diff options
| author | Victor Stinner <victor.stinner@gmail.com> | 2013-04-09 20:13:33 (GMT) |
|---|---|---|
| committer | Victor Stinner <victor.stinner@gmail.com> | 2013-04-09 20:13:33 (GMT) |
| commit | 63d5c1a14a644a794bff8d7f9ebc789a85925def (patch) | |
| tree | 44e215686fbdb5c81c8e8fe312b13720e1583046 /Objects/unicodeobject.c | |
| parent | a85af502a4e11d910701f5b7fe16b44c400bf57b (diff) | |
| download | cpython-63d5c1a14a644a794bff8d7f9ebc789a85925def.zip cpython-63d5c1a14a644a794bff8d7f9ebc789a85925def.tar.gz cpython-63d5c1a14a644a794bff8d7f9ebc789a85925def.tar.bz2 | |
Optimize PyUnicode_DecodeCharmap()
Avoid expensive PyUnicode_READ() and PyUnicode_WRITE(), manipulate pointers
instead.
Diffstat (limited to 'Objects/unicodeobject.c')
| -rw-r--r-- | Objects/unicodeobject.c | 16 |
1 files changed, 9 insertions, 7 deletions
diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c index e0b507f..bf49ce5 100644 --- a/Objects/unicodeobject.c +++ b/Objects/unicodeobject.c @@ -7351,27 +7351,29 @@ PyUnicode_DecodeCharmap(const char *s, while (s < e) { if (mapkind == PyUnicode_2BYTE_KIND && maplen >= 256) { enum PyUnicode_Kind outkind = writer.kind; - void *outdata = writer.data; + Py_UCS2 *mapdata_ucs2 = (Py_UCS2 *)mapdata; if (outkind == PyUnicode_1BYTE_KIND) { + Py_UCS1 *outdata = (Py_UCS1 *)writer.data; Py_UCS4 maxchar = writer.maxchar; while (s < e) { - unsigned char ch = *s; - x = PyUnicode_READ(PyUnicode_2BYTE_KIND, mapdata, ch); + ch = *s; + x = mapdata_ucs2[ch]; if (x > maxchar) goto Error; - PyUnicode_WRITE(PyUnicode_1BYTE_KIND, outdata, writer.pos, x); + outdata[writer.pos] = x; writer.pos++; ++s; } break; } else if (outkind == PyUnicode_2BYTE_KIND) { + Py_UCS2 *outdata = (Py_UCS2 *)writer.data; while (s < e) { - unsigned char ch = *s; - x = PyUnicode_READ(PyUnicode_2BYTE_KIND, mapdata, ch); + ch = *s; + x = mapdata_ucs2[ch]; if (x == 0xFFFE) goto Error; - PyUnicode_WRITE(PyUnicode_2BYTE_KIND, outdata, writer.pos, x); + outdata[writer.pos] = x; writer.pos++; ++s; } |