
> +{
> +	range_set_release(&diff->parent);
> +	range_set_release(&diff->target);
> +}

Unused.

> +static struct line_log_data *
> +search_line_log_data(struct line_log_data *list, const char *path)
> +{
> +	return search_line_log_data_1(list, path, NULL);
> +}

Unused.

> +static void line_log_data_sort (struct line_log_data *list)
> +{
> +	struct line_log_data *p = list;
> +	while (p) {
> +		sort_and_merge_range_set(&p->ranges);
> +		p = p->next;
> +	}
> +}

Unused.

> +static int ranges_overlap (struct range *a, struct range *b)
> +{
> +	return !(a->end <= b->start || b->end <= a->start);
> +}
> +
> +/*
> + * Given a diff and the set of interesting ranges, determine all hunks
> + * of the diff which touch (overlap) at least one of the interesting
> + * ranges in the target.
> + */
> +static void diff_ranges_filter_touched (struct diff_ranges *out,
> +					struct diff_ranges *diff,
> +					struct range_set *rs)
> +{
> +	int i, j = 0;
> +
> +	assert(out->target.nr == 0);
> +
> +	for (i = 0; i < diff->target.nr; i++) {
> +		while (diff->target.ranges[i].start > rs->ranges[j].end) {
> +			j++;
> +			if (j == rs->nr)
> +				return;
> +		}
> +		if (ranges_overlap(&diff->target.ranges[i], &rs->ranges[j])) {
> +			range_set_append(&out->parent,
> +					 diff->parent.ranges[i].start,
> +					 diff->parent.ranges[i].end);
> +			range_set_append(&out->target,
> +					 diff->target.ranges[i].start,
> +					 diff->target.ranges[i].end);
> +		}
> +	}
> +}

Shouldn't the ranges be merged, not just appended?

> diff --git a/log-tree.c b/log-tree.c
> index c894930..3fb025d 100644
> --- a/log-tree.c
> +++ b/log-tree.c
> @@ -809,6 +809,9 @@ int log_tree_commit(struct rev_info *opt, struct commit *commit)
>  	log.parent = NULL;
>  	opt->loginfo = &log;
>  
> +	if (opt->line_level_traverse)
> +		return line_log_print(opt, commit);
> +

#include "line.h" is missing from the beginning of this file.

> diff --git a/revision.h b/revision.h
> index 6d09550..01902cf 100644
> --- a/revision.h
> +++ b/revision.h
> @@ -92,7 +92,8 @@ struct rev_info {
>  			cherry_mark:1,
>  			bisect:1,
>  			ancestry_path:1,
> -			first_parent_only:1;
> +			first_parent_only:1,
> +			line_level_traverse:1;
>  
>  	/* Diff flags */
>  	unsigned int	diff:1,
> @@ -168,6 +169,9 @@ struct rev_info {
>  	int count_left;
>  	int count_right;
>  	int count_same;
> +
> +	/* line level range that we are chasing */
> +	struct decoration line_log_data;

Good use of decoration.

>  };
>  
>  #define REV_TREE_SAME		0

Re: [PATCH v7 3/5] Export three functions from diff.c

[Junio C Hamano]

Thomas Rast <trast@student.ethz.ch> writes:

> From: Bo Yang <struggleyb.nku@gmail.com>
>
> Use fill_metainfo to fill the line level diff meta data,
> emit_line to print out a line and quote_two to quote
> paths.
>
> Signed-off-by: Bo Yang <struggleyb.nku@gmail.com>
> Signed-off-by: Thomas Rast <trast@student.ethz.ch>
> ---

Why?  Neither 4/5 or 5/5 seem to use any of these.

Re: [PATCH v7 2/5] blame: introduce $ as "end of file" in -L syntax

[Thomas Rast]

Junio C Hamano <gitster@pobox.com> writes:

> Thomas Rast <trast@student.ethz.ch> writes:
>
>> To save the user a lookup of the last line number, introduce $ as a
>> shorthand for the last line.  This is mostly useful to spell "until
>> the end of the file" as '-L<begin>,$'.
>
> Hmph.  This is mostly useful not to error out when user (perhaps
> mistakenly) expects that the end of file is spelled as "$"; both
> "git blame -L<begin>" and "git blame L<begin>," have always meant
> "til the end", IIRC.

Heh.  That's actually a very good point; I think neither Bo nor me have
thought about allowing git log -L <begin>:<file>.

-- 
Thomas Rast
trast@{inf,student}.ethz.ch

Re: [PATCH v7 5/5] Implement line-history search (git log -L)

[Thomas Rast]

Junio C Hamano <gitster@pobox.com> writes:

> Thomas Rast <trast@student.ethz.ch> writes:
>
>> This is a rewrite of much of Bo's work, mainly in an effort to split
>> it into smaller, easier to understand routines.
>
> You mentioned "splitting" in the cover letter, but it does seem to
> need a bit more work.

Yes, I think I also mentioned that it's not ready for inclusion ;-)

Most of your points are spot on, however:

>> +static void diff_ranges_release (struct diff_ranges *diff)
>> +{
>> +	range_set_release(&diff->parent);
>> +	range_set_release(&diff->target);
>> +}
>
> Unused.

That should end up being used a few times...

>> +static void diff_ranges_filter_touched (struct diff_ranges *out,
>> +					struct diff_ranges *diff,
>> +					struct range_set *rs)
...
>> +		if (ranges_overlap(&diff->target.ranges[i], &rs->ranges[j])) {
>> +			range_set_append(&out->parent,
>> +					 diff->parent.ranges[i].start,
>> +					 diff->parent.ranges[i].end);
>> +			range_set_append(&out->target,
>> +					 diff->target.ranges[i].start,
>> +					 diff->target.ranges[i].end);
>
> Shouldn't the ranges be merged, not just appended?

If the code ever passed anything but an empty struct diff_ranges as the
'out' argument, yes.  But it doesn't.  In general I'm usually doing the
'out' dance to save one heap allocation.  Perhaps it would be cleaner to
allocate all of them on the heap instead, and return as pointers, dunno.

>> +	/* line level range that we are chasing */
>> +	struct decoration line_log_data;
>
> Good use of decoration.

That was actually Bo's idea.

-- 
Thomas Rast
trast@{inf,student}.ethz.ch

Re: [PATCH v7 5/5] Implement line-history search (git log -L)

[Zbigniew Jędrzejewski-Szmek]

On 06/07/2012 12:23 PM, Thomas Rast wrote:
> The algorithm is then simply to process history in topological order
> from newest to oldest, computing ranges and (partial) diffs.  At
> branch points, we need to merge the ranges we are watching.  We will
> find that many commits do not affect the chosen ranges, and mark them
> TREESAME (in addition to those already filtered by pathspec limiting).
> Another pass of history simplification then gets rid of such commits.

Hi,

this is absolutely great.

When I run your example invocations, nothing is displayed for a very
long time. I understand that this is because of the extra
'simplification pass', which means that whole results need to be ready
before anything is displayed. Adding something like '-10' doesn't seem
to have any effect, so I guess it is ignored. I hope that making '-<n>'
work would not be to complicated, but more important would be getting
incremental results. Will it be possible?

Zbyszek

Re: [PATCH v7 0/5] git log -L, all new and shiny

[Junio C Hamano]

Thomas Rast <trast@student.ethz.ch> writes:

> I too thought it would never happen -- but then again this is still
> not ready, I'm just trying to give it some exposure.
> ...
> There's also a longer-term wishlist hinted at in the commit message of
> the main patch: the diff machinery currently makes no provisions for
> chaining its various bells and whistles.

I am not convinced that it is "diff machinery makes no provivsions"
that is the problem. Isn't it coming from the way the series limits
the output line range and reimplements its own output routine?

All the "bells and whistles" like diffstat, word coloring, etc. go
through the xdi_diff_outf() interface, so isn't it the matter of
limiting lines that this interface calls back the "bells and
whistles" callback functions with?

When you enter the diff machinery, you have the path and the line
range you are interested in of one side (lets say you are comparing
side A and B, and have line range for side A).

If you

 - add a mechanism to pass the "interesting" line range and path
   down to the callchain from xdi_diff_outf() to xdiff_outf();

 - make one of these functions filter out (i.e. not call the
   callback xdiff_emit_consume_fn) hunks that do not overlap with
   the line range you are interested in (I would presume that they
   would be a few new fields in xdemitconf_t structure); and

 - while recording the corresponding line ranges in the other side
   of the hunks that are output,

that would give you

 - output that is limited to the "interesting" input range of side A;

 - the corresponding "interesting" range in the other side of the
   comparison, so that you can update the "interesting" range to
   feed to the next diff that compares side B with something else; and

 - for whatever processing the various "bells and whistles" callers
   already implement, as all their callbacks see are the lines in
   your "interesting" range.

No?  Am I missing something?

Re: [PATCH v7 0/5] git log -L, all new and shiny

[Thomas Rast]

Junio C Hamano <gitster@pobox.com> writes:

> Thomas Rast <trast@student.ethz.ch> writes:
>
>> I too thought it would never happen -- but then again this is still
>> not ready, I'm just trying to give it some exposure.
>> ...
>> There's also a longer-term wishlist hinted at in the commit message of
>> the main patch: the diff machinery currently makes no provisions for
>> chaining its various bells and whistles.
>
> I am not convinced that it is "diff machinery makes no provivsions"
> that is the problem. Isn't it coming from the way the series limits
> the output line range and reimplements its own output routine?

Well, in a very circular logic sense, yes: I reimplement the output
routine because that's the only way I could think of doing it right now :-)

However, notice that word-diff also reimplements its own output routine,
though it probably has a better standing since it is a different format.

>  - add a mechanism to pass the "interesting" line range and path
>    down to the callchain from xdi_diff_outf() to xdiff_outf();
>
>  - make one of these functions filter out (i.e. not call the
>    callback xdiff_emit_consume_fn) hunks that do not overlap with
>    the line range you are interested in (I would presume that they
>    would be a few new fields in xdemitconf_t structure); and
>
>  - while recording the corresponding line ranges in the other side
>    of the hunks that are output,

Hrm.

This would be the first backwards coupling between the revision-walk and
the diff generation parts, at least that I know of.  Normally the
revision walker just calls out to the (line-wise, not tree-based) diff
engine when it wants to show a commit.  Now suddenly the diff engine is
used (a lot, too) in simplifying the history.

Ideally we would want to reuse diffs that have already been generated,
as this is a very expensive process.  The current log -L implementation
manages to do this at the cost of reimplementing the diff output
routines instead.

You solve it instead by mandating that the diff engine itself updates
the "interesting" ranges, but that needs a lot of inside knowledge: like
in blame, we sometimes explore alternatives (e.g. for merges; or with
-M, though log -L in this version does not implement that feature).

So we would end up with redoing diffs, or a very tight coupling, that
IMHO just makes the mess worse.

Or am I missing something?

I instead have the vision that eventually diffs should be represented
internally as something like my pairs of struct range_set.  Then we
could run more passes on them as needed, and have a "common currency"
between all diff-related work.  Only the last one should then actually
output the diff.

That still doesn't properly account for the case where the data format
is no longer in terms of hunks (such as for word-diff, or the stat
formats), though.

-- 
Thomas Rast
trast@{inf,student}.ethz.ch

Re: [PATCH v7 0/5] git log -L, all new and shiny

[Junio C Hamano]

Thomas Rast <trast@student.ethz.ch> writes:

> Junio C Hamano <gitster@pobox.com> writes:
>
>> Thomas Rast <trast@student.ethz.ch> writes:
>>
>>> I too thought it would never happen -- but then again this is still
>>> not ready, I'm just trying to give it some exposure.
>>> ...
>>> There's also a longer-term wishlist hinted at in the commit message of
>>> the main patch: the diff machinery currently makes no provisions for
>>> chaining its various bells and whistles.
>>
>> I am not convinced that it is "diff machinery makes no provivsions"
>> that is the problem. Isn't it coming from the way the series limits
>> the output line range and reimplements its own output routine?
>
> Well, in a very circular logic sense, yes: I reimplement the output
> routine because that's the only way I could think of doing it right now :-)
>
> However, notice that word-diff also reimplements its own output routine,
> though it probably has a better standing since it is a different format.

Also notice that word-diff uses the same xdi_diff_outf() machinery
to grab line-oriented diff as its input (cf. fn_out_consume), and
then does its thing on it.  If you limit what fn_out_consume sees,
you can have word-diff do exactly what you want, no?

> This would be the first backwards coupling between the revision-walk and
> the diff generation parts, at least that I know of.

I am not convinced if you need to have any unusual back-coupling to
begin with, by the way.

If you say "git log -p [--options] -- pathspec", the revision
machinery does filter commits that do not touch any paths that patch
pathspec with the TREESAME logic, but that does not necessarily mean
you will see _all_ the commits that are not TREESAME.  If you for
example use ignore-space-change options, even the preimage and the
postimage differ at the object name level (hence not TREESAME), the
diff machinery already knows how to tell the revision machinery not
to show the log message and stuff, causing the commit to be skipped
from the output, no?

I do not know why you think you would need to do the filtering
"range comparison and union" computation more than necessary.  If
the user asks "log -p", you need to do it once per parent-child pair
that is not TREESAME at the place the current code calls run_diff().
I suspect that "log -p --stat" could be improved to eliminate the
separate call to run_diffstat() by restructuring the code so that
the statistics is gathered inside run_diff(), but that is independent
of this series. If this series hooked into the level I hinted in my
earlier message, such an optimization will reduce calls to your
"range comparision and union" computation for free.

Re: [PATCH v7 0/5] git log -L, all new and shiny

[Junio C Hamano]

Junio C Hamano <gitster@pobox.com> writes:

> Thomas Rast <trast@student.ethz.ch> writes:
>
>> This would be the first backwards coupling between the revision-walk and
>> the diff generation parts, at least that I know of.
>
> I am not convinced if you need to have any unusual back-coupling to
> begin with, by the way.
>
> If you say "git log -p [--options] -- pathspec", the revision
> machinery does filter commits that do not touch any paths that patch
> pathspec with the TREESAME logic, but that does not necessarily mean
> you will see _all_ the commits that are not TREESAME.

Let's clarify this part a bit.

Imagine you have three commits on a single strand of pearls.

	A---B---C

Between A and B, you only corrected indentation errors in file F as
a clean-up step, and then between B and C, you did a real change.
"git diff A B -- F" gives changes, "git diff -w A B -- F" does not.
Both "git diff B C -- F" and "git diff -w B C -- F" give you some
output.

Now, "git log --no-root -p -w C -- F" will give you C but not B.

Let's see how that happens.

The revision machinery looks at C and finds its parent B.  It runs
object level tree comparison and finds that their trees are
different at path F.  It makes a mental note that it may need to
show the log message of C, and asks the diff machinery to run
diff-tree between B and C.  The diff machinery finds that it needs
to show something even in the presense of -w option by actual
comparison, and just before showing the very first line of patch
output, it shows the log message of C (due to the earlier "mental
note").

Then the revision machinery looks at B.  It does the same between B
and A, but this time around, the diff machinery finds that, even
though A and B were _not_ TREESAME at the revision traversal level,
there is nothing to be shown after filtering with the -w option.
Hence no patch is shown and log message for B is not shown, either.

The interesting part of the above is that we did not bother using
the -w comparison to affect TREESAME check at the revision traversal
level.

I consider your "-L bottom,top" essentially the same "two blobs may
be different at the object name level, but after filtering the diff,
there may be no output" filter just like the "-w" option is.
Instead of filtering hunks that have changes only in whitespaces, it
filters hunks that do not overlap the given line range.  So if you
replace "-w" in the above example with "-L bottom,top", exactly the
same thing should happen.

The current code structure does not consider the content level
filtering done by the "-w" when computing TREESAME.  This DOES
affect how the merges are simplified.  If you and I start from a
commit X with full of indentation mistakes, you fix the earlier half
of these mistakes and make commit T while I fix the later half of
these mistakes and make commit J, and we merge our results into
merge M:

      J---M
     /   /
    X---T

none of "diff -w J M", "diff -w T M", "diff -w X J", "diff -w X T"
will output anything.  "git diff -p -w M" will however traverse both
branches, even though in this example nothing will be shown in the
output, because no two trees in these four commits are TREESAME.  

In the longer term, this _may_ be something we want to fix, but
teaching the revision machinery about only "-L bottom,top" is not
the way to do it.  Instead, we should devise a new mechanism to call
into the diff_patch machinery from the place where the revision
machinery computes TREESAME, and let it ask the filtered content
level differences if any of these options (these flags flip the
DIFF_FROM_CONTENTS diff option) are in use.  As a part of the
implementation of that new mechanism, we may want to cache the patch
such an early comparison produces, and reuse it when we actually
produce output, as an optimization.  But I think that is not limited
to the "-L bottom,top" option and an orthogonal issue, as it will
work equally well for existing "-w" filter (there may be others).

The above is where my recommendations are coming from.  The first
step for "-L bottom,top" should be to hook it as a new kind of
DIFF_FROM_CONTENTS option that compares two non-identical blobs but
possibly yield empty result into the diff machinery at the same
level as "-w".  Once that is solidly done, we can think about
updating the merge simplification logic to take DIFF_FROM_CONTENTS
changes into account when it computes TREESAME.

Re: [PATCH v7 0/5] git log -L, all new and shiny

[Thomas Rast]

Junio C Hamano <gitster@pobox.com> writes:

> Junio C Hamano <gitster@pobox.com> writes:
>
>> Thomas Rast <trast@student.ethz.ch> writes:
>>
>>> This would be the first backwards coupling between the revision-walk and
>>> the diff generation parts, at least that I know of.
>>
>> I am not convinced if you need to have any unusual back-coupling to
>> begin with, by the way.
>>
>> If you say "git log -p [--options] -- pathspec", the revision
>> machinery does filter commits that do not touch any paths that patch
>> pathspec with the TREESAME logic, but that does not necessarily mean
>> you will see _all_ the commits that are not TREESAME.
[...]
> The revision machinery looks at C and finds its parent B.  It runs
> object level tree comparison and finds that their trees are
> different at path F.  It makes a mental note that it may need to
> show the log message of C, and asks the diff machinery to run
> diff-tree between B and C.  The diff machinery finds that it needs
> to show something even in the presense of -w option by actual
> comparison, and just before showing the very first line of patch
> output, it shows the log message of C (due to the earlier "mental
> note").
>
> Then the revision machinery looks at B.  It does the same between B
> and A, but this time around, the diff machinery finds that, even
> though A and B were _not_ TREESAME at the revision traversal level,
> there is nothing to be shown after filtering with the -w option.
> Hence no patch is shown and log message for B is not shown, either.

Thanks for the great explanations.

Having spent some time letting this sink in (and being busy doing other
things), I think it's actually a good idea.  It forces us to go back and
change it around so that the diff machinery gets a say _before_ we
simplify history.  I think this bit will be important for log -L history
to make sense, and it's a bug waiting to happen for the -w case.

-- 
Thomas Rast
trast@{inf,student}.ethz.ch

Re: [PATCH v7 0/5] git log -L, all new and shiny

[Junio C Hamano]

Thomas Rast <trast@inf.ethz.ch> writes:

>> Then the revision machinery looks at B.  It does the same between B
>> and A, but this time around, the diff machinery finds that, even
>> though A and B were _not_ TREESAME at the revision traversal level,
>> there is nothing to be shown after filtering with the -w option.
>> Hence no patch is shown and log message for B is not shown, either.
>
> Thanks for the great explanations.
>
> Having spent some time letting this sink in (and being busy doing other
> things), I think it's actually a good idea.  It forces us to go back and
> change it around so that the diff machinery gets a say _before_ we
> simplify history.  I think this bit will be important for log -L history
> to make sense, and it's a bug waiting to happen for the -w case.

Note that this is not limited to "diff_patch() already filters -w".

If you are running with --diff-filter=A to grab only the additions,
for example, you may want the merge simplification to know about
this filtering as well.

So it is likely that you would want to hook diffcore_std(), not just
diff_flush(), to the TREESAME machinery.  Obviously you would want
to do this only for the merge commits; there is no point doing this
for single strand of pearls where the output phase already knows how
to squelch output correctly.