wc might be an empty string

[choldrim]

wcwidth/wcwidth/wcwidth.py

Lines 104 to 182 in c71459e

	def wcwidth(wc):
	r"""
	Given one unicode character, return its printable length on a terminal.
	The wcwidth() function returns 0 if the wc argument has no printable effect
	on a terminal (such as NUL '\0'), -1 if wc is not printable, or has an
	indeterminate effect on the terminal, such as a control character.
	Otherwise, the number of column positions the character occupies on a
	graphic terminal (1 or 2) is returned.
	The following have a column width of -1:
	- C0 control characters (U+001 through U+01F).
	- C1 control characters and DEL (U+07F through U+0A0).
	The following have a column width of 0:
	- Non-spacing and enclosing combining characters (general
	category code Mn or Me in the Unicode database).
	- NULL (U+0000, 0).
	- COMBINING GRAPHEME JOINER (U+034F).
	- ZERO WIDTH SPACE (U+200B) through
	RIGHT-TO-LEFT MARK (U+200F).
	- LINE SEPERATOR (U+2028) and
	PARAGRAPH SEPERATOR (U+2029).
	- LEFT-TO-RIGHT EMBEDDING (U+202A) through
	RIGHT-TO-LEFT OVERRIDE (U+202E).
	- WORD JOINER (U+2060) through
	INVISIBLE SEPARATOR (U+2063).
	The following have a column width of 1:
	- SOFT HYPHEN (U+00AD) has a column width of 1.
	- All remaining characters (including all printable
	ISO 8859-1 and WGL4 characters, Unicode control characters,
	etc.) have a column width of 1.
	The following have a column width of 2:
	- Spacing characters in the East Asian Wide (W) or East Asian
	Full-width (F) category as defined in Unicode Technical
	Report #11 have a column width of 2.
	"""
	# pylint: disable=C0103
	# Invalid argument name "wc"
	ucs = ord(wc)
	# NOTE: created by hand, there isn't anything identifiable other than
	# general Cf category code to identify these, and some characters in Cf
	# category code are of non-zero width.
	# pylint: disable=too-many-boolean-expressions
	# Too many boolean expressions in if statement (7/5)
	if (ucs == 0 or
	ucs == 0x034F or
	0x200B <= ucs <= 0x200F or
	ucs == 0x2028 or
	ucs == 0x2029 or
	0x202A <= ucs <= 0x202E or
	0x2060 <= ucs <= 0x2063):
	return 0
	# C0/C1 control characters
	if ucs < 32 or 0x07F <= ucs < 0x0A0:
	return -1
	# combining characters with zero width
	if _bisearch(ucs, ZERO_WIDTH):
	return 0
	return 1 + _bisearch(ucs, WIDE_EASTASIAN)

if wc is an empty string, an TypeError: ord() expected a character... exception will be raised

if there would be a statement if len(wc) == 0: return 0 before ord(wc), it will be better, I think.

[serhiy-storchaka]

From docstring: "Given one unicode character, ...".

Empty string is not one unicode character. You pass value which does satisfy function contract and get an error. Similar error you will get when pass an unicode string containing more than one character.

[jquast]

I agree and thanks for the report I will work on this as soon as I am able

[wjandrea]

@serhiy-storchaka, The problem is that there should be a specific error about it. For example, it could be as simple as subbing out ord for wcwidth in that error:

if len(wc) != 1:
    msg = 'wcwidth() expected a character, but string of length {} found'.format(len(wc))
    raise TypeError(msg)
ucs = ord(wc)

Example run:

>>> wcwidth('')
Traceback (most recent call last):
  ...
TypeError: wcwidth() expected a character, but string of length 0 found
>>> wcwidth('ab')
Traceback (most recent call last):
  ...
TypeError: wcwidth() expected a character, but string of length 2 found

[jquast]

This is fixed in today's release. On empty string, wcwidth returns 0. Thanks!