Add the unicodedata module

Makefile

@@ -291,7 +291,7 @@ STDLIB_OBJS := stdlib.bc.o stdlib.stripped.bc.o
 STDLIB_RELEASE_OBJS := stdlib.release.bc.o
 ASM_SRCS := $(wildcard src/runtime/*.S)
 
-STDMODULE_SRCS := errnomodule.c shamodule.c sha256module.c sha512module.c _math.c mathmodule.c md5.c md5module.c _randommodule.c _sre.c operator.c binascii.c pwdmodule.c posixmodule.c _struct.c datetimemodule.c _functoolsmodule.c _collectionsmodule.c itertoolsmodule.c resource.c signalmodule.c selectmodule.c fcntlmodule.c timemodule.c arraymodule.c zlibmodule.c _codecsmodule.c socketmodule.c $(EXTRA_STDMODULE_SRCS)
+STDMODULE_SRCS := errnomodule.c shamodule.c sha256module.c sha512module.c _math.c mathmodule.c md5.c md5module.c _randommodule.c _sre.c operator.c binascii.c pwdmodule.c posixmodule.c _struct.c datetimemodule.c _functoolsmodule.c _collectionsmodule.c itertoolsmodule.c resource.c signalmodule.c selectmodule.c fcntlmodule.c timemodule.c arraymodule.c zlibmodule.c _codecsmodule.c socketmodule.c unicodedata.c $(EXTRA_STDMODULE_SRCS)
 STDOBJECT_SRCS := structseq.c capsule.c stringobject.c exceptions.c unicodeobject.c unicodectype.c bytearrayobject.c bytes_methods.c $(EXTRA_STDOBJECT_SRCS)
 STDPYTHON_SRCS := pyctype.c getargs.c formatter_string.c pystrtod.c dtoa.c formatter_unicode.c $(EXTRA_STDPYTHON_SRCS)
 FROM_CPYTHON_SRCS := $(addprefix from_cpython/Modules/,$(STDMODULE_SRCS)) $(addprefix from_cpython/Objects/,$(STDOBJECT_SRCS)) $(addprefix from_cpython/Python/,$(STDPYTHON_SRCS))

from_cpython/CMakeLists.txt

@@ -15,7 +15,7 @@ endforeach(STDLIB_FILE)
 add_custom_target(copy_stdlib ALL DEPENDS ${STDLIB_TARGETS})
 
 # compile specified files in from_cpython/Modules
-file(GLOB_RECURSE STDMODULE_SRCS Modules errnomodule.c shamodule.c sha256module.c sha512module.c _math.c mathmodule.c md5.c md5module.c _randommodule.c _sre.c operator.c binascii.c pwdmodule.c posixmodule.c _struct.c datetimemodule.c _functoolsmodule.c _collectionsmodule.c itertoolsmodule.c resource.c signalmodule.c selectmodule.c fcntlmodule.c timemodule.c arraymodule.c zlibmodule.c _codecsmodule.c socketmodule.c)
+file(GLOB_RECURSE STDMODULE_SRCS Modules errnomodule.c shamodule.c sha256module.c sha512module.c _math.c mathmodule.c md5.c md5module.c _randommodule.c _sre.c operator.c binascii.c pwdmodule.c posixmodule.c _struct.c datetimemodule.c _functoolsmodule.c _collectionsmodule.c itertoolsmodule.c resource.c signalmodule.c selectmodule.c fcntlmodule.c timemodule.c arraymodule.c zlibmodule.c _codecsmodule.c socketmodule.c unicodedata.c)
 
 # compile specified files in from_cpython/Objects
 file(GLOB_RECURSE STDOBJECT_SRCS Objects structseq.c capsule.c stringobject.c exceptions.c unicodeobject.c unicodectype.c bytearrayobject.c bytes_methods.c)

from_cpython/Modules/unicodedata.c

@@ -1252,6 +1252,10 @@ initunicodedata(void)
 
     Py_TYPE(&UCD_Type) = &PyType_Type;
 
+    // Pyston change: let the GC know about our type
+    if (PyType_Ready(&UCD_Type))
+        return;
+
     m = Py_InitModule3(
         "unicodedata", unicodedata_functions, unicodedata_docstring);
     if (!m)

src/capi/modsupport.cpp

@@ -289,9 +289,8 @@ extern "C" PyObject* Py_InitModule4(const char* name, PyMethodDef* methods, cons
                                     int apiver) noexcept {
     BoxedModule* module = createModule(name, "__builtin__");
 
+    // Pass self as is, even if NULL we are not allowed to change it to None
     Box* passthrough = static_cast<Box*>(self);
-    if (!passthrough)
-        passthrough = None;
 
     while (methods && methods->ml_name) {
         if (VERBOSITY())

src/runtime/types.cpp

@@ -65,6 +65,7 @@ extern "C" void initzlib();
 extern "C" void init_codecs();
 extern "C" void init_socket();
 extern "C" void _PyUnicode_Init();
+extern "C" void initunicodedata();
 
 namespace pyston {
 
@@ -1445,6 +1446,7 @@ void setupRuntime() {
     initzlib();
     init_codecs();
     init_socket();
+    initunicodedata();
 
     setupSysEnd();
 

test/tests/unicodedata_test.py

+# skip-if: '-x' in EXTRA_JIT_ARGS
+import unicodedata
+print unicodedata.lookup("EURO SIGN") == u"\u20ac"
+print unicodedata.name(u"/")
+print unicodedata.category(u"A")