def descr__new__(space, w_stringtype, w_object=''):
# NB. the default value of w_object is really a *wrapped* empty string:
# there is gateway magic at work
from pypy.objspace.std.stringobject import W_StringObject
w_obj = space.str(w_object)
if space.is_w(w_stringtype, space.w_str):
return w_obj # XXX might be reworked when space.str() typechecks
value = space.str_w(w_obj)
if space.config.objspace.std.withrope:
from pypy.objspace.std.ropeobject import rope, W_RopeObject
w_obj = space.allocate_instance(W_RopeObject, w_stringtype)
W_RopeObject.__init__(w_obj, rope.LiteralStringNode(value))
return w_obj
else:
w_obj = space.allocate_instance(W_StringObject, w_stringtype)
W_StringObject.__init__(w_obj, value)
return w_obj
def descr__new__(space, w_stringtype, w_object=''):
# NB. the default value of w_object is really a *wrapped* empty string:
# there is gateway magic at work
from pypy.objspace.std.stringobject import W_StringObject
w_obj = space.str(w_object)
if space.is_w(w_stringtype, space.w_str):
return w_obj # XXX might be reworked when space.str() typechecks
value = space.str_w(w_obj)
if space.config.objspace.std.withrope:
from pypy.objspace.std.ropeobject import rope, W_RopeObject
w_obj = space.allocate_instance(W_RopeObject, w_stringtype)
W_RopeObject.__init__(w_obj, rope.LiteralStringNode(value))
return w_obj
else:
w_obj = space.allocate_instance(W_StringObject, w_stringtype)
W_StringObject.__init__(w_obj, value)
return w_obj
def encode_unicode(space, w_unistr, encoding, errors):
from pypy.objspace.std.unicodetype import getdefaultencoding, \
_get_encoding_and_errors, encode_object
from pypy.objspace.std.ropeobject import W_RopeObject
if errors is None or errors == "strict":
node = w_unistr._node
if encoding == 'ascii':
result = rope.unicode_encode_ascii(node)
if result is not None:
return W_RopeObject(result)
elif encoding == 'latin-1':
result = rope.unicode_encode_latin1(node)
if result is not None:
return W_RopeObject(result)
elif encoding == "utf-8":
result = rope.unicode_encode_utf8(node)
if result is not None:
return W_RopeObject(result)
return encode_object(space, w_unistr, encoding, errors)
def wrapchar(space, c):
from pypy.objspace.std.stringobject import W_StringObject
from pypy.objspace.std.ropeobject import rope, W_RopeObject
if space.config.objspace.std.withprebuiltchar:
if space.config.objspace.std.withrope:
return W_RopeObject.PREBUILT[ord(c)]
return W_StringObject.PREBUILT[ord(c)]
else:
if space.config.objspace.std.withrope:
return W_RopeObject(rope.LiteralStringNode(c))
return W_StringObject(c)
def wrapstr(space, s):
from pypy.objspace.std.stringobject import W_StringObject
from pypy.objspace.std.ropeobject import rope, W_RopeObject
if space.config.objspace.std.sharesmallstr:
if space.config.objspace.std.withprebuiltchar:
# share characters and empty string
if len(s) <= 1:
if len(s) == 0:
if space.config.objspace.std.withrope:
return W_RopeObject.EMPTY
return W_StringObject.EMPTY
else:
s = s[0] # annotator hint: a single char
return wrapchar(space, s)
else:
# only share the empty string
if len(s) == 0:
if space.config.objspace.std.withrope:
return W_RopeObject.EMPTY
return W_StringObject.EMPTY
if space.config.objspace.std.withrope:
return W_RopeObject(rope.LiteralStringNode(s))
return W_StringObject(s)
def repr__RopeUnicode(space, w_unicode):
hexdigits = "0123456789abcdef"
node = w_unicode._node
size = node.length()
singlequote = doublequote = False
iter = rope.ItemIterator(node)
for i in range(size):
c = iter.nextunichar()
if singlequote and doublequote:
break
if c == u'\'':
singlequote = True
elif c == u'"':
doublequote = True
if singlequote and not doublequote:
quote = '"'
else:
quote = '\''
result = ['u', quote]
iter = rope.ItemIterator(node)
j = 0
while j < size:
code = iter.nextint()
if code >= 0x10000:
result.extend([
'\\',
"U",
hexdigits[(code >> 28) & 0xf],
hexdigits[(code >> 24) & 0xf],
hexdigits[(code >> 20) & 0xf],
hexdigits[(code >> 16) & 0xf],
hexdigits[(code >> 12) & 0xf],
hexdigits[(code >> 8) & 0xf],
hexdigits[(code >> 4) & 0xf],
hexdigits[(code >> 0) & 0xf],
])
j += 1
continue
if code >= 0xD800 and code < 0xDC00:
if j < size - 1:
code2 = iter.nextint()
# XXX this is wrong: if the next if is false,
# code2 is lost
if code2 >= 0xDC00 and code2 <= 0xDFFF:
code = (((code & 0x03FF) << 10) |
(code2 & 0x03FF)) + 0x00010000
result.extend([
'\\',
"U",
hexdigits[(code >> 28) & 0xf],
hexdigits[(code >> 24) & 0xf],
hexdigits[(code >> 20) & 0xf],
hexdigits[(code >> 16) & 0xf],
hexdigits[(code >> 12) & 0xf],
hexdigits[(code >> 8) & 0xf],
hexdigits[(code >> 4) & 0xf],
hexdigits[(code >> 0) & 0xf],
])
j += 2
continue
if code >= 0x100:
result.extend([
'\\',
"u",
hexdigits[(code >> 12) & 0xf],
hexdigits[(code >> 8) & 0xf],
hexdigits[(code >> 4) & 0xf],
hexdigits[(code >> 0) & 0xf],
])
j += 1
continue
if code == ord('\\') or code == ord(quote):
result.append('\\')
result.append(chr(code))
j += 1
continue
if code == ord('\t'):
result.append('\\')
result.append('t')
j += 1
continue
if code == ord('\r'):
result.append('\\')
result.append('r')
j += 1
continue
if code == ord('\n'):
result.append('\\')
result.append('n')
j += 1
continue
if code < ord(' ') or code >= 0x7f:
result.extend([
'\\',
"x",
hexdigits[(code >> 4) & 0xf],
hexdigits[(code >> 0) & 0xf],
])
j += 1
continue
result.append(chr(code))
j += 1
result.append(quote)
return W_RopeObject(rope.rope_from_charlist(result))