[pypy-commit] pypy unicode-utf8: translation fixes

[rlamy]

Author: Ronan Lamy <ronan.lamy at gmail.com>
Branch: unicode-utf8
Changeset: r93321:598f10607a50
Date: 2017-12-09 02:44 +0000
http://bitbucket.org/pypy/pypy/changeset/598f10607a50/

Log:	translation fixes

diff --git a/pypy/module/_pypyjson/interp_decoder.py b/pypy/module/_pypyjson/interp_decoder.py
--- a/pypy/module/_pypyjson/interp_decoder.py
+++ b/pypy/module/_pypyjson/interp_decoder.py
@@ -3,6 +3,7 @@
 from rpython.rlib.objectmodel import specialize, always_inline, r_dict
 from rpython.rlib import rfloat, runicode, rutf8
 from rpython.rtyper.lltypesystem import lltype, rffi
+from rpython.rlib.rarithmetic import r_uint
 from pypy.interpreter.error import oefmt
 from pypy.interpreter import unicodehelper
 
@@ -366,7 +367,7 @@
             return # help the annotator to know that we'll never go beyond
                    # this point
         #
-        utf8_ch = rutf8.unichr_as_utf8(val, allow_surrogates=True)
+        utf8_ch = rutf8.unichr_as_utf8(r_uint(val), allow_surrogates=True)
         builder.append(utf8_ch)
         return i
 
@@ -400,7 +401,7 @@
                 break
             elif ch == '\\' or ch < '\x20':
                 self.pos = i-1
-                return self.space.unicode_w(self.decode_string_escaped(start))
+                return self.decode_string_escaped(start)
             strhash = intmask((1000003 * strhash) ^ ord(ll_chars[i]))
             bits |= ord(ch)
         length = i - start - 1
diff --git a/pypy/module/_rawffi/alt/type_converter.py b/pypy/module/_rawffi/alt/type_converter.py
--- a/pypy/module/_rawffi/alt/type_converter.py
+++ b/pypy/module/_rawffi/alt/type_converter.py
@@ -128,7 +128,7 @@
         intval: lltype.Signed
         """
         self.error(w_ffitype, w_obj)
-        
+
     def handle_unichar(self, w_ffitype, w_obj, intval):
         """
         intval: lltype.Signed
@@ -174,7 +174,7 @@
     def handle_struct_rawffi(self, w_ffitype, w_structinstance):
         """
         This method should be killed as soon as we remove support for _rawffi structures
-        
+
         w_structinstance: W_StructureInstance
         """
         self.error(w_ffitype, w_structinstance)
@@ -228,7 +228,7 @@
             return space.newbytes(chr(ucharval))
         elif w_ffitype.is_unichar():
             wcharval = self.get_unichar(w_ffitype)
-            return space.newutf8(rutf8.unichr_as_utf8(wcharval), 1)
+            return space.newutf8(rutf8.unichr_as_utf8(r_uint(wcharval)), 1)
         elif w_ffitype.is_double():
             return self._float(w_ffitype)
         elif w_ffitype.is_singlefloat():
@@ -349,7 +349,7 @@
     def get_struct_rawffi(self, w_ffitype, w_structdescr):
         """
         This should be killed as soon as we kill support for _rawffi structures
-        
+
         Return type: lltype.Unsigned
         (the address of the structure)
         """
diff --git a/pypy/module/_rawffi/interp_rawffi.py b/pypy/module/_rawffi/interp_rawffi.py
--- a/pypy/module/_rawffi/interp_rawffi.py
+++ b/pypy/module/_rawffi/interp_rawffi.py
@@ -596,9 +596,9 @@
         return space.w_None
     wcharp_addr = rffi.cast(rffi.CWCHARP, address)
     if maxlength == -1:
-        s = rffi.wcharp2utf8(wcharp_addr)
+        s = rffi.wcharp2unicode(wcharp_addr)
     else:
-        s = rffi.wcharpsize2utf8(wcharp_addr, maxlength)
+        s = rffi.wcharp2unicoden(wcharp_addr, maxlength)
     return space.newunicode(s)
 
 @unwrap_spec(address=r_uint, maxlength=int)
diff --git a/pypy/module/array/interp_array.py b/pypy/module/array/interp_array.py
--- a/pypy/module/array/interp_array.py
+++ b/pypy/module/array/interp_array.py
@@ -1,7 +1,7 @@
 from rpython.rlib import jit, rgc, rutf8
 from rpython.rlib.buffer import RawBuffer
 from rpython.rlib.objectmodel import keepalive_until_here
-from rpython.rlib.rarithmetic import ovfcheck, widen
+from rpython.rlib.rarithmetic import ovfcheck, widen, r_uint
 from rpython.rlib.unroll import unrolling_iterable
 from rpython.rtyper.annlowlevel import llstr
 from rpython.rtyper.lltypesystem import lltype, rffi
@@ -1013,7 +1013,7 @@
             elif mytype.typecode == 'c':
                 return space.newbytes(item)
             elif mytype.typecode == 'u':
-                code = ord(item)
+                code = r_uint(ord(item))
                 return space.newutf8(rutf8.unichr_as_utf8(code), 1)
             assert 0, "unreachable"
 
diff --git a/pypy/module/pyexpat/interp_pyexpat.py b/pypy/module/pyexpat/interp_pyexpat.py
--- a/pypy/module/pyexpat/interp_pyexpat.py
+++ b/pypy/module/pyexpat/interp_pyexpat.py
@@ -483,7 +483,7 @@
             except rutf8.CheckError:
                 from pypy.interpreter import unicodehelper
                 # get the correct error msg
-                unicodehelper.str_decode_utf8(s, len(s), 'string', True,
+                unicodehelper.str_decode_utf8(s, 'string', True,
                     unicodehelper.decode_error_handler(space))
                 assert False, "always raises"
         else:
diff --git a/pypy/objspace/std/formatting.py b/pypy/objspace/std/formatting.py
--- a/pypy/objspace/std/formatting.py
+++ b/pypy/objspace/std/formatting.py
@@ -3,7 +3,7 @@
 
 from rpython.rlib import jit, rutf8
 from rpython.rlib.objectmodel import specialize
-from rpython.rlib.rarithmetic import INT_MAX
+from rpython.rlib.rarithmetic import INT_MAX, r_uint
 from rpython.rlib.rfloat import DTSF_ALT, formatd, isnan, isinf
 from rpython.rlib.rstring import StringBuilder
 from rpython.rlib.unroll import unrolling_iterable
@@ -330,7 +330,7 @@
             space = self.space
             if do_unicode:
                 cp = rutf8.codepoint_at_pos(self.fmt, self.fmtpos - 1)
-                w_s = space.newutf8(rutf8.unichr_as_utf8(cp), 1)
+                w_s = space.newutf8(rutf8.unichr_as_utf8(r_uint(cp)), 1)
             else:
                 cp = ord(self.fmt[self.fmtpos - 1])
                 w_s = space.newbytes(chr(cp))
@@ -466,7 +466,7 @@
                 n = space.int_w(w_value)
                 if do_unicode:
                     try:
-                        c = rutf8.unichr_as_utf8(n)
+                        c = rutf8.unichr_as_utf8(r_uint(n))
                     except ValueError:
                         raise oefmt(space.w_OverflowError,
                                     "unicode character code out of range")