On 4/15/06, Talin <[EMAIL PROTECTED]> wrote:
> Another grab-bag of language ideas / wishes.
In the future, please use one idea/wish per email and use a subject
line indicative of the subject (so *don't* use "Another wish" :-)
--
--Guido van Rossum (home page: http://www.python.org/~guido/)
__
On 4/16/06, Aahz <[EMAIL PROTECTED]> wrote:
> On Sun, Apr 16, 2006, Greg Ewing wrote:
> > Jim Jewett wrote:
> >>
> >> (I wouldn't want to give up slicing, though, which might make the
> >> no-iteration trickier.)
> >
> > You'd want to allow slicing but not indexing -- i.e.
> > s[i:j] is okay but no
>> Because then the re module wouldn't be thread safe.
talin> True, although I suppose that you could add a "last match" to
talin> regular expression objects:
talin> re_pattern = re.compile( ... )
talin> if re_pattern.match( string ):
talin>text = re_pattern.lastmatc
On 4/16/06, Guido van Rossum <[EMAIL PROTECTED]> wrote:
> I think you may be right. I implemented this (it was really simple to
> do) but then found I had to fix tons of places that iterate over
> strings.
I'm abandoning this project; the patch is SF patch 1471291. I'm no
longer in favor of this i
Talin <[EMAIL PROTECTED]> wrote:
>
> Talin acm.org> writes:
>
> > 2) A suggestion which I've seen others bring up before is the use of
> > the * operator for tuple packing / unpacking operations, i.e.:
> >
> > a, *b = (1, 2, 3)
>
> I wanted to add another case that I run across in my code
[EMAIL PROTECTED] wrote:
> Giovanni> I'm sure you were talking more generically, but as for
> this Giovanni> specific case (which is indeed very common), I
> always wondered Giovanni> why re couldn't export wrappers around
> bound methods of the Giovanni> last match object, so to a
Giovanni> Can't you put the global last-match object into TLS?
What if you're not using threads?
Skip
___
Python-3000 mailing list
[email protected]
http://mail.python.org/mailman/listinfo/python-3000
Unsubscribe:
http://mail.python.org/mailm
[EMAIL PROTECTED] wrote:
> Giovanni> Can't you put the global last-match object into TLS?
>
> What if you're not using threads?
Hey, that was *my* point.
For all the Python programs which don't use multi-threading, having a global
match object is a very important shortcut. It allows two main
On 4/16/06, Giovanni Bajo <[EMAIL PROTECTED]> wrote:
> For all the Python programs which don't use multi-threading, having a global
> match object is a very important shortcut. It allows two main things: having
> multiple if/elif regexp chains which do not add an indent level each time, and
> have
On 4/15/06, Talin <[EMAIL PROTECTED]> wrote:
> Another grab-bag of language ideas / wishes. Some of these are
> items that have been raised before on other lists - I want to get them
> written down so that they can be rejected quickly :)
>
> 1) I don't know if this is already on the table, but I su
On 4/14/06, Kendall Clark <[EMAIL PROTECTED]> wrote:
> Fair enough. My primary use case is using Python as a host for a
> Domain Specific Language. [...]
Then I suggest that having a good syntax is important; you don't want
your DSL to look ugly because that would not contribute to having fun.
Le
On 4/16/06, Greg Ewing <[EMAIL PROTECTED]> wrote:
> Ian Bicking wrote:
>
> > Class.some_method(self, blah) seems like a corner case. How often do
> > you do that?
>
> Very frequently, when calling inherited __init__ methods.
>
> > If it is calling a superclass method, then super(Class,
> > self).s
On 4/15/06, Ian D. Bollinger <[EMAIL PROTECTED]> wrote:
> Personally I like the idea (stolen from Perl, and others) of allowing
> underscores in numbers as to group digits, for instance:
> one_million = 1_000_000
> maximum = 0xFF_FF_FF_FF
>
> Once four or more identical digits appear in a row, I fi
If someone really wants AST access in Python 3000, they should design
and implement an API and propose it. I see it as a borderline feature
from the POV of my own agenda for Python 3000, but not as something
I'd likely reject if it were to be offered on a silver platter;
especially if there was a c
Some of the coding projects I've recently been proposing as challenges
to certain wild feature proposals might make good topics for the
Google Summer of Code (see http://code.google.com/soc/ ).
We'd need someone to volunteer as a mentor, and a student to do the
work over the summer; the student wo
On Apr 16, 2006, at 4:06 PM, Guido van Rossum wrote:
> On 4/14/06, Kendall Clark <[EMAIL PROTECTED]> wrote:
>> Fair enough. My primary use case is using Python as a host for a
>> Domain Specific Language. [...]
>
> Then I suggest that having a good syntax is important; you don't want
> your DSL t
On 4/16/06, Kendall Clark <[EMAIL PROTECTED]> wrote:
> A perfectly reasonable suggestion. I like this better than writing a
> PEP. Alas, that doesn't mean I have time to do it, which sucks. I
> guess I shouldn't have proposed a feature for Py3K w/out the time to
> back it up with code.
Perhaps you
I'm not against a better reload() if a reasonable set of heuristics
can be defined (e.g. many people have asked for class objects to be
*edited* instead of replaced; and of course there's the dependencies
graph problem). I'm even OK with changing the definition of the
language (slightly) to make e.
On 4/16/06, Guido van Rossum <[EMAIL PROTECTED]> wrote:
On 4/15/06, Talin <[EMAIL PROTECTED]> wrote:> Another grab-bag of language ideas / wishes. Some of these are> items that have been raised before on other lists - I want to get them
> written down so that they can be rejected quickly :)>> 1) I
Giovanni> Can't you put the global last-match object into TLS?
>> What if you're not using threads?
Giovanni> Hey, that was *my* point.
Giovanni> For all the Python programs which don't use multi-threading,
Giovanni> having a global match object is a very important shortcut.
Guido van Rossum python.org> writes:
> On 4/15/06, Talin acm.org> wrote:
> > Another grab-bag of language ideas / wishes. Some of these are
> > items that have been raised before on other lists - I want to get them
> > written down so that they can be rejected quickly :)
> >
> > 1) I don't know
pobox.com> writes:
> This really isn't Python 3000 material, folks. I thought my first response
> would trigger an "Oh yeah, I hadn't thought of that", not continual
> ill-conceived "yeah, but you can do this instead when the situation is
> right". Please take it c.l.py, hash it out there, and
On 4/16/06, Guido van Rossum <[EMAIL PROTECTED]> wrote:
> Some of the coding projects I've recently been proposing as challenges
> to certain wild feature proposals might make good topics for the
> Google Summer of Code (see http://code.google.com/soc/ ).
>
> (I can't volunteer to be a mentor mysel
On 4/16/06, Neal Norwitz <[EMAIL PROTECTED]> wrote:
> On 4/16/06, Guido van Rossum <[EMAIL PROTECTED]> wrote:
> > Some of the coding projects I've recently been proposing as challenges
> > to certain wild feature proposals might make good topics for the
> > Google Summer of Code (see http://code.go
On 4/16/06, Brett Cannon <[EMAIL PROTECTED]> wrote:
>
> again, if my host at Google wants me to spend my working hours being a
> good mentor I could possibly be convinced to make sure I get assigned
> one student (two turned out to be too much). =)
I heard your host is a slave driver and an assho
"Talin" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> Unfortunately, if the paragraph only contains a single word,
> this blows up. So what you end up having to do is:
>
> parts = paragraph.split( ' ', 1 )
> if len( parts ) > 1:
> first, rest = parts
> else:
> firs
On 4/16/06, Neal Norwitz <[EMAIL PROTECTED]> wrote:
> On 4/16/06, Brett Cannon <[EMAIL PROTECTED]> wrote:
> >
> > again, if my host at Google wants me to spend my working hours being a
> > good mentor I could possibly be convinced to make sure I get assigned
> > one student (two turned out to be to
On Apr 16, 2006, at 2:34 PM, Guido van Rossum wrote:
> Some of the coding projects I've recently been proposing as challenges
> to certain wild feature proposals might make good topics for the
> Google Summer of Code (see http://code.google.com/soc/ ).
>
> We'd need someone to volunteer as a ment
Guido van Rossum wrote:
> (I'm surprised by this. Why would s[i] be wrong?)
Because if we're to be consistent about the notion that
a string isn't a sequence of characters, s[i] is really
a slice of length one, and should be expressed that
way.
Also, if len(s) and s[i] are both legal, it becomes
On Mon, Apr 17, 2006, Greg Ewing wrote:
> Guido van Rossum wrote:
>>
>> I think you may be right. I implemented this (it was really simple
>> to do) but then found I had to fix tons of places that iterate over
>> strings.
>
> I wonder if the stdlib might be a bit unusual here, since it's the
> pla
Guido van Rossum wrote:
> On 4/16/06, Greg Ewing <[EMAIL PROTECTED]> wrote:
> > In the case of __init__
> > methods, most of the time it simply doesn't work,
> > because different __init__ methods rarely have the
> > same signature.
>
> How does that prevent you from using super()? (Hint: it does
On 16/04/06 Aahz said:
> Anything that would make this solution more difficult to explain would
> make my life more difficult. I know there are many better and more
> efficient mechanisms (particularly if one is not restricted to Python
> 2.2), but it seems to me that we cannot afford to let Pyt
On 17/04/06 Greg Ewing said:
> The other possible reason for using super() is so
> you don't have to write the name of the base class
> into all your inherited method calls. But that's a
> separate issue that would be better addressed by a
> different mechanism, rather than conflating the
> two in
On Sun, Apr 16, 2006, Michael P. Soulier wrote:
> On 17/04/06 Greg Ewing said:
>>
>> The other possible reason for using super() is so you don't have
>> to write the name of the base class into all your inherited method
>> calls. But that's a separate issue that would be better addressed by
>> a d
[Guido]
>> ...
>> I implemented this (it was really simple to do) but then found I had
>> to fix tons of places that iterate over strings.
[Greg Ewing]
> I wonder if the stdlib might be a bit unusual here,
> since it's the place where low-level things are
> implemented. I don't think there would b
On Sun, Apr 16, 2006, Tim Peters wrote:
>
> Now that I think of it, though, I've been burned perhaps twice in my
> total Python life by recursing on a string when I didn't intend to.
> Apart from flatten()-ish functions, I'm not sure I've written anything
> vulnerable to that.
I've been burned a
Talin wrote:
> Guido van Rossum python.org> writes:
>
>> On 4/15/06, Talin acm.org> wrote:
>>> Another grab-bag of language ideas / wishes. Some of these are
>>> items that have been raised before on other lists - I want to get them
>>> written down so that they can be rejected quickly :)
>>>
>>
Aahz wrote:
> On Sun, Apr 16, 2006, Tim Peters wrote:
>> Now that I think of it, though, I've been burned perhaps twice in my
>> total Python life by recursing on a string when I didn't intend to.
>> Apart from flatten()-ish functions, I'm not sure I've written anything
>> vulnerable to that.
>
>
> >> On 4/15/06, Talin acm.org> wrote:
> >>> 1) I don't know if this is already on the table, but I sure would like
> >>> to be able to have more options as far as mixing positional and
> >>> keyword arguments.
> > Guido van Rossum python.org> writes:
> >> I agree. Please produce a patch that im
On 4/17/06, Neal Norwitz <[EMAIL PROTECTED]> wrote:
> A, I was planning to rope you in as mentor. :-) I plan to mentor,
> I'm not sure how many projects I can handle though.
Don't plan on any for yourself; you're too busy already with the 2.5
release and with managing the mentors.
> I hope a
On 4/16/06, Talin <[EMAIL PROTECTED]> wrote:
> I'm really more interested in the general case where you have a test, and
> then you have the results of the test. It would be nice to put the conditional
> statement first, and then once you know that the test succeeded, start to
> look at the specifi
Nick Coghlan gmail.com> writes:
> A question for Guido: Is this flexibility for argument passing only, or does
> it apply to parameter declarations as well?
Actually, parameter declarations are what I am talking about. Now go back
and re-read the whole message with that in mind. :)
(Its probab
42 matches
Mail list logo
