Mercurial > cpython-lru_cache-weakref
diff Python/peephole.c @ 69370:4679d0fef389 3.1
Issue #5057: fix a bug in the peepholer that led to non-portable pyc files between narrow and wide builds while optimizing BINARY_SUBSCR on non-BMP chars (e.g. "\U00012345"[0]).
| author | Ezio Melotti <ezio.melotti@gmail.com> |
|---|---|
| date | Fri, 15 Apr 2011 16:38:34 +0300 |
| parents | f197407bdba2 |
| children | 503578ddf286 |
line wrap: on
line diff
--- a/Python/peephole.c Fri Apr 15 18:20:24 2011 +0800 +++ b/Python/peephole.c Fri Apr 15 16:38:34 2011 +0300 @@ -124,6 +124,24 @@ break; case BINARY_SUBSCR: newconst = PyObject_GetItem(v, w); + /* #5057: if v is unicode, there might be differences between + wide and narrow builds in cases like '\U00012345'[0]. + Wide builds will return a non-BMP char, whereas narrow builds + will return a surrogate. In both the cases skip the + optimization in order to produce compatible pycs. + */ + if (newconst != NULL && + PyUnicode_Check(v) && PyUnicode_Check(newconst)) { + Py_UNICODE ch = PyUnicode_AS_UNICODE(newconst)[0]; +#ifdef Py_UNICODE_WIDE + if (ch > 0xFFFF) { +#else + if (ch >= 0xD800 && ch <= 0xDFFF) { +#endif + Py_DECREF(newconst); + return 0; + } + } break; case BINARY_LSHIFT: newconst = PyNumber_Lshift(v, w);
