Hi Jean.
One general comment: it's more useful to the (commandline) caller to
return an appropriate errno when things go wrong. This will help in
testing, and could also be useful if the DC is called from a script.
Python doesn't do this when it raises an exception that isn't caught, as
shown in the following example:
strongheart:/usr/share/distro_const> python
>>> import os
>>> try:
... os.chmod("/tmp/abcdefg", 0777);
... except Exception, e:
... raise SystemExit(e)
...
[Errno 2] No such file or directory: '/tmp/abcdefg'
strongheart:/usr/share/distro_const> echo $?
1
But in the main, you can retrieve the errno from an exception and call
sys.exit with it, like this:
import os
import traceback
import sys
try:
... os.chmod("/tmp/abcdefg", 0777);
... except Exception, e:
... traceback.print_exc()
... sys.exit(e.args[0])
...
Traceback (most recent call last):
File "<stdin>", line 2, in ?
OSError: [Errno 2] No such file or directory: '/tmp/abcdefg'
strongheart:/usr/share/distro_const> echo $?
2
So you see here that 2 (errno) is returned instead of 1.
Thanks,
Jack
On 12/16/08 12:28, Jean McCormack wrote:
> Karen Tung wrote:
>
>> Jean McCormack wrote:
>>
>>> Please review the following webrev for defects 4352 & 4354
>>>
>>> Defects:
>>> http://defect.opensolaris.org/bz/show_bug.cgi?id=4352
>>> http://defect.opensolaris.org/bz/show_bug.cgi?id=4354
>>>
>>> Webrev:
>>> http://cr.opensolaris.org/~jeanm/slim_4352_4354/
>>>
>>> Jean
>>> _______________________________________________
>>> caiman-discuss mailing list
>>> caiman-discuss at opensolaris.org
>>> http://mail.opensolaris.org/mailman/listinfo/caiman-discuss
>>>
>>>
>> Hi Jean,
>>
>> Here are my comments:
>>
>> distro_const.py, line 51:
>> - What's changed in this line? I don't see anything, but it indicated
>> that it's changed.
>>
>>
> manifest <underscore> file was changed to manifest<dash>file. This now
> agrees with the
> rest of the usage statement.
>
> distro_const.py, line 574:
>
>> - While the code works the way it is now,
>> I am concern about the possibility of information lost by just
>> returning 2 for
>> all the SystemExit. It's true that right now, the usage() is the only
>> thing that returns 2.
>> What about in the future? What about other part of the DC code that
>> might
>> call SystemExit() with an error message that we want to see?
>> IMO, it is more correct to define an exception specific to usage
>> related errors,
>> and catch that exception, assign the appropriate return code, and
>> leave SystemExit()
>> the way it is.
>>
> That sounds good.
>
> Jean
>
>> Thanks,
>>
>> --Karen
>>
>
> _______________________________________________
> caiman-discuss mailing list
> caiman-discuss at opensolaris.org
> http://mail.opensolaris.org/mailman/listinfo/caiman-discuss
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://mail.opensolaris.org/pipermail/caiman-discuss/attachments/20081216/4c7532cc/attachment.html>
