[BUG, PATCH 0/3] Fix {blame,cat-file} --textconv for cases with symlinks

@@ -17,10 +17,12 @@ chmod +x helper
 test_expect_success 'setup ' '
 	echo "bin: test 1" >one.bin &&
 	echo "bin: test number 2" >two.bin &&
+	ln -s one.bin symlink.bin &&
 	git add . &&
 	GIT_AUTHOR_NAME=Number1 git commit -a -m First --date="2010-01-01 18:00:00" &&
 	echo "bin: test 1 version 2" >one.bin &&
 	echo "bin: test number 2 version 2" >>two.bin &&
+	ln -sf two.bin symlink.bin &&
 	GIT_AUTHOR_NAME=Number2 git commit -a -m Second --date="2010-01-01 20:00:00"
 '
 

@@ -78,4 +80,46 @@ test_expect_success 'blame from previous revision' '
 	test_cmp expected result
 '
 
+cat >expected <<EOF
+(Number2 2010-01-01 20:00:00 +0000 1) two.bin
+EOF
+
+test_expect_success 'blame with --no-textconv (on symlink)' '
+	git blame --no-textconv symlink.bin >blame &&
+	find_blame <blame >result &&
+	test_cmp expected result
+'
+
+# fails with '...symlink.bin is not "binary" file'
+test_expect_failure 'blame --textconv (on symlink)' '
+	git blame --textconv symlink.bin >blame &&
+	find_blame <blame >result &&
+	test_cmp expected result
+'
+
+# cp two.bin three.bin  and make small tweak
+# (this will direct blame -C -C three.bin to consider two.bin and symlink.bin)
+test_expect_success 'make another new commit' '
+	echo "bin: test number 2" >three.bin &&
+	echo "bin: test number 2 version 2" >>three.bin &&
+	echo "bin: test number 2 version 3" >>three.bin &&
+	echo "bin: test number 3" >>three.bin &&
+	git add three.bin &&
+	GIT_AUTHOR_NAME=Number4 git commit -a -m Fourth --date="2010-01-01 23:00:00"
+'
+
+cat >expected <<EOF
+(Number1 2010-01-01 18:00:00 +0000 1) converted: test number 2
+(Number2 2010-01-01 20:00:00 +0000 2) converted: test number 2 version 2
+(Number3 2010-01-01 22:00:00 +0000 3) converted: test number 2 version 3
+(Number4 2010-01-01 23:00:00 +0000 4) converted: test number 3
+EOF
+
+# fails with '...symlink.bin is not "binary" file'
+test_expect_failure 'blame on last commit (-C -C, symlink)' '
+	git blame -C -C three.bin >blame &&
+	find_blame <blame >result &&
+	test_cmp expected result
+'
+
 test_done

@@ -12,6 +12,7 @@ chmod +x helper
 
 test_expect_success 'setup ' '
 	echo "bin: test" >one.bin &&
+	ln -s one.bin symlink.bin &&
 	git add . &&
 	GIT_AUTHOR_NAME=Number1 git commit -a -m First --date="2010-01-01 18:00:00" &&
 	echo "bin: test version 2" >one.bin &&

@@ -68,4 +69,32 @@ test_expect_success 'cat-file --textconv on previous commit' '
 	git cat-file --textconv HEAD^:one.bin >result &&
 	test_cmp expected result
 '
+
+echo -n "one.bin" >expected
+
+test_expect_success 'cat-file without --textconv (symlink)' '
+	git cat-file blob :symlink.bin >result &&
+	test_cmp expected result
+'
+
+cat >expected <<EOF
+fatal: git cat-file --textconv: unable to run textconv on :symlink.bin
+EOF
+
+# fails because cat-file tries to run converter on symlink.bin
+test_expect_failure 'cat-file --textconv on index (symlink)' '
+	! git cat-file --textconv :symlink.bin 2>result &&
+	test_cmp expected result
+'
+
+cat >expected <<EOF
+fatal: git cat-file --textconv: unable to run textconv on HEAD:symlink.bin
+EOF
+
+# fails because cat-file tries to run converter on symlink.bin
+test_expect_failure 'cat-file --textconv on HEAD (symlink)' '
+	! git cat-file --textconv HEAD:symlink.bin 2>result &&
+	test_cmp expected result
+'
+
 test_done

@@ -5,18 +5,19 @@ test_description='test textconv caching'
 
 cat >helper <<'EOF'
 #!/bin/sh
-sed 's/^/converted: /' "$@" >helper.out
+grep -q '^bin: ' "$@" || { echo "E: $@ is not \"binary\" file" 1>&2; exit 1; }
+sed 's/^bin:/converted:/' "$@" >helper.out
 cat helper.out
 EOF
 chmod +x helper
 
 test_expect_success 'setup' '
-	echo foo content 1 >foo.bin &&
-	echo bar content 1 >bar.bin &&
+	echo "bin: foo content 1" >foo.bin &&
+	echo "bin: bar content 1" >bar.bin &&
 	git add . &&
 	git commit -m one &&
-	echo foo content 2 >foo.bin &&
-	echo bar content 2 >bar.bin &&
+	echo "bin: foo content 2" >foo.bin &&
+	echo "bin: bar content 2" >bar.bin &&
 	git commit -a -m two &&
 	echo "*.bin diff=magic" >.gitattributes &&
 	git config diff.magic.textconv ./helper &&

@@ -25,14 +26,14 @@ test_expect_success 'setup' '
 
 cat >expect <<EOF
 diff --git a/bar.bin b/bar.bin
-index fcf9166..28283d5 100644
+index 628fb83..f64d847 100644
 --- a/bar.bin
 +++ b/bar.bin
 @@ -1 +1 @@
 -converted: bar content 1
 +converted: bar content 2
 diff --git a/foo.bin b/foo.bin
-index d5b9fe3..1345db2 100644
+index 255496b..ad450ff 100644
 --- a/foo.bin
 +++ b/foo.bin
 @@ -1 +1 @@

@@ -59,7 +60,7 @@ test_expect_success 'cached textconv does not run helper' '
 
 cat >expect <<EOF
 diff --git a/bar.bin b/bar.bin
-index fcf9166..28283d5 100644
+index 628fb83..f64d847 100644
 --- a/bar.bin
 +++ b/bar.bin
 @@ -1,2 +1,2 @@

@@ -67,7 +68,7 @@ index fcf9166..28283d5 100644
 -converted: bar content 1
 +converted: bar content 2
 diff --git a/foo.bin b/foo.bin
-index d5b9fe3..1345db2 100644
+index 255496b..ad450ff 100644
 --- a/foo.bin
 +++ b/foo.bin
 @@ -1,2 +1,2 @@

@@ -76,7 +77,7 @@ index d5b9fe3..1345db2 100644
 +converted: foo content 2
 EOF
 test_expect_success 'changing textconv invalidates cache' '
-	echo other >other &&
+	echo "bin: other" >other &&
 	git config diff.magic.textconv "./helper other" &&
 	git diff HEAD^ HEAD >actual &&
 	test_cmp expect actual

@@ -84,7 +85,7 @@ test_expect_success 'changing textconv invalidates cache' '
 
 cat >expect <<EOF
 diff --git a/bar.bin b/bar.bin
-index fcf9166..28283d5 100644
+index 628fb83..f64d847 100644
 --- a/bar.bin
 +++ b/bar.bin
 @@ -1,2 +1,2 @@

@@ -92,7 +93,7 @@ index fcf9166..28283d5 100644
 -converted: bar content 1
 +converted: bar content 2
 diff --git a/foo.bin b/foo.bin
-index d5b9fe3..1345db2 100644
+index 255496b..ad450ff 100644
 --- a/foo.bin
 +++ b/foo.bin
 @@ -1 +1 @@

@@ -9,22 +9,23 @@ find_blame() {
 
 cat >helper <<'EOF'
 #!/bin/sh
-sed 's/^/converted: /' "$@"
+grep -q '^bin: ' "$@" || { echo "E: $@ is not \"binary\" file" 1>&2; exit 1; }
+sed 's/^bin:/converted:/' "$@"
 EOF
 chmod +x helper
 
 test_expect_success 'setup ' '
-	echo test 1 >one.bin &&
-	echo test number 2 >two.bin &&
+	echo "bin: test 1" >one.bin &&
+	echo "bin: test number 2" >two.bin &&
 	git add . &&
 	GIT_AUTHOR_NAME=Number1 git commit -a -m First --date="2010-01-01 18:00:00" &&
-	echo test 1 version 2 >one.bin &&
-	echo test number 2 version 2 >>two.bin &&
+	echo "bin: test 1 version 2" >one.bin &&
+	echo "bin: test number 2 version 2" >>two.bin &&
 	GIT_AUTHOR_NAME=Number2 git commit -a -m Second --date="2010-01-01 20:00:00"
 '
 
 cat >expected <<EOF
-(Number2 2010-01-01 20:00:00 +0000 1) test 1 version 2
+(Number2 2010-01-01 20:00:00 +0000 1) bin: test 1 version 2
 EOF
 
 test_expect_success 'no filter specified' '

@@ -67,7 +68,7 @@ test_expect_success 'blame --textconv going through revisions' '
 '
 
 test_expect_success 'make a new commit' '
-	echo "test number 2 version 3" >>two.bin &&
+	echo "bin: test number 2 version 3" >>two.bin &&
 	GIT_AUTHOR_NAME=Number3 git commit -a -m Third --date="2010-01-01 22:00:00"
 '
 

@@ -5,15 +5,16 @@ test_description='git cat-file textconv support'
 
 cat >helper <<'EOF'
 #!/bin/sh
-sed 's/^/converted: /' "$@"
+grep -q '^bin: ' "$@" || { echo "E: $@ is not \"binary\" file" 1>&2; exit 1; }
+sed 's/^bin:/converted:/' "$@"
 EOF
 chmod +x helper
 
 test_expect_success 'setup ' '
-	echo test >one.bin &&
+	echo "bin: test" >one.bin &&
 	git add . &&
 	GIT_AUTHOR_NAME=Number1 git commit -a -m First --date="2010-01-01 18:00:00" &&
-	echo test version 2 >one.bin &&
+	echo "bin: test version 2" >one.bin &&
 	GIT_AUTHOR_NAME=Number2 git commit -a -m Second --date="2010-01-01 20:00:00"
 '
 

@@ -33,7 +34,7 @@ test_expect_success 'setup textconv filters' '
 '
 
 cat >expected <<EOF
-test version 2
+bin: test version 2
 EOF
 
 test_expect_success 'cat-file without --textconv' '

@@ -42,7 +43,7 @@ test_expect_success 'cat-file without --textconv' '
 '
 
 cat >expected <<EOF
-test
+bin: test
 EOF
 
 test_expect_success 'cat-file without --textconv on previous commit' '

@@ -35,7 +35,7 @@ void finish_copy_notes_for_rewrite(struct notes_rewrite_cfg *c);
 
 extern int check_pager_config(const char *cmd);
 
-extern int textconv_object(const char *path, const unsigned char *sha1, char **buf, unsigned long *buf_size);
+extern int textconv_object(const char *path, unsigned mode, const unsigned char *sha1, char **buf, unsigned long *buf_size);
 
 extern int cmd_add(int argc, const char **argv, const char *prefix);
 extern int cmd_annotate(int argc, const char **argv, const char *prefix);

@@ -83,6 +83,7 @@ struct origin {
 	struct commit *commit;
 	mmfile_t file;
 	unsigned char blob_sha1[20];
+	unsigned mode;
 	char path[FLEX_ARRAY];
 };
 

@@ -92,6 +93,7 @@ struct origin {
  * Return 1 if the conversion succeeds, 0 otherwise.
  */
 int textconv_object(const char *path,
+		    unsigned mode,
 		    const unsigned char *sha1,
 		    char **buf,
 		    unsigned long *buf_size)

@@ -100,7 +102,7 @@ int textconv_object(const char *path,
 	struct userdiff_driver *textconv;
 
 	df = alloc_filespec(path);
-	fill_filespec(df, sha1, S_IFREG | 0664);
+	fill_filespec(df, sha1, mode);
 	textconv = get_textconv(df);
 	if (!textconv) {
 		free_filespec(df);

@@ -125,7 +127,7 @@ static void fill_origin_blob(struct diff_options *opt,
 
 		num_read_blob++;
 		if (DIFF_OPT_TST(opt, ALLOW_TEXTCONV) &&
-		    textconv_object(o->path, o->blob_sha1, &file->ptr, &file_size))
+		    textconv_object(o->path, o->mode, o->blob_sha1, &file->ptr, &file_size))
 			;
 		else
 			file->ptr = read_sha1_file(o->blob_sha1, &type, &file_size);

@@ -313,21 +315,23 @@ static struct origin *get_origin(struct scoreboard *sb,
  * for an origin is also used to pass the blame for the entire file to
  * the parent to detect the case where a child's blob is identical to
  * that of its parent's.
+ *
+ * This also fills origin->mode for correspoinding tree path.
  */
-static int fill_blob_sha1(struct origin *origin)
+static int fill_blob_sha1_and_mode(struct origin *origin)
 {
-	unsigned mode;
 	if (!is_null_sha1(origin->blob_sha1))
 		return 0;
 	if (get_tree_entry(origin->commit->object.sha1,
 			   origin->path,
-			   origin->blob_sha1, &mode))
+			   origin->blob_sha1, &origin->mode))
 		goto error_out;
 	if (sha1_object_info(origin->blob_sha1, NULL) != OBJ_BLOB)
 		goto error_out;
 	return 0;
  error_out:
 	hashclr(origin->blob_sha1);
+	origin->mode = S_IFINVALID;
 	return -1;
 }
 

@@ -360,12 +364,14 @@ static struct origin *find_origin(struct scoreboard *sb,
 			/*
 			 * If the origin was newly created (i.e. get_origin
 			 * would call make_origin if none is found in the
-			 * scoreboard), it does not know the blob_sha1,
+			 * scoreboard), it does not know the blob_sha1/mode,
 			 * so copy it.  Otherwise porigin was in the
-			 * scoreboard and already knows blob_sha1.
+			 * scoreboard and already knows blob_sha1/mode.
 			 */
-			if (porigin->refcnt == 1)
+			if (porigin->refcnt == 1) {
 				hashcpy(porigin->blob_sha1, cached->blob_sha1);
+				porigin->mode = cached->mode;
+			}
 			return porigin;
 		}
 		/* otherwise it was not very useful; free it */

@@ -400,6 +406,7 @@ static struct origin *find_origin(struct scoreboard *sb,
 		/* The path is the same as parent */
 		porigin = get_origin(sb, parent, origin->path);
 		hashcpy(porigin->blob_sha1, origin->blob_sha1);
+		porigin->mode = origin->mode;
 	} else {
 		/*
 		 * Since origin->path is a pathspec, if the parent

@@ -425,6 +432,7 @@ static struct origin *find_origin(struct scoreboard *sb,
 		case 'M':
 			porigin = get_origin(sb, parent, origin->path);
 			hashcpy(porigin->blob_sha1, p->one->sha1);
+			porigin->mode = p->one->mode;
 			break;
 		case 'A':
 		case 'T':

@@ -444,6 +452,7 @@ static struct origin *find_origin(struct scoreboard *sb,
 
 		cached = make_origin(porigin->commit, porigin->path);
 		hashcpy(cached->blob_sha1, porigin->blob_sha1);
+		cached->mode = porigin->mode;
 		parent->util = cached;
 	}
 	return porigin;

@@ -486,6 +495,7 @@ static struct origin *find_rename(struct scoreboard *sb,
 		    !strcmp(p->two->path, origin->path)) {
 			porigin = get_origin(sb, parent, p->one->path);
 			hashcpy(porigin->blob_sha1, p->one->sha1);
+			porigin->mode = p->one->mode;
 			break;
 		}
 	}

@@ -1099,6 +1109,7 @@ static int find_copy_in_parent(struct scoreboard *sb,
 
 			norigin = get_origin(sb, parent, p->one->path);
 			hashcpy(norigin->blob_sha1, p->one->sha1);
+			norigin->mode = p->one->mode;
 			fill_origin_blob(&sb->revs->diffopt, norigin, &file_p);
 			if (!file_p.ptr)
 				continue;

@@ -2075,7 +2086,7 @@ static struct commit *fake_working_tree_commit(struct diff_options *opt,
 		switch (st.st_mode & S_IFMT) {
 		case S_IFREG:
 			if (DIFF_OPT_TST(opt, ALLOW_TEXTCONV) &&
-			    textconv_object(read_from, null_sha1, &buf.buf, &buf_len))
+			    textconv_object(read_from, mode, null_sha1, &buf.buf, &buf_len))
 				buf.len = buf_len;
 			else if (strbuf_read_file(&buf, read_from, st.st_size) != st.st_size)
 				die_errno("cannot open or read '%s'", read_from);

@@ -2455,11 +2466,11 @@ parse_done:
 	}
 	else {
 		o = get_origin(&sb, sb.final, path);
-		if (fill_blob_sha1(o))
+		if (fill_blob_sha1_and_mode(o))
 			die("no such path %s in %s", path, final_commit_name);
 
 		if (DIFF_OPT_TST(&sb.revs->diffopt, ALLOW_TEXTCONV) &&
-		    textconv_object(path, o->blob_sha1, (char **) &sb.final_buf,
+		    textconv_object(path, o->mode, o->blob_sha1, (char **) &sb.final_buf,
 				    &sb.final_buf_size))
 			;
 		else

@@ -143,7 +143,7 @@ static int cat_one_file(int opt, const char *exp_type, const char *obj_name)
 			die("git cat-file --textconv %s: <object> must be <sha1:path>",
 			    obj_name);
 
-		if (!textconv_object(obj_context.path, sha1, &buf, &size))
+		if (!textconv_object(obj_context.path, obj_context.mode, sha1, &buf, &size))
 			die("git cat-file --textconv: unable to run textconv on %s",
 			    obj_name);
 		break;

@@ -1068,7 +1068,7 @@ int get_sha1_with_context_1(const char *name, unsigned char *sha1,
 		struct cache_entry *ce;
 		int pos;
 		if (namelen > 2 && name[1] == '/')
-			return get_sha1_oneline(name + 2, sha1);
+			return get_sha1_oneline(name + 2, sha1);    /* XXX also mode? */
 		if (namelen < 3 ||
 		    name[2] != ':' ||
 		    name[1] < '0' || '3' < name[1])

@@ -1095,6 +1095,7 @@ int get_sha1_with_context_1(const char *name, unsigned char *sha1,
 				break;
 			if (ce_stage(ce) == stage) {
 				hashcpy(sha1, ce->sha1);
+				oc->mode = ce->ce_mode; /* XXX ok? */
 				return 0;
 			}
 			pos++;

@@ -90,8 +90,7 @@ test_expect_success 'blame with --no-textconv (on symlink)' '
 	test_cmp expected result
 '
 
-# fails with '...symlink.bin is not "binary" file'
-test_expect_failure 'blame --textconv (on symlink)' '
+test_expect_success 'blame --textconv (on symlink)' '
 	git blame --textconv symlink.bin >blame &&
 	find_blame <blame >result &&
 	test_cmp expected result

@@ -115,8 +114,7 @@ cat >expected <<EOF
 (Number4 2010-01-01 23:00:00 +0000 4) converted: test number 3
 EOF
 
-# fails with '...symlink.bin is not "binary" file'
-test_expect_failure 'blame on last commit (-C -C, symlink)' '
+test_expect_success 'blame on last commit (-C -C, symlink)' '
 	git blame -C -C three.bin >blame &&
 	find_blame <blame >result &&
 	test_cmp expected result

@@ -81,8 +81,7 @@ cat >expected <<EOF
 fatal: git cat-file --textconv: unable to run textconv on :symlink.bin
 EOF
 
-# fails because cat-file tries to run converter on symlink.bin
-test_expect_failure 'cat-file --textconv on index (symlink)' '
+test_expect_success 'cat-file --textconv on index (symlink)' '
 	! git cat-file --textconv :symlink.bin 2>result &&
 	test_cmp expected result
 '

@@ -91,8 +90,7 @@ cat >expected <<EOF
 fatal: git cat-file --textconv: unable to run textconv on HEAD:symlink.bin
 EOF
 
-# fails because cat-file tries to run converter on symlink.bin
-test_expect_failure 'cat-file --textconv on HEAD (symlink)' '
+test_expect_success 'cat-file --textconv on HEAD (symlink)' '
 	! git cat-file --textconv HEAD:symlink.bin 2>result &&
 	test_cmp expected result
 '

@@ -1062,6 +1062,7 @@ int get_sha1_with_context_1(const char *name, unsigned char *sha1,
 	/* sha1:path --> object name of path in ent sha1
 	 * :path -> object name of path in index
 	 * :[0-3]:path -> object name of path in index at stage
+	 * :/foo -> last commit whose subject starts with foo
 	 */
 	if (name[0] == ':') {
 		int stage = 0;