Revision: caf67bedf5af
Branch: default
Author: Robot Framework Developers (robotframew...@gmail.com)
Date: Thu Oct 10 09:50:02 2013 UTC
Log: Support \U escapes with ordinals over 0xFFFF also with narror
builds.
Also fixed \x, \u and \U not to work like \x+1 or \u+123.
Update issue 1524
Small fixes.y
http://code.google.com/p/robotframework/source/detail?r=caf67bedf5af
Modified:
/src/robot/utils/escaping.py
/utest/utils/test_escaping.py
=======================================
--- /src/robot/utils/escaping.py Wed Sep 18 07:58:48 2013 UTC
+++ /src/robot/utils/escaping.py Thu Oct 10 09:50:02 2013 UTC
@@ -80,23 +80,23 @@
def _unescape_character(self, text, length, escape):
try:
- char = self._get_character(text, length)
+ char = self._get_character(text[:length], length)
except ValueError:
return escape + text
else:
return char + text[length:]
def _get_character(self, text, length):
- ordinal = self._get_ordinal(text, length)
- try:
- return unichr(ordinal)
- except (OverflowError, TypeError):
+ if len(text) < length or not text.isalnum():
raise ValueError
-
- def _get_ordinal(self, text, length):
- if len(text) < length:
+ ordinal = int(text, 16)
+ # No Unicode code points above 0x10FFFF
+ if ordinal > 0x10FFFF:
raise ValueError
- return int(text[:length], 16)
+ # unichr only supports ordinals up to 0xFFFF with narrow Python
builds
+ if ordinal > 0xFFFF:
+ return eval("u'\\U%08x'" % ordinal)
+ return unichr(ordinal)
class EscapeFinder(object):
=======================================
--- /utest/utils/test_escaping.py Wed Sep 18 09:01:36 2013 UTC
+++ /utest/utils/test_escaping.py Thu Oct 10 09:50:02 2013 UTC
@@ -80,7 +80,7 @@
assert_unescape(inp, exp)
def test_invalid_x(self):
- for inp in r'\x \xxx xx\xxx \x0 \x0g \X00'.split():
+ for inp in r'\x \xxx xx\xxx \x0 \x0g \X00 \x-1 \x+1'.split():
assert_unescape(inp, inp.replace('\\', ''))
def test_valid_x(self):
@@ -90,7 +90,15 @@
assert_unescape(inp, exp)
def test_invalid_u(self):
- for inp in r'\u \ukekkonen b\uu \u0 \u123 \u123x'.split():
+ for inp in r'''\u
+ \ukekkonen
+ b\uu
+ \u0
+ \u123
+ \u123x
+ \u-123
+ \u+123
+ \u1.23'''.split():
assert_unescape(inp, inp.replace('\\', ''))
def test_valid_u(self):
@@ -100,21 +108,27 @@
assert_unescape(inp, exp)
def test_invalid_U(self):
- for inp in r'\U \Ukekkonen b\Uu \U0 \U1234567 \U1234567x'.split():
+ for inp in r'''\U
+ \Ukekkonen
+ b\Uu
+ \U0
+ \U1234567
+ \U1234567x
+ \U-1234567
+ \U+1234567
+ \U1.234567'''.split():
assert_unescape(inp, inp.replace('\\', ''))
def test_valid_U(self):
- try:
- u00010905 = unichr(int('00010905', 16)) # PHOENICIAN LETTER
WAU
- except ValueError: # occurs on "narrow" Python builds
- u00010905 = 'U00010905'
for inp, exp in [(r'\U00000000', u'\x00'),
(r'\U0000ABba', u'\uabba'),
- (r'\U00010905', u00010905),
+ (r'\U0001f3e9', u'\U0001f3e9'),
+ (r'\U0010FFFF', u'\U0010ffff'),
(r'\U000000e4iti', u'\xe4iti')]:
assert_unescape(inp, exp)
def test_U_above_valid_range(self):
+ assert_unescape(r'\U00110000', 'U00110000')
assert_unescape(r'\U12345678', 'U12345678')
assert_unescape(r'\UffffFFFF', 'UffffFFFF')
--
---
You received this message because you are subscribed to the Google Groups "robotframework-commit" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to robotframework-commit+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
