Mentre io pensavo ad una intro simpatica "Peter Hansen" scriveva:
> Miki Tebeka wrote:
>>>Hi all, I'm trying to write a multiplatform function that tries to
>>>return the actual user home directory.
>>>...
>>
>> What's wrong with:
>> from user import home
>> which does about what your code does.
>
> :-)
>
> I suspect he simply didn't know about it. I didn't either...
That's true :-D
But as I said in the other post os.environ["HOME"] doesn't work on my
Win2000 box.
> Nemesis, please use the above recipe instead, as it makes
> the more reasonable (IMHO) choice of checking for a HOME
> environment variable before trying the expanduser("~")
> approach. This covers folks like me who, though stuck
> using Windows, despise the ridiculous Microsoft convention
> of "user folders" named like "C:\Documents and Settings\Peter"
> and prefer to create sensible folder like c:\users\peter
> and set a HOME variable to point to it. Your approach
> ignores our HOME variable.
If you look at ntpath.py (I think this is the 'path' module on
Windows 2000) in you Lib dir you'll see that expanduser does try
os.environ["HOME"]
So I'm not ignoring it, maybe it is redundant in my function.
The problem is that expanduser and user.home doesn't test if the value
returned is really a directory.
> c:\>python
> Python 2.4 (#60, Nov 30 2004, 11:49:19) [MSC v.1310 32 bit (Intel)] on win32
> >>> from user import home
> >>> print home
> c:\users\peter
>
> Yay! :-)
on my box it returns "%USERPROFILE%" ;-)
--
Forgive your enemies, but never forget their names.
|\ | |HomePage : http://nem01.altervista.org
| \|emesis |XPN (my nr): http://xpn.altervista.org
--
http://mail.python.org/mailman/listinfo/python-list
