[Python-Dev] Python 3.0: can we make dbm's .keys() return an iterator?

2008-09-11 Thread Gerhard Häring
As far as I can see, the specification of the dbm interface is the 
module docstring in dbm/__init__.py, which reads:


"""
[...]
It has the following interface (key and data are strings):

d[key] = data   # store data at key (may override data at
# existing key)
data = d[key]   # retrieve data at key (raise KeyError if no
# such key)
del d[key]  # delete data stored at key (raises KeyError
# if no such key)
flag = key in d # true if the key exists
list = d.keys() # return a list of all existing keys (slow!)
"""

Now I thought that in Python 3.0, keys(), values() and friends should 
return iterators. Can we change at least the specification of the dbm 
module? We could then later in 3.1 change the implementations to return 
iterators instead of lists, too.


I stumbled upon it cos I'm trying to help Skip with the SQLite-based 
implementation.


-- Gerhard
___
Python-Dev mailing list
[email protected]
http://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
http://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] HTTPS read-only SVN access is denied?

2008-09-11 Thread Olemis Lang
2008/9/7, "Martin v. Löwis" <[EMAIL PROTECTED]>:
> Not necessarily - as you say, it's undocumented (and will remain so);
>  in any case, I have now granted anonymous read access to that
>  repository, through https.
>

Thnx a lot... Formerly  I could not access anything because of the
aforementioned authentication issue... Now public access is ok but...

I have recently tried to check out the latest versions of the PEPs and
I still can't do it. Something like this is what happens...



$ svn co https://svn.python.org/projects/peps/trunk/ .

Apep-0100.txt
Apep-0102.txt

... # Many other files "added"

Apep-0297.txt
Apep-0299.txt
 U   .

Fetching external item into 'docutils'
svn: Can't connect to host 'svn.berlios.de': A connection attempt
failed because the connected party did not properly respond after a
period of time, or established connection failed because connected
host has failed to respond.

--

Please I need some help so as to finally check out all the PEPs at once.

Is this a consequence of my systems settings (SVN, etc...)? Otherwise,
why is this "lovely" message shown? Please... how can I overcome this
situation?

Thnx.

-- 
Regards,

Olemis.
___
Python-Dev mailing list
[email protected]
http://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
http://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] HTTPS read-only SVN access is denied?

2008-09-11 Thread Fredrik Lundh

Olemis Lang wrote:


Fetching external item into 'docutils'
svn: Can't connect to host 'svn.berlios.de': A connection attempt
failed because the connected party did not properly respond after a
period of time, or established connection failed because connected
host has failed to respond.


> Please I need some help so as to finally check out all the PEPs at
> once.

I suspect you already have all the PEP:s; the checkout command gets 
stuck when trying to fetch some additional software from an external

server.

try adding the --ignore-externals option to the checkout command, and 
see if this gets you any further.


then check if you can reach http://svn.berlios.de via your browser, or 
if some firewall rule gets in the way.




___
Python-Dev mailing list
[email protected]
http://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
http://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] HTTPS read-only SVN access is denied?

2008-09-11 Thread Martin v. Löwis

> then check if you can reach http://svn.berlios.de via your browser, or
> if some firewall rule gets in the way.

He probably can, but the firewall still gets in the way when he tries
to do the svn checkout - his firewall is incapable of forwarding OPTIONS
and other methods used by subversion. Hence he needs to use https, as
the firewall then puts through all traffic unmodified, thanks to the
CONNECT method.

So there will be no chance that he can checkout the berlios externals
directly, as their URL is defined in the Python repository (and will
remain at http, for the sake of most of us with sane networking
environments)

Regards,
Martin
___
Python-Dev mailing list
[email protected]
http://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
http://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com


Re: [Python-Dev] Python 3.0: can we make dbm's .keys() return an iterator?

2008-09-11 Thread Josiah Carlson
On Thu, Sep 11, 2008 at 2:40 AM, Gerhard Häring <[EMAIL PROTECTED]> wrote:
> As far as I can see, the specification of the dbm interface is the module
> docstring in dbm/__init__.py, which reads:
>
> """
> [...]
> It has the following interface (key and data are strings):
>
>d[key] = data   # store data at key (may override data at
># existing key)
>data = d[key]   # retrieve data at key (raise KeyError if no
># such key)
>del d[key]  # delete data stored at key (raises KeyError
># if no such key)
>flag = key in d # true if the key exists
>list = d.keys() # return a list of all existing keys (slow!)
> """
>
> Now I thought that in Python 3.0, keys(), values() and friends should return
> iterators. Can we change at least the specification of the dbm module? We
> could then later in 3.1 change the implementations to return iterators
> instead of lists, too.
>
> I stumbled upon it cos I'm trying to help Skip with the SQLite-based
> implementation.

If it's to be following a variant of the dictionary API, the object
should return a view object:
http://docs.python.org/dev/3.0/library/stdtypes.html#dict .

I've updated my version of the library in the tracker, which includes
generic Keys, Values, and Items classes that implement the view
interface.

 - Josiah
___
Python-Dev mailing list
[email protected]
http://mail.python.org/mailman/listinfo/python-dev
Unsubscribe: 
http://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com