Merge branch 'jc/spht'

* jc/spht:
  Use gitattributes to define per-path whitespace rule
  core.whitespace: documentation updates.
  builtin-apply: teach whitespace_rules
  builtin-apply: rename "whitespace" variables and fix styles
  core.whitespace: add test for diff whitespace error highlighting
  git-diff: complain about >=8 consecutive spaces in initial indent
  War on whitespace: first, a bit of retreat.

Conflicts:

	cache.h
	config.c
	diff.c

Documentation/config.txt

@@ -295,6 +295,20 @@ core.pager::
 	The command that git will use to paginate output.  Can be overridden
 	with the `GIT_PAGER` environment variable.
 
+core.whitespace::
+	A comma separated list of common whitespace problems to
+	notice.  `git diff` will use `color.diff.whitespace` to
+	highlight them, and `git apply --whitespace=error` will
+	consider them as errors:
++
+* `trailing-space` treats trailing whitespaces at the end of the line
+  as an error (enabled by default).
+* `space-before-tab` treats a space character that appears immediately
+  before a tab character in the initial indent part of the line as an
+  error (enabled by default).
+* `indent-with-non-tab` treats a line that is indented with 8 or more
+  space characters that can be replaced with tab characters.
+
 alias.*::
 	Command aliases for the gitlink:git[1] command wrapper - e.g.
 	after defining "alias.last = cat-file commit HEAD", the invocation
@@ -387,8 +401,8 @@ color.diff.<slot>::
 	which part of the patch to use the specified color, and is one
 	of `plain` (context text), `meta` (metainformation), `frag`
 	(hunk header), `old` (removed lines), `new` (added lines),
-	`commit` (commit headers), or `whitespace` (highlighting dubious
+	`commit` (commit headers), or `whitespace` (highlighting
-	whitespace).  The values of these variables may be specified as
+	whitespace errors). The values of these variables may be specified as
 	in color.branch.<slot>.
 
 color.interactive::

Documentation/git-apply.txt

@@ -13,7 +13,7 @@ SYNOPSIS
 	  [--apply] [--no-add] [--build-fake-ancestor <file>] [-R | --reverse]
 	  [--allow-binary-replacement | --binary] [--reject] [-z]
 	  [-pNUM] [-CNUM] [--inaccurate-eof] [--cached]
-	  [--whitespace=<nowarn|warn|error|error-all|strip>]
+	  [--whitespace=<nowarn|warn|fix|error|error-all>]
 	  [--exclude=PATH] [--verbose] [<patch>...]
 
 DESCRIPTION
@@ -135,25 +135,32 @@ discouraged.
 	be useful when importing patchsets, where you want to exclude certain
 	files or directories.
 
---whitespace=<option>::
+--whitespace=<action>::
-	When applying a patch, detect a new or modified line
+	When applying a patch, detect a new or modified line that has
-	that ends with trailing whitespaces (this includes a
+	whitespace errors.  What are considered whitespace errors is
-	line that solely consists of whitespaces).  By default,
+	controlled by `core.whitespace` configuration.  By default,
-	the command outputs warning messages and applies the
+	trailing whitespaces (including lines that solely consist of
-	patch.
+	whitespaces) and a space character that is immediately followed
-	When gitlink:git-apply[1] is used for statistics and not applying a
+	by a tab character inside the initial indent of the line are
-	patch, it defaults to `nowarn`.
+	considered whitespace errors.
-	You can use different `<option>` to control this
++
-	behavior:
+By default, the command outputs warning messages but applies the patch.
+When gitlink:git-apply[1] is used for statistics and not applying a
+patch, it defaults to `nowarn`.
++
+You can use different `<action>` to control this
+behavior:
 +
 * `nowarn` turns off the trailing whitespace warning.
 * `warn` outputs warnings for a few such errors, but applies the
-  patch (default).
+  patch as-is (default).
+* `fix` outputs warnings for a few such errors, and applies the
+  patch after fixing them (`strip` is a synonym --- the tool
+  used to consider only trailing whitespaces as errors, and the
+  fix involved 'stripping' them, but modern gits do more).
 * `error` outputs warnings for a few such errors, and refuses
   to apply the patch.
 * `error-all` is similar to `error` but shows all errors.
-* `strip` outputs warnings for a few such errors, strips out the
-  trailing whitespaces and applies the patch.
 
 --inaccurate-eof::
 	Under certain circumstances, some versions of diff do not correctly

Documentation/gitattributes.txt

@@ -361,6 +361,37 @@ When left unspecified, the driver itself is used for both
 internal merge and the final merge.
 
 
+Checking whitespace errors
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+`whitespace`
+^^^^^^^^^^^^
+
+The `core.whitespace` configuration variable allows you to define what
+`diff` and `apply` should consider whitespace errors for all paths in
+the project (See gitlink:git-config[1]).  This attribute gives you finer
+control per path.
+
+Set::
+
+	Notice all types of potential whitespace errors known to git.
+
+Unset::
+
+	Do not notice anything as error.
+
+Unspecified::
+
+	Use the value of `core.whitespace` configuration variable to
+	decide what to notice as error.
+
+String::
+
+	Specify a comma separate list of common whitespace problems to
+	notice in the same format as `core.whitespace` configuration
+	variable.
+
+
 EXAMPLE
 -------
 

Makefile

@@ -313,7 +313,7 @@ LIB_OBJS = \
 	alloc.o merge-file.o path-list.o help.o unpack-trees.o $(DIFF_OBJS) \
 	color.o wt-status.o archive-zip.o archive-tar.o shallow.o utf8.o \
 	convert.o attr.o decorate.o progress.o mailmap.o symlinks.o remote.o \
-	transport.o bundle.o walker.o parse-options.o
+	transport.o bundle.o walker.o parse-options.o ws.o
 
 BUILTIN_OBJS = \
 	builtin-add.o \

builtin-apply.c

@@ -45,14 +45,14 @@ static const char *fake_ancestor;
 static int line_termination = '\n';
 static unsigned long p_context = ULONG_MAX;
 static const char apply_usage[] =
-"git-apply [--stat] [--numstat] [--summary] [--check] [--index] [--cached] [--apply] [--no-add] [--index-info] [--allow-binary-replacement] [--reverse] [--reject] [--verbose] [-z] [-pNUM] [-CNUM] [--whitespace=<nowarn|warn|error|error-all|strip>] <patch>...";
+"git-apply [--stat] [--numstat] [--summary] [--check] [--index] [--cached] [--apply] [--no-add] [--index-info] [--allow-binary-replacement] [--reverse] [--reject] [--verbose] [-z] [-pNUM] [-CNUM] [--whitespace=<nowarn|warn|fix|error|error-all>] <patch>...";
 
-static enum whitespace_eol {
+static enum ws_error_action {
-	nowarn_whitespace,
+	nowarn_ws_error,
-	warn_on_whitespace,
+	warn_on_ws_error,
-	error_on_whitespace,
+	die_on_ws_error,
-	strip_whitespace,
+	correct_ws_error,
-} new_whitespace = warn_on_whitespace;
+} ws_error_action = warn_on_ws_error;
 static int whitespace_error;
 static int squelch_whitespace_errors = 5;
 static int applied_after_fixing_ws;
@@ -61,28 +61,28 @@ static const char *patch_input_file;
 static void parse_whitespace_option(const char *option)
 {
 	if (!option) {
-		new_whitespace = warn_on_whitespace;
+		ws_error_action = warn_on_ws_error;
 		return;
 	}
 	if (!strcmp(option, "warn")) {
-		new_whitespace = warn_on_whitespace;
+		ws_error_action = warn_on_ws_error;
 		return;
 	}
 	if (!strcmp(option, "nowarn")) {
-		new_whitespace = nowarn_whitespace;
+		ws_error_action = nowarn_ws_error;
 		return;
 	}
 	if (!strcmp(option, "error")) {
-		new_whitespace = error_on_whitespace;
+		ws_error_action = die_on_ws_error;
 		return;
 	}
 	if (!strcmp(option, "error-all")) {
-		new_whitespace = error_on_whitespace;
+		ws_error_action = die_on_ws_error;
 		squelch_whitespace_errors = 0;
 		return;
 	}
-	if (!strcmp(option, "strip")) {
+	if (!strcmp(option, "strip") || !strcmp(option, "fix")) {
-		new_whitespace = strip_whitespace;
+		ws_error_action = correct_ws_error;
 		return;
 	}
 	die("unrecognized whitespace option '%s'", option);
@@ -90,11 +90,8 @@ static void parse_whitespace_option(const char *option)
 
 static void set_default_whitespace_mode(const char *whitespace_option)
 {
-	if (!whitespace_option && !apply_default_whitespace) {
+	if (!whitespace_option && !apply_default_whitespace)
-		new_whitespace = (apply
+		ws_error_action = (apply ? warn_on_ws_error : nowarn_ws_error);
-				  ? warn_on_whitespace
-				  : nowarn_whitespace);
-	}
 }
 
 /*
@@ -137,11 +134,17 @@ struct fragment {
 #define BINARY_DELTA_DEFLATED	1
 #define BINARY_LITERAL_DEFLATED 2
 
+/*
+ * This represents a "patch" to a file, both metainfo changes
+ * such as creation/deletion, filemode and content changes represented
+ * as a series of fragments.
+ */
 struct patch {
 	char *new_name, *old_name, *def_name;
 	unsigned int old_mode, new_mode;
 	int is_new, is_delete;	/* -1 = unknown, 0 = false, 1 = true */
 	int rejected;
+	unsigned ws_rule;
 	unsigned long deflate_origlen;
 	int lines_added, lines_deleted;
 	int score;
@@ -158,7 +161,8 @@ struct patch {
 	struct patch *next;
 };
 
-static void say_patch_name(FILE *output, const char *pre, struct patch *patch, const char *post)
+static void say_patch_name(FILE *output, const char *pre,
+			   struct patch *patch, const char *post)
 {
 	fputs(pre, output);
 	if (patch->old_name && patch->new_name &&
@@ -229,7 +233,8 @@ static char *find_name(const char *line, char *def, int p_value, int terminate)
 	if (*line == '"') {
 		struct strbuf name;
 
-		/* Proposed "new-style" GNU patch/diff format; see
+		/*
+		 * Proposed "new-style" GNU patch/diff format; see
 		 * http://marc.theaimsgroup.com/?l=git&m=112927316408690&w=2
 		 */
 		strbuf_init(&name, 0);
@@ -499,7 +504,8 @@ static int gitdiff_dissimilarity(const char *line, struct patch *patch)
 
 static int gitdiff_index(const char *line, struct patch *patch)
 {
-	/* index line is N hexadecimal, "..", N hexadecimal,
+	/*
+	 * index line is N hexadecimal, "..", N hexadecimal,
 	 * and optional space with octal mode.
 	 */
 	const char *ptr, *eol;
@@ -550,7 +556,8 @@ static const char *stop_at_slash(const char *line, int llen)
 	return NULL;
 }
 
-/* This is to extract the same name that appears on "diff --git"
+/*
+ * This is to extract the same name that appears on "diff --git"
  * line.  We do not find and return anything if it is a rename
  * patch, and it is OK because we will find the name elsewhere.
  * We need to reliably find name only when it is mode-change only,
@@ -584,7 +591,8 @@ static char *git_header_name(char *line, int llen)
 			goto free_and_fail1;
 		strbuf_remove(&first, 0, cp + 1 - first.buf);
 
-		/* second points at one past closing dq of name.
+		/*
+		 * second points at one past closing dq of name.
 		 * find the second name.
 		 */
 		while ((second < line + llen) && isspace(*second))
@@ -627,7 +635,8 @@ static char *git_header_name(char *line, int llen)
 		return NULL;
 	name++;
 
-	/* since the first name is unquoted, a dq if exists must be
+	/*
+	 * since the first name is unquoted, a dq if exists must be
 	 * the beginning of the second name.
 	 */
 	for (second = name; second < line + llen; second++) {
@@ -758,7 +767,7 @@ static int parse_num(const char *line, unsigned long *p)
 }
 
 static int parse_range(const char *line, int len, int offset, const char *expect,
-			unsigned long *p1, unsigned long *p2)
+		       unsigned long *p1, unsigned long *p2)
 {
 	int digits, ex;
 
@@ -867,14 +876,14 @@ static int find_header(char *line, unsigned long size, int *hdrsize, struct patc
 			return offset;
 		}
 
-		/** --- followed by +++ ? */
+		/* --- followed by +++ ? */
 		if (memcmp("--- ", line,  4) || memcmp("+++ ", line + len, 4))
 			continue;
 
 		/*
 		 * We only accept unified patches, so we want it to
 		 * at least have "@@ -a,b +c,d @@\n", which is 14 chars
-		 * minimum
+		 * minimum ("@@ -0,0 +1 @@\n" is the shortest).
 		 */
 		nextlen = linelen(line + len, size - len);
 		if (size < nextlen + 14 || memcmp("@@ -", line + len + nextlen, 4))
@@ -889,7 +898,7 @@ static int find_header(char *line, unsigned long size, int *hdrsize, struct patc
 	return -1;
 }
 
-static void check_whitespace(const char *line, int len)
+static void check_whitespace(const char *line, int len, unsigned ws_rule)
 {
 	const char *err = "Adds trailing whitespace";
 	int seen_space = 0;
@@ -901,23 +910,35 @@ static void check_whitespace(const char *line, int len)
 	 * this function.  That is, an addition of an empty line would
 	 * check the '+' here.  Sneaky...
 	 */
-	if (isspace(line[len-2]))
+	if ((ws_rule & WS_TRAILING_SPACE) && isspace(line[len-2]))
 		goto error;
 
 	/*
 	 * Make sure that there is no space followed by a tab in
 	 * indentation.
 	 */
-	err = "Space in indent is followed by a tab";
+	if (ws_rule & WS_SPACE_BEFORE_TAB) {
-	for (i = 1; i < len; i++) {
+		err = "Space in indent is followed by a tab";
-		if (line[i] == '\t') {
+		for (i = 1; i < len; i++) {
-			if (seen_space)
+			if (line[i] == '\t') {
-				goto error;
+				if (seen_space)
+					goto error;
+			}
+			else if (line[i] == ' ')
+				seen_space = 1;
+			else
+				break;
 		}
-		else if (line[i] == ' ')
+	}
-			seen_space = 1;
+
-		else
+	/*
-			break;
+	 * Make sure that the indentation does not contain more than
+	 * 8 spaces.
+	 */
+	if ((ws_rule & WS_INDENT_WITH_NON_TAB) &&
+	    (8 < len) && !strncmp("+        ", line, 9)) {
+		err = "Indent more than 8 places with spaces";
+		goto error;
 	}
 	return;
 
@@ -931,14 +952,14 @@ static void check_whitespace(const char *line, int len)
 			err, patch_input_file, linenr, len-2, line+1);
 }
 
-
 /*
  * Parse a unified diff. Note that this really needs to parse each
  * fragment separately, since the only way to know the difference
  * between a "---" that is part of a patch, and a "---" that starts
  * the next patch is to look at the line counts..
  */
-static int parse_fragment(char *line, unsigned long size, struct patch *patch, struct fragment *fragment)
+static int parse_fragment(char *line, unsigned long size,
+			  struct patch *patch, struct fragment *fragment)
 {
 	int added, deleted;
 	int len = linelen(line, size), offset;
@@ -979,22 +1000,23 @@ static int parse_fragment(char *line, unsigned long size, struct patch *patch, s
 			break;
 		case '-':
 			if (apply_in_reverse &&
-					new_whitespace != nowarn_whitespace)
+			    ws_error_action != nowarn_ws_error)
-				check_whitespace(line, len);
+				check_whitespace(line, len, patch->ws_rule);
 			deleted++;
 			oldlines--;
 			trailing = 0;
 			break;
 		case '+':
 			if (!apply_in_reverse &&
-					new_whitespace != nowarn_whitespace)
+			    ws_error_action != nowarn_ws_error)
-				check_whitespace(line, len);
+				check_whitespace(line, len, patch->ws_rule);
 			added++;
 			newlines--;
 			trailing = 0;
 			break;
 
-                /* We allow "\ No newline at end of file". Depending
+		/*
+		 * We allow "\ No newline at end of file". Depending
                  * on locale settings when the patch was produced we
                  * don't know what this line looks like. The only
                  * thing we do know is that it begins with "\ ".
@@ -1012,7 +1034,8 @@ static int parse_fragment(char *line, unsigned long size, struct patch *patch, s
 	fragment->leading = leading;
 	fragment->trailing = trailing;
 
-	/* If a fragment ends with an incomplete line, we failed to include
+	/*
+	 * If a fragment ends with an incomplete line, we failed to include
 	 * it in the above loop because we hit oldlines == newlines == 0
 	 * before seeing it.
 	 */
@@ -1140,7 +1163,8 @@ static struct fragment *parse_binary_hunk(char **buf_p,
 					  int *status_p,
 					  int *used_p)
 {
-	/* Expect a line that begins with binary patch method ("literal"
+	/*
+	 * Expect a line that begins with binary patch method ("literal"
 	 * or "delta"), followed by the length of data before deflating.
 	 * a sequence of 'length-byte' followed by base-85 encoded data
 	 * should follow, terminated by a newline.
@@ -1189,7 +1213,8 @@ static struct fragment *parse_binary_hunk(char **buf_p,
 			size--;
 			break;
 		}
-		/* Minimum line is "A00000\n" which is 7-byte long,
+		/*
+		 * Minimum line is "A00000\n" which is 7-byte long,
 		 * and the line length must be multiple of 5 plus 2.
 		 */
 		if ((llen < 7) || (llen-2) % 5)
@@ -1240,7 +1265,8 @@ static struct fragment *parse_binary_hunk(char **buf_p,
 
 static int parse_binary(char *buffer, unsigned long size, struct patch *patch)
 {
-	/* We have read "GIT binary patch\n"; what follows is a line
+	/*
+	 * We have read "GIT binary patch\n"; what follows is a line
 	 * that says the patch method (currently, either "literal" or
 	 * "delta") and the length of data before deflating; a
 	 * sequence of 'length-byte' followed by base-85 encoded data
@@ -1270,7 +1296,8 @@ static int parse_binary(char *buffer, unsigned long size, struct patch *patch)
 	if (reverse)
 		used += used_1;
 	else if (status) {
-		/* not having reverse hunk is not an error, but having
+		/*
+		 * Not having reverse hunk is not an error, but having
 		 * a corrupt reverse hunk is.
 		 */
 		free((void*) forward->patch);
@@ -1291,7 +1318,12 @@ static int parse_chunk(char *buffer, unsigned long size, struct patch *patch)
 	if (offset < 0)
 		return offset;
 
-	patchsize = parse_single_patch(buffer + offset + hdrsize, size - offset - hdrsize, patch);
+	patch->ws_rule = whitespace_rule(patch->new_name
+					 ? patch->new_name
+					 : patch->old_name);
+
+	patchsize = parse_single_patch(buffer + offset + hdrsize,
+				       size - offset - hdrsize, patch);
 
 	if (!patchsize) {
 		static const char *binhdr[] = {
@@ -1367,8 +1399,10 @@ static void reverse_patches(struct patch *p)
 	}
 }
 
-static const char pluses[] = "++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++";
+static const char pluses[] =
-static const char minuses[]= "----------------------------------------------------------------------";
+"++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++";
+static const char minuses[]=
+"----------------------------------------------------------------------";
 
 static void show_stats(struct patch *patch)
 {
@@ -1437,7 +1471,9 @@ static int read_old_data(struct stat *st, const char *path, struct strbuf *buf)
 	}
 }
 
-static int find_offset(const char *buf, unsigned long size, const char *fragment, unsigned long fragsize, int line, int *lines)
+static int find_offset(const char *buf, unsigned long size,
+		       const char *fragment, unsigned long fragsize,
+		       int line, int *lines)
 {
 	int i;
 	unsigned long start, backwards, forwards;
@@ -1536,9 +1572,11 @@ static void remove_last_line(const char **rbuf, int *rsize)
 	*rsize = offset + 1;
 }
 
-static int apply_line(char *output, const char *patch, int plen)
+static int apply_line(char *output, const char *patch, int plen,
+		      unsigned ws_rule)
 {
-	/* plen is number of bytes to be copied from patch,
+	/*
+	 * plen is number of bytes to be copied from patch,
 	 * starting at patch+1 (patch[0] is '+').  Typically
 	 * patch[plen] is '\n', unless this is the incomplete
 	 * last line.
@@ -1551,13 +1589,17 @@ static int apply_line(char *output, const char *patch, int plen)
 	int need_fix_leading_space = 0;
 	char *buf;
 
-	if ((new_whitespace != strip_whitespace) || !whitespace_error ||
+	if ((ws_error_action != correct_ws_error) || !whitespace_error ||
 	    *patch != '+') {
 		memcpy(output, patch + 1, plen);
 		return plen;
 	}
 
-	if (1 < plen && isspace(patch[plen-1])) {
+	/*
+	 * Strip trailing whitespace
+	 */
+	if ((ws_rule & WS_TRAILING_SPACE) &&
+	    (1 < plen && isspace(patch[plen-1]))) {
 		if (patch[plen] == '\n')
 			add_nl_to_tail = 1;
 		plen--;
@@ -1566,15 +1608,23 @@ static int apply_line(char *output, const char *patch, int plen)
 		fixed = 1;
 	}
 
+	/*
+	 * Check leading whitespaces (indent)
+	 */
 	for (i = 1; i < plen; i++) {
 		char ch = patch[i];
 		if (ch == '\t') {
 			last_tab_in_indent = i;
-			if (0 <= last_space_in_indent)
+			if ((ws_rule & WS_SPACE_BEFORE_TAB) &&
+			    0 <= last_space_in_indent)
+			    need_fix_leading_space = 1;
+		} else if (ch == ' ') {
+			last_space_in_indent = i;
+			if ((ws_rule & WS_INDENT_WITH_NON_TAB) &&
+			    last_tab_in_indent < 0 &&
+			    8 <= i)
 				need_fix_leading_space = 1;
 		}
-		else if (ch == ' ')
-			last_space_in_indent = i;
 		else
 			break;
 	}
@@ -1582,10 +1632,21 @@ static int apply_line(char *output, const char *patch, int plen)
 	buf = output;
 	if (need_fix_leading_space) {
 		int consecutive_spaces = 0;
-		/* between patch[1..last_tab_in_indent] strip the
+		int last = last_tab_in_indent + 1;
-		 * funny spaces, updating them to tab as needed.
+
+		if (ws_rule & WS_INDENT_WITH_NON_TAB) {
+			/* have "last" point at one past the indent */
+			if (last_tab_in_indent < last_space_in_indent)
+				last = last_space_in_indent + 1;
+			else
+				last = last_tab_in_indent + 1;
+		}
+
+		/*
+		 * between patch[1..last], strip the funny spaces,
+		 * updating them to tab as needed.
 		 */
-		for (i = 1; i < last_tab_in_indent; i++, plen--) {
+		for (i = 1; i < last; i++, plen--) {
 			char ch = patch[i];
 			if (ch != ' ') {
 				consecutive_spaces = 0;
@@ -1598,8 +1659,10 @@ static int apply_line(char *output, const char *patch, int plen)
 				}
 			}
 		}
+		while (0 < consecutive_spaces--)
+			*output++ = ' ';
 		fixed = 1;
-		i = last_tab_in_indent;
+		i = last;
 	}
 	else
 		i = 1;
@@ -1612,7 +1675,8 @@ static int apply_line(char *output, const char *patch, int plen)
 	return output + plen - buf;
 }
 
-static int apply_one_fragment(struct strbuf *buf, struct fragment *frag, int inaccurate_eof)
+static int apply_one_fragment(struct strbuf *buf, struct fragment *frag,
+			      int inaccurate_eof, unsigned ws_rule)
 {
 	int match_beginning, match_end;
 	const char *patch = frag->patch;
@@ -1671,7 +1735,7 @@ static int apply_one_fragment(struct strbuf *buf, struct fragment *frag, int ina
 		case '+':
 			if (first != '+' || !no_add) {
 				int added = apply_line(new + newsize, patch,
-						       plen);
+						       plen, ws_rule);
 				newsize += added;
 				if (first == '+' &&
 				    added == 1 && new[newsize-1] == '\n')
@@ -1694,8 +1758,9 @@ static int apply_one_fragment(struct strbuf *buf, struct fragment *frag, int ina
 		size -= len;
 	}
 
-	if (inaccurate_eof && oldsize > 0 && old[oldsize - 1] == '\n' &&
+	if (inaccurate_eof &&
-			newsize > 0 && new[newsize - 1] == '\n') {
+	    oldsize > 0 && old[oldsize - 1] == '\n' &&
+	    newsize > 0 && new[newsize - 1] == '\n') {
 		oldsize--;
 		newsize--;
 	}
@@ -1732,7 +1797,7 @@ static int apply_one_fragment(struct strbuf *buf, struct fragment *frag, int ina
 		if (match_beginning && offset)
 			offset = -1;
 		if (offset >= 0) {
-			if (new_whitespace == strip_whitespace &&
+			if (ws_error_action == correct_ws_error &&
 			    (buf->len - oldsize - offset == 0)) /* end of file? */
 				newsize -= new_blank_lines_at_end;
 
@@ -1757,9 +1822,10 @@ static int apply_one_fragment(struct strbuf *buf, struct fragment *frag, int ina
 			match_beginning = match_end = 0;
 			continue;
 		}
-		/* Reduce the number of context lines
+		/*
-		 * Reduce both leading and trailing if they are equal
+		 * Reduce the number of context lines; reduce both
-		 * otherwise just reduce the larger context.
+		 * leading and trailing if they are equal otherwise
+		 * just reduce the larger context.
 		 */
 		if (leading >= trailing) {
 			remove_first_line(&oldlines, &oldsize);
@@ -1819,7 +1885,8 @@ static int apply_binary(struct strbuf *buf, struct patch *patch)
 	const char *name = patch->old_name ? patch->old_name : patch->new_name;
 	unsigned char sha1[20];
 
-	/* For safety, we require patch index line to contain
+	/*
+	 * For safety, we require patch index line to contain
 	 * full 40-byte textual SHA1 for old and new, at least for now.
 	 */
 	if (strlen(patch->old_sha1_prefix) != 40 ||
@@ -1830,7 +1897,8 @@ static int apply_binary(struct strbuf *buf, struct patch *patch)
 			     "without full index line", name);
 
 	if (patch->old_name) {
-		/* See if the old one matches what the patch
+		/*
+		 * See if the old one matches what the patch
 		 * applies to.
 		 */
 		hash_sha1_file(buf->buf, buf->len, blob_type, sha1);
@@ -1867,7 +1935,8 @@ static int apply_binary(struct strbuf *buf, struct patch *patch)
 		/* XXX read_sha1_file NUL-terminates */
 		strbuf_attach(buf, result, size, size + 1);
 	} else {
-		/* We have verified buf matches the preimage;
+		/*
+		 * We have verified buf matches the preimage;
 		 * apply the patch data to it, which is stored
 		 * in the patch->fragments->{patch,size}.
 		 */
@@ -1889,12 +1958,14 @@ static int apply_fragments(struct strbuf *buf, struct patch *patch)
 {
 	struct fragment *frag = patch->fragments;
 	const char *name = patch->old_name ? patch->old_name : patch->new_name;
+	unsigned ws_rule = patch->ws_rule;
+	unsigned inaccurate_eof = patch->inaccurate_eof;
 
 	if (patch->is_binary)
 		return apply_binary(buf, patch);
 
 	while (frag) {
-		if (apply_one_fragment(buf, frag, patch->inaccurate_eof)) {
+		if (apply_one_fragment(buf, frag, inaccurate_eof, ws_rule)) {
 			error("patch failed: %s:%ld", name, frag->oldpos);
 			if (!apply_with_reject)
 				return -1;
@@ -2066,7 +2137,8 @@ static int check_patch(struct patch *patch, struct patch *prev_patch)
 
 	if (new_name && prev_patch && 0 < prev_patch->is_delete &&
 	    !strcmp(prev_patch->old_name, new_name))
-		/* A type-change diff is always split into a patch to
+		/*
+		 * A type-change diff is always split into a patch to
 		 * delete old, immediately followed by a patch to
 		 * create new (see diff.c::run_diff()); in such a case
 		 * it is Ok that the entry to be deleted by the
@@ -2670,7 +2742,7 @@ static int apply_patch(int fd, const char *filename, int inaccurate_eof)
 		offset += nr;
 	}
 
-	if (whitespace_error && (new_whitespace == error_on_whitespace))
+	if (whitespace_error && (ws_error_action == die_on_ws_error))
 		apply = 0;
 
 	update_index = check_index && apply;
@@ -2865,7 +2937,7 @@ int cmd_apply(int argc, const char **argv, const char *unused_prefix)
 				squelched,
 				squelched == 1 ? "" : "s");
 		}
-		if (new_whitespace == error_on_whitespace)
+		if (ws_error_action == die_on_ws_error)
 			die("%d line%s add%s whitespace errors.",
 			    whitespace_error,
 			    whitespace_error == 1 ? "" : "s",

cache.h

@@ -644,6 +644,18 @@ extern int diff_auto_refresh_index;
 /* match-trees.c */
 void shift_tree(const unsigned char *, const unsigned char *, unsigned char *, int);
 
+/*
+ * whitespace rules.
+ * used by both diff and apply
+ */
+#define WS_TRAILING_SPACE	01
+#define WS_SPACE_BEFORE_TAB	02
+#define WS_INDENT_WITH_NON_TAB	04
+#define WS_DEFAULT_RULE (WS_TRAILING_SPACE|WS_SPACE_BEFORE_TAB)
+extern unsigned whitespace_rule_cfg;
+extern unsigned whitespace_rule(const char *);
+extern unsigned parse_whitespace_rule(const char *);
+
 /* ls-files */
 int pathspec_match(const char **spec, char *matched, const char *filename, int skiplen);
 int report_path_error(const char *ps_matched, const char **pathspec, int prefix_offset);

config.c

@@ -439,6 +439,11 @@ int git_default_config(const char *var, const char *value)
 		return 0;
 	}
 
+	if (!strcmp(var, "core.whitespace")) {
+		whitespace_rule_cfg = parse_whitespace_rule(value);
+		return 0;
+	}
+
 	/* Add other config variables here and to Documentation/config.txt. */
 	return 0;
 }

diff.c

@@ -454,6 +454,7 @@ static void diff_words_show(struct diff_words_data *diff_words)
 struct emit_callback {
 	struct xdiff_emit_state xm;
 	int nparents, color_diff;
+	unsigned ws_rule;
 	const char **label_path;
 	struct diff_words_data *diff_words;
 	int *found_changesp;
@@ -493,8 +494,8 @@ static void emit_line(const char *set, const char *reset, const char *line, int
 }
 
 static void emit_line_with_ws(int nparents,
-		const char *set, const char *reset, const char *ws,
+			      const char *set, const char *reset, const char *ws,
-		const char *line, int len)
+			      const char *line, int len, unsigned ws_rule)
 {
 	int col0 = nparents;
 	int last_tab_in_indent = -1;
@@ -502,13 +503,17 @@ static void emit_line_with_ws(int nparents,
 	int i;
 	int tail = len;
 	int need_highlight_leading_space = 0;
-	/* The line is a newly added line.  Does it have funny leading
+	/*
-	 * whitespaces?  In indent, SP should never precede a TAB.
+	 * The line is a newly added line.  Does it have funny leading
+	 * whitespaces?  In indent, SP should never precede a TAB.  In
+	 * addition, under "indent with non tab" rule, there should not
+	 * be more than 8 consecutive spaces.
 	 */
 	for (i = col0; i < len; i++) {
 		if (line[i] == '\t') {
 			last_tab_in_indent = i;
-			if (0 <= last_space_in_indent)
+			if ((ws_rule & WS_SPACE_BEFORE_TAB) &&
+			    0 <= last_space_in_indent)
 				need_highlight_leading_space = 1;
 		}
 		else if (line[i] == ' ')
@@ -516,6 +521,13 @@ static void emit_line_with_ws(int nparents,
 		else
 			break;
 	}
+	if ((ws_rule & WS_INDENT_WITH_NON_TAB) &&
+	    0 <= last_space_in_indent &&
+	    last_tab_in_indent < 0 &&
+	    8 <= (i - col0)) {
+		last_tab_in_indent = i;
+		need_highlight_leading_space = 1;
+	}
 	fputs(set, stdout);
 	fwrite(line, col0, 1, stdout);
 	fputs(reset, stdout);
@@ -540,10 +552,12 @@ static void emit_line_with_ws(int nparents,
 	tail = len - 1;
 	if (line[tail] == '\n' && i < tail)
 		tail--;
-	while (i < tail) {
+	if (ws_rule & WS_TRAILING_SPACE) {
-		if (!isspace(line[tail]))
+		while (i < tail) {
-			break;
+			if (!isspace(line[tail]))
-		tail--;
+				break;
+			tail--;
+		}
 	}
 	if ((i < tail && line[tail + 1] != '\n')) {
 		/* This has whitespace between tail+1..len */
@@ -565,7 +579,7 @@ static void emit_add_line(const char *reset, struct emit_callback *ecbdata, cons
 		emit_line(set, reset, line, len);
 	else
 		emit_line_with_ws(ecbdata->nparents, set, reset, ws,
-				line, len);
+				  line, len, ecbdata->ws_rule);
 }
 
 static void fn_out_consume(void *priv, char *line, unsigned long len)
@@ -981,6 +995,7 @@ struct checkdiff_t {
 	struct xdiff_emit_state xm;
 	const char *filename;
 	int lineno, color_diff;
+	unsigned ws_rule;
 };
 
 static void checkdiff_consume(void *priv, char *line, unsigned long len)
@@ -1016,7 +1031,8 @@ static void checkdiff_consume(void *priv, char *line, unsigned long len)
 			if (white_space_at_end)
 				printf("white space at end");
 			printf(":%s ", reset);
-			emit_line_with_ws(1, set, reset, ws, line, len);
+			emit_line_with_ws(1, set, reset, ws, line, len,
+					  data->ws_rule);
 		}
 
 		data->lineno++;
@@ -1317,6 +1333,7 @@ static void builtin_diff(const char *name_a,
 		ecbdata.label_path = lbl;
 		ecbdata.color_diff = DIFF_OPT_TST(o, COLOR_DIFF);
 		ecbdata.found_changesp = &o->found_changes;
+		ecbdata.ws_rule = whitespace_rule(name_b ? name_b : name_a);
 		xpp.flags = XDF_NEED_MINIMAL | o->xdl_opts;
 		xecfg.ctxlen = o->context;
 		xecfg.flags = XDL_EMIT_FUNCNAMES;
@@ -1410,6 +1427,7 @@ static void builtin_checkdiff(const char *name_a, const char *name_b,
 	data.filename = name_b ? name_b : name_a;
 	data.lineno = 0;
 	data.color_diff = DIFF_OPT_TST(o, COLOR_DIFF);
+	data.ws_rule = whitespace_rule(data.filename);
 
 	if (fill_mmfile(&mf1, one) < 0 || fill_mmfile(&mf2, two) < 0)
 		die("unable to read files to diff");

environment.c

@@ -36,6 +36,7 @@ int pager_use_color = 1;
 char *editor_program;
 char *excludes_file;
 int auto_crlf = 0;	/* 1: both ways, -1: only when adding git objects */
+unsigned whitespace_rule_cfg = WS_DEFAULT_RULE;
 
 /* This is set by setup_git_dir_gently() and/or git_default_config() */
 char *git_work_tree_cfg;

t/t4019-diff-wserror.sh

@@ -0,0 +1,123 @@
+#!/bin/sh
+
+test_description='diff whitespace error detection'
+
+. ./test-lib.sh
+
+test_expect_success setup '
+
+	git config diff.color.whitespace "blue reverse" &&
+	>F &&
+	git add F &&
+	echo "         Eight SP indent" >>F &&
+	echo " 	HT and SP indent" >>F &&
+	echo "With trailing SP " >>F &&
+	echo "No problem" >>F
+
+'
+
+blue_grep='7;34m' ;# ESC [ 7 ; 3 4 m
+
+test_expect_success default '
+
+	git diff --color >output
+	grep "$blue_grep" output >error
+	grep -v "$blue_grep" output >normal
+
+	grep Eight normal >/dev/null &&
+	grep HT error >/dev/null &&
+	grep With error >/dev/null &&
+	grep No normal >/dev/null
+
+'
+
+test_expect_success 'without -trail' '
+
+	git config core.whitespace -trail
+	git diff --color >output
+	grep "$blue_grep" output >error
+	grep -v "$blue_grep" output >normal
+
+	grep Eight normal >/dev/null &&
+	grep HT error >/dev/null &&
+	grep With normal >/dev/null &&
+	grep No normal >/dev/null
+
+'
+
+test_expect_success 'without -trail (attribute)' '
+
+	git config --unset core.whitespace
+	echo "F whitespace=-trail" >.gitattributes
+	git diff --color >output
+	grep "$blue_grep" output >error
+	grep -v "$blue_grep" output >normal
+
+	grep Eight normal >/dev/null &&
+	grep HT error >/dev/null &&
+	grep With normal >/dev/null &&
+	grep No normal >/dev/null
+
+'
+
+test_expect_success 'without -space' '
+
+	rm -f .gitattributes
+	git config core.whitespace -space
+	git diff --color >output
+	grep "$blue_grep" output >error
+	grep -v "$blue_grep" output >normal
+
+	grep Eight normal >/dev/null &&
+	grep HT normal >/dev/null &&
+	grep With error >/dev/null &&
+	grep No normal >/dev/null
+
+'
+
+test_expect_success 'without -space (attribute)' '
+
+	git config --unset core.whitespace
+	echo "F whitespace=-space" >.gitattributes
+	git diff --color >output
+	grep "$blue_grep" output >error
+	grep -v "$blue_grep" output >normal
+
+	grep Eight normal >/dev/null &&
+	grep HT normal >/dev/null &&
+	grep With error >/dev/null &&
+	grep No normal >/dev/null
+
+'
+
+test_expect_success 'with indent-non-tab only' '
+
+	rm -f .gitattributes
+	git config core.whitespace indent,-trailing,-space
+	git diff --color >output
+	grep "$blue_grep" output >error
+	grep -v "$blue_grep" output >normal
+
+	grep Eight error >/dev/null &&
+	grep HT normal >/dev/null &&
+	grep With normal >/dev/null &&
+	grep No normal >/dev/null
+
+'
+
+test_expect_success 'with indent-non-tab only (attribute)' '
+
+	git config --unset core.whitespace
+	echo "F whitespace=indent,-trailing,-space" >.gitattributes
+	git diff --color >output
+	grep "$blue_grep" output >error
+	grep -v "$blue_grep" output >normal
+
+	grep Eight error >/dev/null &&
+	grep HT normal >/dev/null &&
+	grep With normal >/dev/null &&
+	grep No normal >/dev/null
+
+'
+
+test_done

t/t4124-apply-ws-rule.sh

@@ -0,0 +1,151 @@
+#!/bin/sh
+
+test_description='core.whitespace rules and git-apply'
+
+. ./test-lib.sh
+
+prepare_test_file () {
+
+	# A line that has character X is touched iff RULE is in effect:
+	#       X  RULE
+	#   	!  trailing-space
+	#   	@  space-before-tab
+	#   	#  indent-with-non-tab
+	sed -e "s/_/ /g" -e "s/>/	/" <<-\EOF
+		An_SP in an ordinary line>and a HT.
+		>A HT.
+		_>A SP and a HT (@).
+		_>_A SP, a HT and a SP (@).
+		_______Seven SP.
+		________Eight SP (#).
+		_______>Seven SP and a HT (@).
+		________>Eight SP and a HT (@#).
+		_______>_Seven SP, a HT and a SP (@).
+		________>_Eight SP, a HT and a SP (@#).
+		_______________Fifteen SP (#).
+		_______________>Fifteen SP and a HT (@#).
+		________________Sixteen SP (#).
+		________________>Sixteen SP and a HT (@#).
+		_____a__Five SP, a non WS, two SP.
+		A line with a (!) trailing SP_
+		A line with a (!) trailing HT>
+	EOF
+}
+
+apply_patch () {
+	>target &&
+	sed -e "s|\([ab]\)/file|\1/target|" <patch |
+	git apply "$@"
+}
+
+test_fix () {
+
+	# fix should not barf
+	apply_patch --whitespace=fix || return 1
+
+	# find touched lines
+	diff file target | sed -n -e "s/^> //p" >fixed
+
+	# the changed lines are all expeced to change
+	fixed_cnt=$(wc -l <fixed)
+	case "$1" in
+	'') expect_cnt=$fixed_cnt ;;
+	?*) expect_cnt=$(grep "[$1]" <fixed | wc -l) ;;
+	esac
+	test $fixed_cnt -eq $expect_cnt || return 1
+
+	# and we are not missing anything
+	case "$1" in
+	'') expect_cnt=0 ;;
+	?*) expect_cnt=$(grep "[$1]" <file | wc -l) ;;
+	esac
+	test $fixed_cnt -eq $expect_cnt || return 1
+
+	# Get the patch actually applied
+	git diff-files -p target >fixed-patch
+	test -s fixed-patch && return 0
+
+	# Make sure it is complaint-free
+	>target
+	git apply --whitespace=error-all <fixed-patch
+
+}
+
+test_expect_success setup '
+
+	>file &&
+	git add file &&
+	prepare_test_file >file &&
+	git diff-files -p >patch &&
+	>target &&
+	git add target
+
+'
+
+test_expect_success 'whitespace=nowarn, default rule' '
+
+	apply_patch --whitespace=nowarn &&
+	diff file target
+
+'
+
+test_expect_success 'whitespace=warn, default rule' '
+
+	apply_patch --whitespace=warn &&
+	diff file target
+
+'
+
+test_expect_success 'whitespace=error-all, default rule' '
+
+	apply_patch --whitespace=error-all && return 1
+	test -s target && return 1
+	: happy
+
+'
+
+test_expect_success 'whitespace=error-all, no rule' '
+
+	git config core.whitespace -trailing,-space-before,-indent &&
+	apply_patch --whitespace=error-all &&
+	diff file target
+
+'
+
+test_expect_success 'whitespace=error-all, no rule (attribute)' '
+
+	git config --unset core.whitespace &&
+	echo "target -whitespace" >.gitattributes &&
+	apply_patch --whitespace=error-all &&
+	diff file target
+
+'
+
+for t in - ''
+do
+	case "$t" in '') tt='!' ;; *) tt= ;; esac
+	for s in - ''
+	do
+		case "$s" in '') ts='@' ;; *) ts= ;; esac
+		for i in - ''
+		do
+			case "$i" in '') ti='#' ;; *) ti= ;; esac
+			rule=${t}trailing,${s}space,${i}indent
+
+			rm -f .gitattributes
+			test_expect_success "rule=$rule" '
+				git config core.whitespace "$rule" &&
+				test_fix "$tt$ts$ti"
+			'
+
+			test_expect_success "rule=$rule (attributes)" '
+				git config --unset core.whitespace &&
+				echo "target whitespace=$rule" >.gitattributes &&
+				test_fix "$tt$ts$ti"
+			'
+
+		done
+	done
+done
+
+test_done

ws.c

@@ -0,0 +1,96 @@
+/*
+ * Whitespace rules
+ *
+ * Copyright (c) 2007 Junio C Hamano
+ */
+
+#include "cache.h"
+#include "attr.h"
+
+static struct whitespace_rule {
+	const char *rule_name;
+	unsigned rule_bits;
+} whitespace_rule_names[] = {
+	{ "trailing-space", WS_TRAILING_SPACE },
+	{ "space-before-tab", WS_SPACE_BEFORE_TAB },
+	{ "indent-with-non-tab", WS_INDENT_WITH_NON_TAB },
+};
+
+unsigned parse_whitespace_rule(const char *string)
+{
+	unsigned rule = WS_DEFAULT_RULE;
+
+	while (string) {
+		int i;
+		size_t len;
+		const char *ep;
+		int negated = 0;
+
+		string = string + strspn(string, ", \t\n\r");
+		ep = strchr(string, ',');
+		if (!ep)
+			len = strlen(string);
+		else
+			len = ep - string;
+
+		if (*string == '-') {
+			negated = 1;
+			string++;
+			len--;
+		}
+		if (!len)
+			break;
+		for (i = 0; i < ARRAY_SIZE(whitespace_rule_names); i++) {
+			if (strncmp(whitespace_rule_names[i].rule_name,
+				    string, len))
+				continue;
+			if (negated)
+				rule &= ~whitespace_rule_names[i].rule_bits;
+			else
+				rule |= whitespace_rule_names[i].rule_bits;
+			break;
+		}
+		string = ep;
+	}
+	return rule;
+}
+
+static void setup_whitespace_attr_check(struct git_attr_check *check)
+{
+	static struct git_attr *attr_whitespace;
+
+	if (!attr_whitespace)
+		attr_whitespace = git_attr("whitespace", 10);
+	check[0].attr = attr_whitespace;
+}
+
+unsigned whitespace_rule(const char *pathname)
+{
+	struct git_attr_check attr_whitespace_rule;
+
+	setup_whitespace_attr_check(&attr_whitespace_rule);
+	if (!git_checkattr(pathname, 1, &attr_whitespace_rule)) {
+		const char *value;
+
+		value = attr_whitespace_rule.value;
+		if (ATTR_TRUE(value)) {
+			/* true (whitespace) */
+			unsigned all_rule = 0;
+			int i;
+			for (i = 0; i < ARRAY_SIZE(whitespace_rule_names); i++)
+				all_rule |= whitespace_rule_names[i].rule_bits;
+			return all_rule;
+		} else if (ATTR_FALSE(value)) {
+			/* false (-whitespace) */
+			return 0;
+		} else if (ATTR_UNSET(value)) {
+			/* reset to default (!whitespace) */
+			return whitespace_rule_cfg;
+		} else {
+			/* string */
+			return parse_whitespace_rule(value);
+		}
+	} else {
+		return whitespace_rule_cfg;
+	}
+}