SyntaxError: Non-ASCII character '\xc2' in file.py

[NikunjBadjatya]

Hi,
I have a list of lists containing texts to be filled into
corresponding rows and cols of xl sheet using .write function.
ex.data=[
[' 1 ','LTP Execution Procedure',' '],
[' ',' ',' '],
[' ','Prerequisite :',' '],
[' ',' ',' Install LTP on host with root login for testing with
glibc'], #line X
['','','# rpm -ivh arm-sony-linux-dev-ltp-20060822-CE_1.5.arm.rpm'],
[' ',' ',' LTP on host with root login for testing with sslibc
'] #line Y
]

I am passing this list into myfun(), which does the job of writing the
above data[] into the rows and columns. I am using xlwt module.

I am stuck with the following error:
SyntaxError: Non-ASCII character '\xc2' in file.py on line X and line
Y , but no encoding declared.

Any idea what is wrong with my code??
Thanks

[John Machin]

On 8/01/2010 3:43 PM, NikunjBadjatya wrote:
> Hi,

Hello again. Did you get your hyperlinks working crossplatformly?

Exactly what the the error message says.

See
http://docs.python.org/reference/lexical_analysis.html#encoding-declarations
(new in Python 2.5) for documentation and
http://www.python.org/dev/peps/pep-0263/ for background.

In each of line X and line Y, there are two non-ASCII characters. What
appears to be a SPACE ("\x20") after each of "LTP" and "login" is
actually a NO-BREAK SPACE (U+00A0) which is "\xc2\xa0" when encoded in
UTF-8.

Note: I'm viewing the unrendered source of your message in Thunderbird;
the message has

Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

in the headers, and I'm seeing e.g.

Install LTP=A0on host with root login=A0for

The evidence ("Non-ASCII character '\xc2'") is consistent with your file
having been encoded in UTF-8 at the time of the syntax error.

If you really want the NO-BREAK SPACEs in there:

(1) code them explicitly using escape sequences e.g.

text = "foo\xc2\xa0bar\xc2\xa0zot"

or

NBSP = "\xc2\xa0"

text = NBSP.join(["foo", "bar", "zot"])

Then you can put an encoding declaration

# coding: utf-8

at the top of your file as documentation (you won't get a Syntax error
without this, as there will be no non-ASCII bytes in your file before
Python converts the escape sequences).

or
(2) insert an encoding declaration at the top of your file

# coding: utf-8

and leave the NBSPs where they are. Note: NOT RECOMMENDED. You see space
but you've got no-break space. Explicit is better.

However if you didn't expect the no-break spaces (or any other non-ASCII
characters), then edit them to become spaces, and don't put an encoding
declaration at the top of your file. That way if any more nassssties
creep in, you'll get a syntax error, and you can eradicate them.

BTW, a blanket assertion: A Python syntax error in one of your scripts
can not possibly be caused by a problem in any module that your script
might be able to import in the future once your script is repaired
sufficiently to make it runnable.

HTH,
John