Fix directory/path manipulation methods

The `dirname` and `dirbase` methods have been replaced with the Android
implementation, which is actually compilant to some kind of standard.

A new method `topdir` has been added, which returns the topmost
directory in a path.

These changes fix issue #49:

	`gitfo_prettify_dir_path` converts "./.git/" to ".git/", so
	the code at src/repository.c:190 goes out of bounds when
	trying to find the topmost directory.

	The new `git__topdir` method handles this gracefully, and the
	fixed `git__dirname` now returns the proper value for the
	repository's working dir.

	E.g.

		/repo/.git/ ==> working dir '/repo/'
		.git/		==> working dir '.'

Signed-off-by: Vicent Marti <tanoku@gmail.com>

src/odb_loose.c

@@ -59,7 +59,7 @@ static int make_temp_file(git_file *fd, char *tmp, size_t n, char *file)
 	size_t tmplen = strlen(template);
 	int dirlen;
 
-	if ((dirlen = git__dirname(tmp, n, file)) < 0)
+	if ((dirlen = git__dirname_r(tmp, n, file)) < 0)
 		return GIT_ERROR;
 
 	if ((dirlen + tmplen) >= n)

src/repository.c

@@ -157,7 +157,9 @@ static int assign_repository_DIRs(git_repository *repo,
 
 static int guess_repository_DIRs(git_repository *repo, const char *repository_path)
 {
-	char path_aux[GIT_PATH_MAX], *last_DIR;
+	char path_aux[GIT_PATH_MAX];
+	const char *topdir;
+
 	int path_len;
 	int error = GIT_SUCCESS;
 
@@ -185,12 +187,10 @@ static int guess_repository_DIRs(git_repository *repo, const char *repository_pa
 
 	path_aux[path_len] = 0;
 
-	last_DIR = (path_aux + path_len - 2);
+	if ((topdir = git__topdir(path_aux)) == NULL)
+		return GIT_EINVALIDPATH;
 
-	while (*last_DIR != '/')
-		last_DIR--;
-
-	if (strcmp(last_DIR, GIT_DIR) == 0) {
+	if (strcmp(topdir, GIT_DIR) == 0) {
 		repo->is_bare = 0;
 
 		/* index file */
@@ -198,8 +198,9 @@ static int guess_repository_DIRs(git_repository *repo, const char *repository_pa
 		repo->path_index = git__strdup(path_aux);
 
 		/* working dir */
-		*(last_DIR + 1) = 0;
-		repo->path_workdir = git__strdup(path_aux);
+		repo->path_workdir = git__dirname(path_aux);
+		if (repo->path_workdir == NULL)
+			return GIT_EINVALIDPATH;
 
 	} else {
 		repo->is_bare = 1;

src/util.c

@@ -36,65 +36,193 @@ int git__suffixcmp(const char *str, const char *suffix)
 	return strcmp(str + (a - b), suffix);
 }
 
-int git__dirname(char *dir, size_t n, char *path)
+/*
+ * Based on the Android implementation, BSD licensed.
+ * Check http://android.git.kernel.org/
+ */
+int git__basename_r(char *buffer, size_t bufflen, const char *path)
 {
-	char *s;
-	size_t len;
+	const char *endp, *startp;
+	int len, result;
+
+	/* Empty or NULL string gets treated as "." */
+	if (path == NULL || *path == '\0') {
+		startp  = ".";
+		len     = 1;
+		goto Exit;
+	}
 
-	assert(dir && n > 1);
+	/* Strip trailing slashes */
+	endp = path + strlen(path) - 1;
+	while (endp > path && *endp == '/')
+		endp--;
 
-	if (!path || !*path || (s = strrchr(path, '/')) == NULL) {
-		strcpy(dir, ".");
-		return 1;
+	/* All slashes becomes "/" */
+	if (endp == path && *endp == '/') {
+		startp = "/";
+		len    = 1;
+		goto Exit;
 	}
 
-	if (s == path) { /* "/[aaa]" */
-		strcpy(dir, "/");
-		return 1;
-	}
+	/* Find the start of the base */
+	startp = endp;
+	while (startp > path && *(startp - 1) != '/')
+		startp--;
 
-	if ((len = s - path) >= n)
-		return GIT_ERROR;
+	len = endp - startp +1;
 
-	memcpy(dir, path, len);
-	dir[len] = '\0';
+Exit:
+	result = len;
+	if (buffer == NULL) {
+		return result;
+	}
+	if (len > (int)bufflen-1) {
+		len    = (int)bufflen-1;
+		result = GIT_ENOMEM;
+	}
 
-	return len;
+	if (len >= 0) {
+		memcpy(buffer, startp, len);
+		buffer[len] = 0;
+	}
+	return result;
 }
 
-int git__basename(char *base, size_t n, char *path)
+/*
+ * Based on the Android implementation, BSD licensed.
+ * Check http://android.git.kernel.org/
+ */
+int git__dirname_r(char *buffer, size_t bufflen, const char *path)
 {
-	char *s;
-	size_t len;
+    const char *endp;
+    int result, len;
+
+    /* Empty or NULL string gets treated as "." */
+    if (path == NULL || *path == '\0') {
+        path = ".";
+        len  = 1;
+        goto Exit;
+    }
 
-	assert(base && n > 1);
+    /* Strip trailing slashes */
+    endp = path + strlen(path) - 1;
+    while (endp > path && *endp == '/')
+        endp--;
 
-	if (!path || !*path) {
-		strcpy(base, ".");
-		return 1;
+    /* Find the start of the dir */
+    while (endp > path && *endp != '/')
+        endp--;
+
+    /* Either the dir is "/" or there are no slashes */
+    if (endp == path) {
+        path = (*endp == '/') ? "/" : ".";
+        len  = 1;
+        goto Exit;
     }
-	len = strlen(path);
 
-	if ((s = strrchr(path, '/')) == NULL) {
-		if (len >= n)
-			return GIT_ERROR;
-		strcpy(base, path);
-		return len;
+    do {
+        endp--;
+    } while (endp > path && *endp == '/');
+
+    len = endp - path +1;
+
+Exit:
+    result = len;
+    if (len+1 > GIT_PATH_MAX) {
+        return GIT_ENOMEM;
     }
+    if (buffer == NULL)
+        return result;
 
-	if (s == path && len == 1) { /* "/" */
-		strcpy(base, "/");
-		return 1;
+    if (len > (int)bufflen-1) {
+        len    = (int)bufflen-1;
+        result = GIT_ENOMEM;
     }
 
-	len -= s - path;
-	if (len >= n)
-		return GIT_ERROR;
+    if (len >= 0) {
+        memcpy(buffer, path, len);
+        buffer[len] = 0;
+    }
+    return result;
+}
+
 
-	memcpy(base, s+1, len);
-	base[len] = '\0';
+char *git__dirname(const char *path)
+{
+    char *dname = NULL;
+    int len;
 
-	return len;
+	len = (path ? strlen(path) : 0) + 2;
+	dname = (char *)git__malloc(len);
+	if (dname == NULL)
+		return NULL;
+
+    if (git__dirname_r(dname, len, path) < GIT_SUCCESS) {
+		free(dname);
+		return NULL;
+	}
+
+    return dname;
+}
+
+char *git__basename(const char *path)
+{
+    char *bname = NULL;
+    int len;
+
+	len = (path ? strlen(path) : 0) + 2;
+	bname = (char *)git__malloc(len);
+	if (bname == NULL)
+		return NULL;
+
+    if (git__basename_r(bname, len, path) < GIT_SUCCESS) {
+		free(bname);
+		return NULL;
+	}
+
+    return bname;
+}
+
+
+const char *git__topdir(const char *path)
+{
+	size_t len;
+	int i;
+
+	assert(path);
+	len = strlen(path);
+
+	if (!len || path[len - 1] != '/')
+		return NULL;
+
+	for (i = len - 2; i >= 0; --i)
+		if (path[i] == '/')
+			break;
+
+	return &path[i + 1];
+}
+
+static char *strtok_raw(char *output, char *src, char *delimit, int keep)
+{
+	while (*src && strchr(delimit, *src) == NULL)
+		*output++ = *src++;
+
+	*output = 0;
+
+	if (keep)
+		return src;
+	else
+		return *src ? src+1 : src;
+}
+
+char *git__strtok(char *output, char *src, char *delimit)
+{
+	return strtok_raw(output, src, delimit, 0);
+}
+
+char *git__strtok_keep(char *output, char *src, char *delimit)
+{
+	return strtok_raw(output, src, delimit, 1);
 }
 
 void git__hexdump(const char *buffer, size_t len)

src/util.h

@@ -19,8 +19,44 @@ extern int git__fmt(char *, size_t, const char *, ...)
 extern int git__prefixcmp(const char *str, const char *prefix);
 extern int git__suffixcmp(const char *str, const char *suffix);
 
-extern int git__dirname(char *dir, size_t n, char *path);
-extern int git__basename(char *base, size_t n, char *path);
+/*
+ * The dirname() function shall take a pointer to a character string
+ * that contains a pathname, and return a pointer to a string that is a
+ * pathname of the parent directory of that file. Trailing '/' characters
+ * in the path are not counted as part of the path.
+ *
+ * If path does not contain a '/', then dirname() shall return a pointer to
+ * the string ".". If path is a null pointer or points to an empty string,
+ * dirname() shall return a pointer to the string "." .
+ *
+ * The `git__dirname` implementation is thread safe. The returned 
+ * string must be manually free'd.
+ *
+ * The `git__dirname_r` implementation expects a string allocated
+ * by the user with big enough size.
+ */
+extern char *git__dirname(const char *path);
+extern int git__dirname_r(char *buffer, size_t bufflen, const char *path);
+
+/*
+ * This function returns the basename of the file, which is the last
+ * part of its full name given by fname, with the drive letter and
+ * leading directories stripped off. For example, the basename of
+ * c:/foo/bar/file.ext is file.ext, and the basename of a:foo is foo.
+ *
+ * Trailing slashes and backslashes are significant: the basename of
+ * c:/foo/bar/ is an empty string after the rightmost slash.
+ *
+ * The `git__basename` implementation is thread safe. The returned 
+ * string must be manually free'd.
+ *
+ * The `git__basename_r` implementation expects a string allocated
+ * by the user with big enough size.
+ */
+extern char *git__basename(const char *path);
+extern int git__basename_r(char *buffer, size_t bufflen, const char *path);
+
+extern const char *git__topdir(const char *path);
 
 extern void git__hexdump(const char *buffer, size_t n);
 extern uint32_t git__hash(const void *key, int len, uint32_t seed);
@@ -40,6 +76,21 @@ GIT_INLINE(int) git__is_sizet(git_off_t p)
 #	define git__rotl(v, s) (uint32_t)(((uint32_t)(v) << (s)) | ((uint32_t)(v) >> (32 - (s))))
 #endif
 
+enum git_splitpath_flags
+{
+   GIT_SPL_PATH = 1,
+   GIT_SPL_FILE = 2,
+   GIT_SPL_EXT  = 4,
+   GIT_SPL_PATH_FILE = GIT_SPL_PATH + GIT_SPL_FILE,
+   GIT_SPL_FILE_EXT  = GIT_SPL_FILE + GIT_SPL_EXT,
+   GIT_SPL_EXT_NO_PERIOD = 8,
+};
+
+
+extern char *git__splitpath(char *path, int flag);
+extern char *git__strtok(char *output, char *src, char *delimit);
+extern char *git__strtok_keep(char *output, char *src, char *delimit);
+
 /*
  * Realloc the buffer pointed at by variable 'x' so that it can hold
  * at least 'nr' entries; the number of entries currently allocated

tests/t00-core.c

@@ -61,61 +61,76 @@ BEGIN_TEST("strutil", suffix_comparison)
 END_TEST
 
 BEGIN_TEST("strutil", dirname)
-	char dir[64];
-
-	must_be_true(!(git__dirname(dir, sizeof(dir), NULL) < 0));
-	must_be_true(!strcmp(dir, "."));
-
-	must_be_true(!(git__dirname(dir, sizeof(dir), "") < 0));
-	must_be_true(!strcmp(dir, "."));
-
-	must_be_true(!(git__dirname(dir, sizeof(dir), "a") < 0));
-	must_be_true(!strcmp(dir, "."));
-
-	must_be_true(!(git__dirname(dir, sizeof(dir), "/") < 0));
-	must_be_true(!strcmp(dir, "/"));
-
-	must_be_true(!(git__dirname(dir, sizeof(dir), "/usr") < 0));
-	must_be_true(!strcmp(dir, "/"));
+	char dir[64], *dir2;
+
+#define DIRNAME_TEST(A, B) { \
+	must_be_true(git__dirname_r(dir, sizeof(dir), A) >= 0); \
+	must_be_true(strcmp(dir, B) == 0);				\
+	must_be_true((dir2 = git__dirname(A)) != NULL);	\
+	must_be_true(strcmp(dir2, B) == 0);				\
+	free(dir2);										\
+}
 
-	/* TODO: should this be "/" instead (ie strip trailing / first) */
-	must_be_true(!(git__dirname(dir, sizeof(dir), "/usr/") < 0));
-	must_be_true(!strcmp(dir, "/usr"));
+	DIRNAME_TEST(NULL, ".");
+	DIRNAME_TEST("", ".");
+	DIRNAME_TEST("a", ".");
+	DIRNAME_TEST("/", "/");
+	DIRNAME_TEST("/usr", "/");
+	DIRNAME_TEST("/usr/", "/");
+	DIRNAME_TEST("/usr/lib", "/usr");
+	DIRNAME_TEST("usr/lib", "usr");
+	DIRNAME_TEST(".git/", ".");
 
-	must_be_true(!(git__dirname(dir, sizeof(dir), "/usr/lib") < 0));
-	must_be_true(!strcmp(dir, "/usr"));
+#undef DIRNAME_TEST
 
-	must_be_true(!(git__dirname(dir, sizeof(dir), "usr/lib") < 0));
-	must_be_true(!strcmp(dir, "usr"));
 END_TEST
 
 BEGIN_TEST("strutil", basename)
-	char base[64];
+	char base[64], *base2;
+
+#define BASENAME_TEST(A, B) { \
+	must_be_true(git__basename_r(base, sizeof(base), A) >= 0); \
+	must_be_true(strcmp(base, B) == 0);					\
+	must_be_true((base2 = git__basename(A)) != NULL);	\
+	must_be_true(strcmp(base2, B) == 0);				\
+	free(base2);										\
+}
 
-	must_be_true(!(git__basename(base, sizeof(base), NULL) < 0));
-	must_be_true(!strcmp(base, "."));
+	BASENAME_TEST(NULL, ".");
+	BASENAME_TEST("", ".");
+	BASENAME_TEST("a", "a");
+	BASENAME_TEST("/", "/");
+	BASENAME_TEST("/usr", "usr");
+	BASENAME_TEST("/usr/", "usr");
+	BASENAME_TEST("/usr/lib", "lib");
+	BASENAME_TEST("usr/lib", "lib");
 
-	must_be_true(!(git__basename(base, sizeof(base), "") < 0));
-	must_be_true(!strcmp(base, "."));
+#undef BASENAME_TEST
 
-	must_be_true(!(git__basename(base, sizeof(base), "a") < 0));
-	must_be_true(!strcmp(base, "a"));
+END_TEST
 
-	must_be_true(!(git__basename(base, sizeof(base), "/") < 0));
-	must_be_true(!strcmp(base, "/"));
+BEGIN_TEST("strutil", topdir)
+	const char *dir;
 
-	must_be_true(!(git__basename(base, sizeof(base), "/usr") < 0));
-	must_be_true(!strcmp(base, "usr"));
+#define TOPDIR_TEST(A, B) { \
+	must_be_true((dir = git__topdir(A)) != NULL);	\
+	must_be_true(strcmp(dir, B) == 0);				\
+}
 
-	/* TODO: should this be "usr" instead (ie strip trailing / first) */
-	must_be_true(!(git__basename(base, sizeof(base), "/usr/") < 0));
-	must_be_true(!strcmp(base, ""));
+	TOPDIR_TEST(".git/", ".git/");
+	TOPDIR_TEST("/.git/", ".git/");
+	TOPDIR_TEST("usr/local/.git/", ".git/");
+	TOPDIR_TEST("./.git/", ".git/");
+	TOPDIR_TEST("/usr/.git/", ".git/");
+	TOPDIR_TEST("/", "/");
+	TOPDIR_TEST("a/", "a/");
 
-	must_be_true(!(git__basename(base, sizeof(base), "/usr/lib") < 0));
-	must_be_true(!strcmp(base, "lib"));
+	must_be_true(git__topdir("/usr/.git") == NULL);
+	must_be_true(git__topdir(".") == NULL);
+	must_be_true(git__topdir("") == NULL);
+	must_be_true(git__topdir("a") == NULL);
 
-	must_be_true(!(git__basename(base, sizeof(base), "usr/lib") < 0));
-	must_be_true(!strcmp(base, "lib"));
+#undef TOPDIR_TEST
 END_TEST
 
 /* Initial size of 1 will cause writing past array bounds prior to fix */
@@ -579,6 +594,7 @@ git_testsuite *libgit2_suite_core(void)
 	ADD_TEST(suite, "strutil", suffix_comparison);
 	ADD_TEST(suite, "strutil", dirname);
 	ADD_TEST(suite, "strutil", basename);
+	ADD_TEST(suite, "strutil", topdir);
 
 	ADD_TEST(suite, "vector", initial_size_one);
 	ADD_TEST(suite, "vector", remove);