Message67651
Review:
* Why is an empty string not printable? In any case, the empty string
should be among the test cases for isprintable().
* Why not use PyUnicode_DecodeASCII instead of
PyUnicode_FromEncodedObject? It should be a bit faster.
* If old-style string formatting gets "%a", .format() must get a "!a"
specifier.
* The ascii() and repr() tests should be expanded so that both test the
same set of objects, and the expected differences. Are there tests for
failing cases?
* This is just "return ascii" (in builtin_ascii):
+ if (ascii == NULL)
+ return NULL;
+
+ return ascii;
* For PyBool_FromLong(1) and PyBool_FromLong(0) there is Py_RETURN_TRUE
and Py_RETURN_FALSE. (You're not to blame, the rest of unicodeobject.c
seems to use them too, probably a legacy.)
* There appear to be some space indentations in tab-indented files like
bltinmodule.c and vice versa (unicodeobject.c).
* C docs/isprintable() docs: The spec
+ Characters defined in the Unicode character database as "Other"
+ or "Separator" other than ASCII space(0x20) are not considered
+ printable.
is unclear, better say "All character except those ... are considered
printable".
* ascii() docs:
+ the non-ASCII
+ characters in the string returned by :func:`ascii`() are hex-escaped
+ to generate a same string as :func:`repr` in Python 2.
should be
"the non-ASCII characters in the string returned by :func:`repr` are
backslash-escaped (with ``\x``, ``\u`` or ``\U``) to generate ...".
* makeunicodedata: len(list(n for n in names if n is not None)) could
better be expressed as sum(1 for n in names if n is not None).
Otherwise, the patch is fine IMO. (I'm surprised that only so few tests
needed adaptation, that's a sign that we're not testing Unicode enough.) |
|
Date |
User |
Action |
Args |
2008-06-03 10:13:57 | georg.brandl | set | spambayes_score: 0.00286902 -> 0.0028690177 recipients:
+ georg.brandl, lemburg, gvanrossum, ishimoto, amaury.forgeotdarc |
2008-06-03 10:13:56 | georg.brandl | set | spambayes_score: 0.00286902 -> 0.00286902 messageid: <1212488036.32.0.623029581987.issue2630@psf.upfronthosting.co.za> |
2008-06-03 10:13:53 | georg.brandl | link | issue2630 messages |
2008-06-03 10:13:50 | georg.brandl | create | |
|