Patch #445762: Support --disable-unicode

- Do not compile unicodeobject, unicodectype, and unicodedata if Unicode is disabled
- check for Py_USING_UNICODE in all places that use Unicode functions
- disables unicode literals, and the builtin functions
- add the types.StringTypes list
- remove Unicode literals from most tests.

5 files changed, 45 insertions, 13 deletions

diff --git a/Modules/_codecsmodule.c b/Modules/_codecsmodule.c
index 61f25d1..a085bcf 100644
--- a/Modules/_codecsmodule.c
+++ b/Modules/_codecsmodule.c
@@ -71,6 +71,7 @@ PyObject *codeclookup(PyObject *self, PyObject *args)
     return NULL;
 }
 
+#ifdef Py_USING_UNICODE
 /* --- Helpers ------------------------------------------------------------ */
 
 static
@@ -621,12 +622,14 @@ mbcs_encode(PyObject *self,
 }
 
 #endif /* MS_WIN32 */
+#endif /* Py_USING_UNICODE */
 
 /* --- Module API --------------------------------------------------------- */
 
 static PyMethodDef _codecs_functions[] = {
     {"register",		codecregister,			1},
     {"lookup",			codeclookup, 			1},
+#ifdef Py_USING_UNICODE
     {"utf_8_encode",		utf_8_encode,			1},
     {"utf_8_decode",		utf_8_decode,			1},
     {"utf_16_encode",		utf_16_encode,			1},
@@ -654,6 +657,7 @@ static PyMethodDef _codecs_functions[] = {
     {"mbcs_encode", 		mbcs_encode,			1},
     {"mbcs_decode", 		mbcs_decode,			1},
 #endif
+#endif /* Py_USING_UNICODE */
     {NULL, NULL}		/* sentinel */
 };
 
diff --git a/Modules/_sre.c b/Modules/_sre.c
index 1776a16..9943c30 100644
--- a/Modules/_sre.c
+++ b/Modules/_sre.c
@@ -63,7 +63,7 @@ static char copyright[] =
 /* defining this one enables tracing */
 #undef VERBOSE
 
-#if PY_VERSION_HEX >= 0x01060000
+#if PY_VERSION_HEX >= 0x01060000 && defined(Py_USING_UNICODE)
 /* defining this enables unicode support (default under 1.6a1 and later) */
 #define HAVE_UNICODE
 #endif
diff --git a/Modules/_tkinter.c b/Modules/_tkinter.c
index 9b73307..eedb0c1 100644
--- a/Modules/_tkinter.c
+++ b/Modules/_tkinter.c
@@ -255,6 +255,7 @@ AsString(PyObject *value, PyObject *tmp)
 {
 	if (PyString_Check(value))
 		return PyString_AsString(value);
+#ifdef Py_USING_UNICODE
 	else if (PyUnicode_Check(value)) {
 		PyObject *v = PyUnicode_AsUTF8String(value);
 		if (v == NULL)
@@ -266,6 +267,7 @@ AsString(PyObject *value, PyObject *tmp)
 		Py_DECREF(v);
 		return PyString_AsString(v);
 	}
+#endif
 	else {
 		PyObject *v = PyObject_Str(value);
 		if (v == NULL)
@@ -520,6 +522,7 @@ AsObj(PyObject *value)
 		ckfree(FREECAST argv);
 		return result;
 	}
+#ifdef Py_USING_UNICODE
 	else if (PyUnicode_Check(value)) {
 #if TKMAJORMINOR <= 8001
 		/* In Tcl 8.1 we must use UTF-8 */
@@ -542,6 +545,7 @@ AsObj(PyObject *value)
 					 PyUnicode_GET_SIZE(value));
 #endif /* TKMAJORMINOR > 8001 */
 	}
+#endif
 	else {
 		PyObject *v = PyObject_Str(value);
 		if (!v)
@@ -616,13 +620,16 @@ Tkapp_Call(PyObject *self, PyObject *args)
 		   so would confuse applications that expect a string. */
 		char *s = Tcl_GetStringResult(interp);
 		char *p = s;
+
 		/* If the result contains any bytes with the top bit set,
 		   it's UTF-8 and we should decode it to Unicode */
+#ifdef Py_USING_UNICODE
 		while (*p != '\0') {
 			if (*p & 0x80)
 				break;
 			p++;
 		}
+
 		if (*p == '\0')
 			res = PyString_FromStringAndSize(s, (int)(p-s));
 		else {
@@ -634,6 +641,10 @@ Tkapp_Call(PyObject *self, PyObject *args)
 			    res = PyString_FromStringAndSize(s, (int)(p-s));
 			}
 		}
+#else
+		p = strchr(p, '\0');
+		res = PyString_FromStringAndSize(s, (int)(p-s));
+#endif
 	}
 
 	LEAVE_OVERLAP_TCL
diff --git a/Modules/cPickle.c b/Modules/cPickle.c
index b27339f..bb0d281 100644
--- a/Modules/cPickle.c
+++ b/Modules/cPickle.c
@@ -1172,6 +1172,7 @@ err:
 }
 
 
+#ifdef Py_USING_UNICODE
 /* A copy of PyUnicode_EncodeRawUnicodeEscape() that also translates
    backslash and newline characters to \uXXXX escapes. */
 static PyObject *
@@ -1289,6 +1290,7 @@ err:
     Py_XDECREF(repr);
     return -1;
 }
+#endif
 
 
 static int
@@ -1824,11 +1826,13 @@ save(Picklerobject *self, PyObject *args, int  pers_save) {
                 goto finally;
             }
 
+#ifdef Py_USING_UNICODE
         case 'u':
             if ((type == &PyUnicode_Type) && (PyString_GET_SIZE(args) < 2)) {
                 res = save_unicode(self, args, 0);
                 goto finally;
             }
+#endif
     }
 
     if (args->ob_refcnt > 1) {
@@ -1857,12 +1861,14 @@ save(Picklerobject *self, PyObject *args, int  pers_save) {
             }
             break;
 
+#ifdef Py_USING_UNICODE
         case 'u':
             if (type == &PyUnicode_Type) {
                 res = save_unicode(self, args, 1);
                 goto finally;
             }
             break;
+#endif
 
         case 't':
             if (type == &PyTuple_Type) {
@@ -2818,6 +2824,7 @@ load_short_binstring(Unpicklerobject *self) {
 }
 
 
+#ifdef Py_USING_UNICODE
 static int
 load_unicode(Unpicklerobject *self) {
     PyObject *str = 0;
@@ -2836,8 +2843,10 @@ load_unicode(Unpicklerobject *self) {
 finally:
     return res;
 }
+#endif
 
 
+#ifdef Py_USING_UNICODE
 static int
 load_binunicode(Unpicklerobject *self) {
     PyObject *unicode;
@@ -2857,6 +2866,7 @@ load_binunicode(Unpicklerobject *self) {
     PDATA_PUSH(self->stack, unicode, -1);
     return 0;
 }
+#endif
 
 
 static int
@@ -3615,6 +3625,7 @@ load(Unpicklerobject *self) {
                     break;
                 continue;
 
+#ifdef Py_USING_UNICODE
             case UNICODE:
                 if (load_unicode(self) < 0)
                     break;
@@ -3624,6 +3635,7 @@ load(Unpicklerobject *self) {
                 if (load_binunicode(self) < 0)
                     break;
                 continue;
+#endif
 
             case EMPTY_TUPLE:
                 if (load_empty_tuple(self) < 0)
@@ -3905,6 +3917,7 @@ noload(Unpicklerobject *self) {
                     break;
                 continue;
 
+#ifdef Py_USING_UNICODE
             case UNICODE:
                 if (load_unicode(self) < 0)
                     break;
@@ -3914,6 +3927,7 @@ noload(Unpicklerobject *self) {
                 if (load_binunicode(self) < 0)
                     break;
                 continue;
+#endif
 
             case EMPTY_TUPLE:
                 if (load_empty_tuple(self) < 0)
diff --git a/Modules/pyexpat.c b/Modules/pyexpat.c
index 4bddc46..3311093 100644
--- a/Modules/pyexpat.c
+++ b/Modules/pyexpat.c
@@ -28,6 +28,11 @@
 #define Py_TPFLAGS_GC 0
 #endif
 
+#if (PY_MAJOR_VERSION == 1 && PY_MINOR_VERSION > 5) || (PY_MAJOR_VERSION == 2 && PY_MINOR_VERSION < 2)
+/* In Python 1.6, 2.0 and  2.1, disabling Unicode was not possible. */
+#define Py_USING_UNICODE
+#endif
+
 enum HandlerTypes {
     StartElement,
     EndElement,
@@ -173,7 +178,7 @@ conv_atts_using_string(XML_Char **atts)
 }
 #endif
 
-#if !(PY_MAJOR_VERSION == 1 && PY_MINOR_VERSION < 6)
+#ifdef Py_USING_UNICODE
 #if EXPAT_VERSION == 0x010200
 static PyObject *
 conv_atts_using_unicode(XML_Char **atts)
@@ -370,7 +375,7 @@ call_with_frame(PyCodeObject *c, PyObject* func, PyObject* args)
     return res;
 }
 
-#if PY_MAJOR_VERSION == 1 && PY_MINOR_VERSION < 6
+#ifndef Py_USING_UNICODE
 #define STRING_CONV_FUNC conv_string_to_utf8
 #else
 /* Python 1.6 and later versions */
@@ -506,7 +511,7 @@ VOID_HANDLER(ProcessingInstruction,
               const XML_Char *data),
              ("(O&O&)",STRING_CONV_FUNC,target, STRING_CONV_FUNC,data))
 
-#if PY_MAJOR_VERSION == 1 && PY_MINOR_VERSION < 6
+#ifndef Py_USING_UNICODE
 VOID_HANDLER(CharacterData, 
              (void *userData, const XML_Char *data, int len), 
              ("(N)", conv_string_len_to_utf8(data,len)))
@@ -531,7 +536,7 @@ VOID_HANDLER(UnparsedEntityDecl,
               STRING_CONV_FUNC,notationName))
 
 #if EXPAT_VERSION >= 0x015f00
-#if PY_MAJOR_VERSION == 1 && PY_MINOR_VERSION < 6
+#ifndef Py_USING_UNICODE
 VOID_HANDLER(EntityDecl,
              (void *userData,
               const XML_Char *entityName,
@@ -608,7 +613,7 @@ conv_content_model_utf8(XML_Content * const model)
     return conv_content_model(model, conv_string_to_utf8);
 }
 
-#if !(PY_MAJOR_VERSION == 1 && PY_MINOR_VERSION < 6)
+#ifdef Py_USING_UNICODE
 static PyObject *
 conv_content_model_unicode(XML_Content * const model)
 {
@@ -678,7 +683,7 @@ VOID_HANDLER(EndCdataSection,
                (void *userData),
 		("()"))
 
-#if PY_MAJOR_VERSION == 1 && PY_MINOR_VERSION < 6
+#ifndef Py_USING_UNICODE
 VOID_HANDLER(Default,
 	      (void *userData,  const XML_Char *s, int len),
 	      ("(N)", conv_string_len_to_utf8(s,len)))
@@ -1064,7 +1069,7 @@ static struct PyMethodDef xmlparse_methods[] = {
 /* ---------- */
 
 
-#if !(PY_MAJOR_VERSION == 1 && PY_MINOR_VERSION < 6)
+#ifdef Py_USING_UNICODE
 
 /* 
     pyexpat international encoding support.
@@ -1158,8 +1163,7 @@ newxmlparseobject(char *encoding, char *namespace_separator)
         return NULL;
     }
     XML_SetUserData(self->itself, (void *)self);
-#if PY_MAJOR_VERSION == 1 && PY_MINOR_VERSION < 6
-#else
+#ifdef Py_USING_UNICODE
     XML_SetUnknownEncodingHandler(self->itself, (XML_UnknownEncodingHandler) PyUnknownEncodingHandler, NULL);
 #endif
 
@@ -1292,7 +1296,7 @@ xmlparse_setattr(xmlparseobject *self, char *name, PyObject *v)
     }
     if (strcmp(name, "returns_unicode") == 0) {
         if (PyObject_IsTrue(v)) {
-#if PY_MAJOR_VERSION == 1 && PY_MINOR_VERSION < 6
+#ifndef Py_USING_UNICODE
             PyErr_SetString(PyExc_ValueError, 
                             "Cannot return Unicode strings in Python 1.5");
             return -1;
@@ -1545,8 +1549,7 @@ MODULE_INITFUNC(void)
                                          info.minor, info.micro));
     }
 #endif
-#if PY_MAJOR_VERSION == 1 && PY_MINOR_VERSION < 6
-#else
+#ifdef Py_USING_UNICODE
     init_template_buffer();
 #endif
     /* XXX When Expat supports some way of figuring out how it was