Pass UTF8 filename to fz_open_document / fz_open_file to remove kludges.

Use _wopen on a UTF8 -> wchar_t decoded filename to support UTF8 filenames
for win32.

7 files changed, 33 insertions, 50 deletions

diff --git a/apps/pdfapp.c b/apps/pdfapp.c
index 66513570..52bb4c48 100644
--- a/apps/pdfapp.c
+++ b/apps/pdfapp.c
@@ -89,26 +89,14 @@ void pdfapp_invert(pdfapp_t *app, fz_bbox rect)
 	fz_invert_pixmap_rect(app->image, rect);
 }
 
-void pdfapp_open(pdfapp_t *app, char *filename, int fd, int reload)
+void pdfapp_open(pdfapp_t *app, char *filename, int reload)
 {
 	fz_context *ctx = app->ctx;
-	fz_stream *file;
 	char *password = "";
 
 	fz_try(ctx)
 	{
-		file = fz_open_fd(ctx, fd);
-
-		if (strstr(filename, ".rels"))
-			app->doc = (fz_document*) xps_open_document(ctx, filename);
-		else if (strstr(filename, ".xps") || strstr(filename, ".XPS") || strstr(filename, ".rels"))
-			app->doc = (fz_document*) xps_open_document_with_stream(file);
-		else if (strstr(filename, ".cbz") || strstr(filename, ".CBZ") || strstr(filename, ".zip") || strstr(filename, ".ZIP"))
-			app->doc = (fz_document*) cbz_open_document_with_stream(file);
-		else
-			app->doc = (fz_document*) pdf_open_document_with_stream(file);
-
-		fz_close(file);
+		app->doc = fz_open_document(ctx, filename);
 
 		if (fz_needs_password(app->doc))
 		{
diff --git a/apps/pdfapp.h b/apps/pdfapp.h
index 6812323b..f93b816a 100644
--- a/apps/pdfapp.h
+++ b/apps/pdfapp.h
@@ -100,7 +100,7 @@ struct pdfapp_s
 };
 
 void pdfapp_init(fz_context *ctx, pdfapp_t *app);
-void pdfapp_open(pdfapp_t *app, char *filename, int fd, int reload);
+void pdfapp_open(pdfapp_t *app, char *filename, int reload);
 void pdfapp_close(pdfapp_t *app);
 
 char *pdfapp_version(pdfapp_t *app);
diff --git a/apps/win_main.c b/apps/win_main.c
index cc71a120..358b21bb 100644
--- a/apps/win_main.c
+++ b/apps/win_main.c
@@ -594,15 +594,8 @@ void windocopy(pdfapp_t *app)
 
 void winreloadfile(pdfapp_t *app)
 {
-	int fd;
-
 	pdfapp_close(app);
-
-	fd = _wopen(wbuf, O_BINARY | O_RDONLY, 0666);
-	if (fd < 0)
-		winerror(&gapp, "cannot reload file");
-
-	pdfapp_open(app, filename, fd, 1);
+	pdfapp_open(app, filename, 1);
 }
 
 void winopenuri(pdfapp_t *app, char *buf)
@@ -834,7 +827,6 @@ WinMain(HINSTANCE hInstance, HINSTANCE hPrevInstance, LPSTR lpCmdLine, int nShow
 	LPWSTR *argv = CommandLineToArgvW(GetCommandLineW(), &argc);
 	char argv0[256];
 	MSG msg;
-	int fd;
 	int code;
 	fz_context *ctx;
 
@@ -861,15 +853,11 @@ WinMain(HINSTANCE hInstance, HINSTANCE hPrevInstance, LPSTR lpCmdLine, int nShow
 			exit(0);
 	}
 
-	fd = _wopen(wbuf, O_BINARY | O_RDONLY, 0666);
-	if (fd < 0)
-		winerror(&gapp, "cannot open file");
-
 	code = WideCharToMultiByte(CP_UTF8, 0, wbuf, -1, filename, sizeof filename, NULL, NULL);
 	if (code == 0)
 		winerror(&gapp, "cannot convert filename to utf-8");
 
-	pdfapp_open(&gapp, filename, fd, 0);
+	pdfapp_open(&gapp, filename, 0);
 
 	while (GetMessage(&msg, NULL, 0, 0))
 	{
diff --git a/apps/x11_main.c b/apps/x11_main.c
index b62f97db..6ab5b43a 100644
--- a/apps/x11_main.c
+++ b/apps/x11_main.c
@@ -532,15 +532,8 @@ void onselreq(Window requestor, Atom selection, Atom target, Atom property, Time
 
 void winreloadfile(pdfapp_t *app)
 {
-	int fd;
-
 	pdfapp_close(app);
-
-	fd = open(filename, O_RDONLY, 0666);
-	if (fd < 0)
-		fz_throw(gapp.ctx, "cannot reload file '%s'", filename);
-
-	pdfapp_open(app, filename, fd, 1);
+	pdfapp_open(app, filename, 1);
 }
 
 void winopenuri(pdfapp_t *app, char *buf)
@@ -612,7 +605,6 @@ int main(int argc, char **argv)
 	int oldy = 0;
 	int resolution = -1;
 	int pageno = 1;
-	int fd;
 	fd_set fds;
 	int width = -1;
 	int height = -1;
@@ -663,11 +655,7 @@ int main(int argc, char **argv)
 	gapp.resolution = resolution;
 	gapp.pageno = pageno;
 
-	fd = open(filename, O_RDONLY, 0666);
-	if (fd < 0)
-		fz_throw(gapp.ctx, "cannot open file '%s'", filename);
-
-	pdfapp_open(&gapp, filename, fd, 0);
+	pdfapp_open(&gapp, filename, 0);
 
 	FD_ZERO(&fds);
 
diff --git a/fitz/doc_document.c b/fitz/doc_document.c
index 15ad4b62..22229892 100644
--- a/fitz/doc_document.c
+++ b/fitz/doc_document.c
@@ -27,7 +27,7 @@ fz_document *
 fz_open_document(fz_context *ctx, char *filename)
 {
 	char *ext = strrchr(filename, '.');
-	if (ext && !fz_strcasecmp(ext, ".xps"))
+	if (ext && (!fz_strcasecmp(ext, ".xps") || !fz_strcasecmp(ext, ".rels")))
 		return (fz_document*) xps_open_document(ctx, filename);
 	if (ext && !fz_strcasecmp(ext, ".cbz"))
 		return (fz_document*) cbz_open_document(ctx, filename);
diff --git a/fitz/fitz.h b/fitz/fitz.h
index 48b67ff8..9e0c0311 100644
--- a/fitz/fitz.h
+++ b/fitz/fitz.h
@@ -534,20 +534,20 @@ int fz_strlcpy(char *dst, const char *src, int n);
 int fz_strlcat(char *dst, const char *src, int n);
 
 /*
-	fz_chartorune: UTF8 decode a string of chars to a rune.
+	fz_chartorune: UTF8 decode a single rune from a sequence of chars.
 
 	rune: Pointer to an int to assign the decoded 'rune' to.
 
-	str: Pointer to a UTF8 encoded string
+	str: Pointer to a UTF8 encoded string.
 
 	Returns the number of bytes consumed. Does not throw exceptions.
 */
 int fz_chartorune(int *rune, char *str);
 
 /*
-	runetochar: UTF8 encode a run to a string of chars.
+	fz_runetochar: UTF8 encode a rune to a sequence of chars.
 
-	str: Pointer to a place to put the UTF8 encoded string.
+	str: Pointer to a place to put the UTF8 encoded character.
 
 	rune: Pointer to a 'rune'.
 
@@ -557,7 +557,7 @@ int fz_chartorune(int *rune, char *str);
 int fz_runetochar(char *str, int rune);
 
 /*
-	fz_runelen: Count many chars are required to represent a rune.
+	fz_runelen: Count how many chars are required to represent a rune.
 
 	rune: The rune to encode.
 
@@ -986,7 +986,12 @@ typedef struct fz_stream_s fz_stream;
 /*
 	fz_open_file: Open the named file and wrap it in a stream.
 
-	filename: Path to a file as it would be given to open(2).
+	filename: Path to a file. On non-Windows machines the filename should
+	be exactly as it would be passed to open(2). On Windows machines, the
+	path should be UTF-8 encoded so that non-ASCII characters can be
+	represented. Other platforms do the encoding as standard anyway (and
+	in most cases, particularly for MacOS and Linux, the encoding they
+	use is UTF-8 anyway).
 */
 fz_stream *fz_open_file(fz_context *ctx, const char *filename);
 
diff --git a/fitz/stm_open.c b/fitz/stm_open.c
index 1e303825..ced32d80 100644
--- a/fitz/stm_open.c
+++ b/fitz/stm_open.c
@@ -129,7 +129,21 @@ fz_open_fd(fz_context *ctx, int fd)
 fz_stream *
 fz_open_file(fz_context *ctx, const char *name)
 {
+#ifdef _WIN32
+	char *s = (char*)name;
+	wchar_t *wname, *d;
+	int c, fd;
+	d = wname = fz_malloc(ctx, (strlen(name)+1) * sizeof(wchar_t));
+	while (*s) {
+		s += fz_chartorune(&c, s);
+		*d++ = c;
+	}
+	*d = 0;
+	fd = _wopen(wname, O_BINARY | O_RDONLY, 0);
+	fz_free(ctx, wname);
+#else
 	int fd = open(name, O_BINARY | O_RDONLY, 0);
+#endif
 	if (fd == -1)
 		fz_throw(ctx, "cannot open %s", name);
 	return fz_open_fd(ctx, fd);