eval() function in List Comprehension doesn't work.
please see the under codes:

canBusType = 'CANdiag'
result = [eval('canBusType') for i in range(3)]
NameError: name 'canBusType' is not defined

It did work in Python2.5 or 2.6. The expected result is 
['CANdiag', 'CANdiag', 'CANdiag'].

I can't reproduce this. For me it works as expected.

I can't reproduce it either, tested with Py3 (on Linux and Windows) and
with Py2.[456], it worked fine everywhere.

Does your eval() work properly outside listcomps?

Hallo Mr. Fürstenau,

thank you for the quick response.
Have you tried it with Python 3.0

The Error report is truely:
NameError: name 'canBusType' is not defined

Someone else reported the same problem in internet too.

Best regards, mit freundlichen Grüßen,

Jiafei Peng

Softwareentwickler / Embedded System Software (EF-F2)
Software developer / Embedded System Software 

IAV GmbH
Nordhoffstr. 5
38518 Gifhorn
GERMANY

Phone: +49 5371  805-2817
Fax:+49 5371  805-1330

E-mail:  <mailto:Jiafei.Peng@iav.de>
Internet: http://www.iav.de

IAV GmbH
Sitz/Registered Office: Berlin
Registergericht/Registration Court: Amtsgericht Charlottenburg
Registernummer/Company Registration Number: HRB 21 280
Geschäftsführer/Managing Directors: Kurt Blumenröder, Michael Schubert

Hagen Fürstenau <report@bugs.python.org> 
13.02.2009 10:54
Bitte antworten an
Python tracker <report@bugs.python.org>

An
jiafei.peng@iav.de
Kopie

Thema
[issue5242] eval() function in List Comprehension doesn't work 

Hagen Fürstenau <hfuerstenau@gmx.net> added the comment:

I can't reproduce this. For me it works as expected.

----------
nosy: +hagen

_______________________________________
Python tracker <report@bugs.python.org>
<http://bugs.python.org/issue5242>
_______________________________________

Yes
it does work properly outside listcomps.

            canBusType = 'CANdiag'
            result1 = eval('canBusType')
            result2 = [eval('canBusType'), eval('canBusType'), eval(
'canBusType')]
            result3 = [eval('canBusType') for i in range(3)]

result1 = 'CANdiag'
result2 =['CANdiag' 'CANdiag' 'CANdiag']
for result3:
NameError: name 'canBusType' is not defined

Best regards, mit freundlichen Grüßen,

Jiafei Peng

Softwareentwickler / Embedded System Software (EF-F2)
Software developer / Embedded System Software 

IAV GmbH
Nordhoffstr. 5
38518 Gifhorn
GERMANY

Phone: +49 5371  805-2817
Fax:+49 5371  805-1330

E-mail:  <mailto:Jiafei.Peng@iav.de>
Internet: http://www.iav.de

IAV GmbH
Sitz/Registered Office: Berlin
Registergericht/Registration Court: Amtsgericht Charlottenburg
Registernummer/Company Registration Number: HRB 21 280
Geschäftsführer/Managing Directors: Kurt Blumenröder, Michael Schubert

Ezio Melotti <report@bugs.python.org> 
13.02.2009 11:04
Bitte antworten an
Python tracker <report@bugs.python.org>

An
jiafei.peng@iav.de
Kopie

Thema
[issue5242] eval() function in List Comprehension doesn't work 

Ezio Melotti <ezio.melotti@gmail.com> added the comment:

I can't reproduce it either, tested with Py3 (on Linux and Windows) and
with Py2.[456], it worked fine everywhere.

Does your eval() work properly outside listcomps?

----------
nosy: +ezio.melotti

_______________________________________
Python tracker <report@bugs.python.org>
<http://bugs.python.org/issue5242>
_______________________________________

Ezio: this happens inside a function, like this:

def f():
    canBusType = 'CANdiag'
    result = [eval('canBusType') for i in range(3)]

This is expected, and won't easily fix.  The reason is that list
comprehensions in 3.x use a function namespace "under the hood" (in 2.x,
they were implemented like a simple for loop). Because inner functions
need to know what names to get from what enclosing namespace, the names
referenced in eval() can't come from enclosing functions. They must
either be locals or globals.

Of course, the question to the OP is why eval() is needed anyway.

If you know what variable you are going to be eval-ing, or at least, 
have a list of those that might be eval-ed, you can get around this 
issue by making sure they are explicitly referenced in the inner scope 
(i.e., in the list comprehension).  For example, even though list 
comprehensions work in 2.x, generator expressions don't, but this hack 
does (on 2.4 at least):

def f():
  canBusType = 'CANdiag'
  return (eval('canBusType') for i in range(3) if True or canBusType)

By putting a semantically vacuous reference to canBusType (and any 
other variables you want) you make sure they are usable from within the 
eval as well.

eval() is probably already an hack, there's no need to add another hack
to make it work. It's better to just get rid of eval() and find a better
way to do what you want to do.

I agree.