def descr__new__(space, w_unicodetype, w_string='', w_encoding=None, w_errors=None):
# NB. the default value of w_string is really a *wrapped* empty string:
# there is gateway magic at work
from pypy.objspace.std.unicodeobject import W_UnicodeObject
w_obj = w_string
w_obj_type = space.type(w_obj)
if space.is_w(w_obj_type, space.w_unicode):
if (not space.is_w(w_encoding, space.w_None) or
not space.is_w(w_errors, space.w_None)):
raise OperationError(space.w_TypeError,
space.wrap('decoding Unicode is not supported'))
if space.is_w(w_unicodetype, space.w_unicode):
return w_obj
w_value = w_obj
elif (space.is_w(w_encoding, space.w_None) and
space.is_w(w_errors, space.w_None)):
if space.is_true(space.isinstance(w_obj, space.w_str)):
w_value = unicode_from_string(space, w_obj)
elif space.is_true(space.isinstance(w_obj, space.w_unicode)):
w_value = w_obj
else:
w_value = unicode_from_object(space, w_obj)
else:
w_value = unicode_from_encoded_object(space, w_obj, w_encoding, w_errors)
# help the annotator! also the ._value depends on W_UnicodeObject layout
assert isinstance(w_value, W_UnicodeObject)
w_newobj = space.allocate_instance(W_UnicodeObject, w_unicodetype)
W_UnicodeObject.__init__(w_newobj, w_value._value)
return w_newobj
def descr_new_(space, w_unicodetype, w_string, w_encoding=None, w_errors=None):
# NB. the default value of w_obj is really a *wrapped* empty string:
# there is gateway magic at work
from pypy.objspace.std.unicodeobject import W_UnicodeObject
w_obj = w_string
encoding, errors = _get_encoding_and_errors(space, w_encoding, w_errors)
# convoluted logic for the case when unicode subclass has a __unicode__
# method, we need to call this method
is_precisely_unicode = space.is_w(space.type(w_obj), space.w_unicode)
if (is_precisely_unicode or
(space.isinstance_w(w_obj, space.w_unicode) and
space.findattr(w_obj, space.wrap('__unicode__')) is None)):
if encoding is not None or errors is not None:
raise OperationError(space.w_TypeError,
space.wrap('decoding Unicode is not supported'))
w_value = w_obj
if is_precisely_unicode and space.is_w(w_unicodetype, space.w_unicode):
return w_value
else:
if encoding is None and errors is None:
w_value = unicode_from_object(space, w_obj)
else:
w_value = unicode_from_encoded_object(space, w_obj,
encoding, errors)
if space.is_w(w_unicodetype, space.w_unicode):
return w_value
assert isinstance(w_value, W_UnicodeObject)
w_newobj = space.allocate_instance(W_UnicodeObject, w_unicodetype)
W_UnicodeObject.__init__(w_newobj, w_value._value)
return w_newobj
def descr__new__(space, w_unicodetype, w_string='', w_encoding=None, w_errors=None):
# NB. the default value of w_obj is really a *wrapped* empty string:
# there is gateway magic at work
from pypy.objspace.std.unicodeobject import W_UnicodeObject
from pypy.objspace.std.ropeunicodeobject import W_RopeUnicodeObject
w_obj = w_string
w_obj_type = space.type(w_obj)
encoding, errors = _get_encoding_and_errors(space, w_encoding, w_errors)
if space.is_w(w_obj_type, space.w_unicode):
if encoding is not None or errors is not None:
raise OperationError(space.w_TypeError,
space.wrap('decoding Unicode is not supported'))
if space.is_w(w_unicodetype, space.w_unicode):
return w_obj
w_value = w_obj
elif encoding is None and errors is None:
if space.is_true(space.isinstance(w_obj, space.w_str)):
w_value = unicode_from_string(space, w_obj)
elif space.is_true(space.isinstance(w_obj, space.w_unicode)):
w_value = w_obj
else:
w_value = unicode_from_object(space, w_obj)
else:
w_value = unicode_from_encoded_object(space, w_obj, encoding, errors)
if space.config.objspace.std.withropeunicode:
assert isinstance(w_value, W_RopeUnicodeObject)
w_newobj = space.allocate_instance(W_RopeUnicodeObject, w_unicodetype)
W_RopeUnicodeObject.__init__(w_newobj, w_value._node)
return w_newobj
assert isinstance(w_value, W_UnicodeObject)
w_newobj = space.allocate_instance(W_UnicodeObject, w_unicodetype)
W_UnicodeObject.__init__(w_newobj, w_value._value)
return w_newobj
def descr_new_(space, w_unicodetype, w_string='', w_encoding=None, w_errors=None):
# NB. the default value of w_obj is really a *wrapped* empty string:
# there is gateway magic at work
from pypy.objspace.std.unicodeobject import W_UnicodeObject
from pypy.objspace.std.ropeunicodeobject import W_RopeUnicodeObject
w_obj = w_string
encoding, errors = _get_encoding_and_errors(space, w_encoding, w_errors)
if space.is_true(space.isinstance(w_obj, space.w_unicode)):
if encoding is not None or errors is not None:
raise OperationError(space.w_TypeError,
space.wrap('decoding Unicode is not supported'))
w_value = w_obj
else:
if encoding is None and errors is None:
if space.is_true(space.isinstance(w_obj, space.w_str)):
w_value = unicode_from_string(space, w_obj)
else:
w_value = unicode_from_object(space, w_obj)
else:
w_value = unicode_from_encoded_object(space, w_obj,
encoding, errors)
if space.is_w(w_unicodetype, space.w_unicode):
return w_value
if space.config.objspace.std.withropeunicode:
assert isinstance(w_value, W_RopeUnicodeObject)
w_newobj = space.allocate_instance(W_RopeUnicodeObject, w_unicodetype)
W_RopeUnicodeObject.__init__(w_newobj, w_value._node)
return w_newobj
assert isinstance(w_value, W_UnicodeObject)
w_newobj = space.allocate_instance(W_UnicodeObject, w_unicodetype)
W_UnicodeObject.__init__(w_newobj, w_value._value)
return w_newobj
def descr_append_slice(self, space, w_s, start, end):
w_unicode = W_UnicodeObject.convert_arg_to_w_unicode(space, w_s)
if not 0 <= start <= end <= w_unicode._len():
raise oefmt(space.w_ValueError, "bad start/stop")
byte_start = w_unicode._index_to_byte(start)
byte_end = w_unicode._index_to_byte(end)
self.builder.append_slice(w_unicode._utf8, byte_start, byte_end)
def descr_append(self, space, w_s):
if isinstance(w_s, W_UnicodeObject):
self.builder.append_utf8(w_s._utf8, w_s._len())
else:
w_unicode = W_UnicodeObject.convert_arg_to_w_unicode(space, w_s)
s = space.utf8_w(w_unicode)
self.builder.append(s)
def unicode_from_string(space, w_str):
# this is a performance and bootstrapping hack
encoding = getdefaultencoding(space)
from pypy.objspace.std.unicodeobject import W_UnicodeObject
if encoding != 'ascii':
return unicode_from_encoded_object(space, w_str, encoding, "strict")
s = space.str_w(w_str)
try:
return W_UnicodeObject(s.decode("ascii"))
except UnicodeDecodeError:
# raising UnicodeDecodeError is messy, "please crash for me"
return unicode_from_encoded_object(space, w_str, "ascii", "strict")
def descr_next(self, space):
from pypy.objspace.std.unicodeobject import W_UnicodeObject
w_seq = self.w_seq
if w_seq is None:
raise OperationError(space.w_StopIteration, space.w_None)
assert isinstance(w_seq, W_UnicodeObject)
index = self.index
if index == w_seq._length:
self.w_seq = None
raise OperationError(space.w_StopIteration, space.w_None)
start = self.byteindex
end = w_seq.next_codepoint_pos_dont_look_inside(start)
w_res = W_UnicodeObject(w_seq._utf8[start:end], 1)
self.byteindex = end
self.index += 1
return w_res
def newunicode(self, uni):
assert uni is not None
assert isinstance(uni, unicode)
return W_UnicodeObject(uni)
def wrapunicode(space, uni):
from pypy.objspace.std.unicodeobject import W_UnicodeObject
from pypy.objspace.std.ropeunicodeobject import wrapunicode
if space.config.objspace.std.withropeunicode:
return wrapunicode(space, uni)
return W_UnicodeObject(uni)
def newutf8(self, utf8s, length):
assert utf8s is not None
assert isinstance(utf8s, str)
return W_UnicodeObject(utf8s, length)
def wrapunicode(space, uni):
from pypy.objspace.std.unicodeobject import W_UnicodeObject
return W_UnicodeObject(uni)
def _wrap_object(self, space, obj):
result = rffi.cast(self.c_type, obj)
u = rffi.cast(lltype.UniChar, result)
return W_UnicodeObject(u.encode('utf8'), 1)