Issue python#9425: Create PyModule_GetFilenameObject() function

Victor Stinner · Victor Stinner · commit 6c00c1464f03 · 2010-08-17T23:37:11.000Z
... to get the filename as a unicode object, instead of a byte string. Function
needed to support unencodable filenames. Deprecate PyModule_GetFilename() in
favor on the new function.
diff --git a/Doc/c-api/module.rst b/Doc/c-api/module.rst
@@ -64,13 +64,24 @@ There are only a few functions special to module objects.
 
 .. cfunction:: char* PyModule_GetFilename(PyObject *module)
 
+   Similar to :cfunc:`PyModule_GetFilenameObject` but return the filename
+   encoded to 'utf-8'.
+
+   .. deprecated:: 3.2
+      :cfunc:`PyModule_GetFilename` raises :ctype:`UnicodeEncodeError` on
+      unencodable filenames, use :cfunc:`PyModule_GetFilenameObject` instead.
+
+
+.. cfunction:: PyObject* PyModule_GetFilenameObject(PyObject *module)
+
    .. index::
       single: __file__ (module attribute)
       single: SystemError (built-in exception)
 
    Return the name of the file from which *module* was loaded using *module*'s
-   :attr:`__file__` attribute.  If this is not defined, or if it is not a string,
-   raise :exc:`SystemError` and return *NULL*.
+   :attr:`__file__` attribute.  If this is not defined, or if it is not a
+   unicode string, raise :exc:`SystemError` and return *NULL*; otherwise return
+   a reference to a :ctype:`PyUnicodeObject`.
 
 
 .. cfunction:: void* PyModule_GetState(PyObject *module)
diff --git a/Include/moduleobject.h b/Include/moduleobject.h
@@ -16,6 +16,7 @@ PyAPI_FUNC(PyObject *) PyModule_New(const char *);
 PyAPI_FUNC(PyObject *) PyModule_GetDict(PyObject *);
 PyAPI_FUNC(const char *) PyModule_GetName(PyObject *);
 PyAPI_FUNC(const char *) PyModule_GetFilename(PyObject *);
+PyAPI_FUNC(PyObject *) PyModule_GetFilenameObject(PyObject *);
 PyAPI_FUNC(void) _PyModule_Clear(PyObject *);
 PyAPI_FUNC(struct PyModuleDef*) PyModule_GetDef(PyObject*);
 PyAPI_FUNC(void*) PyModule_GetState(PyObject*);
diff --git a/Misc/NEWS b/Misc/NEWS
@@ -12,6 +12,11 @@ What's New in Python 3.2 Alpha 2?
 Core and Builtins
 -----------------
 
+- Issue #9425: Create PyModule_GetFilenameObject() function to get the filename
+  as a unicode object, instead of a byte string. Function needed to support
+  unencodable filenames. Deprecate PyModule_GetFilename() in favor on the new
+  function.
+
 - Issue #8063: Call _PyGILState_Init() earlier in Py_InitializeEx().
 
 - Issue #9612: The set object is now 64-bit clean under Windows.
diff --git a/Objects/moduleobject.c b/Objects/moduleobject.c
@@ -188,8 +188,8 @@ PyModule_GetName(PyObject *m)
     return _PyUnicode_AsString(nameobj);
 }
 
-static PyObject*
-module_getfilename(PyObject *m)
+PyObject*
+PyModule_GetFilenameObject(PyObject *m)
 {
     PyObject *d;
     PyObject *fileobj;
@@ -205,17 +205,21 @@ module_getfilename(PyObject *m)
         PyErr_SetString(PyExc_SystemError, "module filename missing");
         return NULL;
     }
+    Py_INCREF(fileobj);
     return fileobj;
 }
 
 const char *
 PyModule_GetFilename(PyObject *m)
 {
     PyObject *fileobj;
-    fileobj = module_getfilename(m);
+    char *utf8;
+    fileobj = PyModule_GetFilenameObject(m);
     if (fileobj == NULL)
         return NULL;
-    return _PyUnicode_AsString(fileobj);
+    utf8 = _PyUnicode_AsString(fileobj);
+    Py_DECREF(fileobj);
+    return utf8;
 }
 
 PyModuleDef*
@@ -346,19 +350,21 @@ static PyObject *
 module_repr(PyModuleObject *m)
 {
     const char *name;
-    PyObject *filename;
+    PyObject *filename, *repr;
 
     name = PyModule_GetName((PyObject *)m);
     if (name == NULL) {
         PyErr_Clear();
         name = "?";
     }
-    filename = module_getfilename((PyObject *)m);
+    filename = PyModule_GetFilenameObject((PyObject *)m);
     if (filename == NULL) {
         PyErr_Clear();
         return PyUnicode_FromFormat("<module '%s' (built-in)>", name);
     }
-    return PyUnicode_FromFormat("<module '%s' from '%U'>", name, filename);
+    repr = PyUnicode_FromFormat("<module '%s' from '%U'>", name, filename);
+    Py_DECREF(filename);
+    return repr;
 }
 
 static int

Original file line number	Diff line number	Diff line change
`@@ -188,8 +188,8 @@ PyModule_GetName(PyObject *m)`
`188`	`188`	`return _PyUnicode_AsString(nameobj);`
`189`	`189`	`}`
`190`	`190`
`191`		`-static PyObject*`
`192`		`-module_getfilename(PyObject *m)`
	`191`	`+PyObject*`
	`192`	`+PyModule_GetFilenameObject(PyObject *m)`
`193`	`193`	`{`
`194`	`194`	`PyObject *d;`
`195`	`195`	`PyObject *fileobj;`
`@@ -205,17 +205,21 @@ module_getfilename(PyObject *m)`
`205`	`205`	`PyErr_SetString(PyExc_SystemError, "module filename missing");`
`206`	`206`	`return NULL;`
`207`	`207`	`}`
	`208`	`+ Py_INCREF(fileobj);`
`208`	`209`	`return fileobj;`
`209`	`210`	`}`
`210`	`211`
`211`	`212`	`const char *`
`212`	`213`	`PyModule_GetFilename(PyObject *m)`
`213`	`214`	`{`
`214`	`215`	`PyObject *fileobj;`
`215`		`- fileobj = module_getfilename(m);`
	`216`	`+ char *utf8;`
	`217`	`+ fileobj = PyModule_GetFilenameObject(m);`
`216`	`218`	`if (fileobj == NULL)`
`217`	`219`	`return NULL;`
`218`		`- return _PyUnicode_AsString(fileobj);`
	`220`	`+ utf8 = _PyUnicode_AsString(fileobj);`
	`221`	`+ Py_DECREF(fileobj);`
	`222`	`+ return utf8;`
`219`	`223`	`}`
`220`	`224`
`221`	`225`	`PyModuleDef*`
`@@ -346,19 +350,21 @@ static PyObject *`
`346`	`350`	`module_repr(PyModuleObject *m)`
`347`	`351`	`{`
`348`	`352`	`const char *name;`
`349`		`- PyObject *filename;`
	`353`	`+ PyObject filename, repr;`
`350`	`354`
`351`	`355`	`name = PyModule_GetName((PyObject *)m);`
`352`	`356`	`if (name == NULL) {`
`353`	`357`	`PyErr_Clear();`
`354`	`358`	`name = "?";`
`355`	`359`	`}`
`356`		`- filename = module_getfilename((PyObject *)m);`
	`360`	`+ filename = PyModule_GetFilenameObject((PyObject *)m);`
`357`	`361`	`if (filename == NULL) {`
`358`	`362`	`PyErr_Clear();`
`359`	`363`	`return PyUnicode_FromFormat("<module '%s' (built-in)>", name);`
`360`	`364`	`}`
`361`		`- return PyUnicode_FromFormat("<module '%s' from '%U'>", name, filename);`
	`365`	`+ repr = PyUnicode_FromFormat("<module '%s' from '%U'>", name, filename);`
	`366`	`+ Py_DECREF(filename);`
	`367`	`+ return repr;`
`362`	`368`	`}`
`363`	`369`
`364`	`370`	`static int`