This issue tracker has been migrated to GitHub, and is currently read-only.
For more information, see the GitHub FAQs in the Python's Developer Guide.

classification
Title: ctypes.c_char gives a misleading error when passed a one-character unicode string
Type: enhancement Stage:
Components: ctypes Versions: Python 3.3
process
Status: closed Resolution: duplicate
Dependencies: Superseder: Remove unsupported code from ctypes
View: 22161
Assigned To: Nosy List: Steven.Barker, amaury.forgeotdarc, jamadagni, meador.inge, vstinner
Priority: normal Keywords:

Created on 2013-05-16 03:21 by Steven.Barker, last changed 2022-04-11 14:57 by admin. This issue is now closed.

Messages (3)
msg189338 - (view) Author: Steven Barker (Steven.Barker) * Date: 2013-05-16 03:21
While investigating a Stack Overflow question (http://stackoverflow.com/questions/16484764/multiprocessing-value-clear-syntax) I came across a misleading error message from the multiprocessing.Value constructor:

>>> import multiprocessing
>>> my_char = "x"
>>> v = multiprocessing.Value("c", my_char)
Traceback (most recent call last):
  File "<pyshell#21>", line 1, in <module>
    v = multiprocessing.Value("c", my_char)
  File "S:\Python33\lib\multiprocessing\__init__.py", line 243, in Value
    return Value(typecode_or_type, *args, lock=lock)
  File "S:\Python33\lib\multiprocessing\sharedctypes.py", line 70, in Value
    obj = RawValue(typecode_or_type, *args)
  File "S:\Python33\lib\multiprocessing\sharedctypes.py", line 47, in RawValue
    obj.__init__(*args)
TypeError: one character string expected

The "one character string expected" message was rather unhelpful, since that seemed to be what I gave it. After being stumped by this for a bit, I realized that it might be having an issue with Unicode and giving an error message more appropriate to Python 2 than Python 3. Sure enough, passing a one-character bytes instance works just fine.

So, at a minimum I think the error message should be updated to say it wants a one-character bytes instance rather than a "string" (which to my mind means a "str" instance). A further enhancement might be to accept unicode strings that contain just a single ASCII character too, but I realize that may be more messy and error prone.

The error message comes from the ctypes module, and can be reproduced easily:

>>> ctypes.c_char("x")
Traceback (most recent call last):
  File "<pyshell#28>", line 1, in <module>
    ctypes.c_char("x")
TypeError: one character string expected

The exception text comes from Modules/_ctypes/cfield.c:1151
msg190487 - (view) Author: Shriramana Sharma (jamadagni) Date: 2013-06-02 14:10
I came upon this too. In Python 2 it used to expect a one character string. Apparently the same error message has been carried forward to Python 3 too, though now the actual expected input is either a one character bytes type and not a str type, or an int corresponding to the ord() value of that char.

Minimal demonstration:

$ python
Python 2.7.4 (default, Apr 19 2013, 18:28:01) 
[GCC 4.7.3] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> from ctypes import *
>>> class test ( Structure ) :
...     _fields_ = [ ( "ch", c_char ) ]
... 
>>> a = test()
>>> a.ch = ord('a')
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: one character string expected
>>> a.ch = 'c'
>>> a.ch
'c'
>>> 

$ python3
Python 3.3.1 (default, Apr 17 2013, 22:30:32) 
[GCC 4.7.3] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> from ctypes import *
>>> class test ( Structure ) :
...     _fields_ = [ ( "ch", c_char ) ]
... 
>>> a = test()
>>> a.ch = 'c'
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: one character string expected
>>> a.ch = b'c'
>>> a.ch
b'c'
>>> a.ch = ord('c')
>>> a.ch
b'c'
>>>
msg247561 - (view) Author: Steven Barker (Steven.Barker) * Date: 2015-07-29 07:00
I was looking over some of the bugs I've contributed to, and it looks like this one has been fixed. It should be marked as a dupe of issue 22161 and closed (I can close, but not set a superseder, it seems).
History
Date User Action Args
2022-04-11 14:57:45adminsetgithub: 62191
2015-07-29 09:51:43vstinnersetsuperseder: Remove unsupported code from ctypes
2015-07-29 07:00:28Steven.Barkersetstatus: open -> closed
resolution: duplicate
messages: + msg247561
2013-06-05 00:49:18vstinnersetnosy: + vstinner
2013-06-02 14:10:55jamadagnisetnosy: + jamadagni
messages: + msg190487
2013-05-17 19:12:17terry.reedysetnosy: + amaury.forgeotdarc, meador.inge
2013-05-16 03:26:01Steven.Barkersetcomponents: + ctypes, - Library (Lib)
2013-05-16 03:21:57Steven.Barkercreate