Am I missing something with Python not having interfaces?

[Bruno Desthuilliers]

Re: Am I missing something with Python not having interfaces?

jmDesktop a écrit :
(snip)

On the dynamic typing,
isn't that the same sort of thing that lots of scripting languages
do?

This is not restricted to scripting languages - that is, unless you'd
call Smalltalk, Erlang or common lisp "scripting languages".

Also and FWIW, static typing in C is mostly here to help the compiler
optimizing, and it's very common to use void pointers and typecast to
get some genericity (cf the sort() function in C stdlib). And you have
the same pattern in Java with "generic" containers storing only "object"
instances, then you have to cast your objects back to the appropriate
type (which may fails, leading to a runtime error).

VBScript doesn't require you to define your variables,

You mean "to declare the type of your variables" ? Neither OCaml nor
Haskell require this, and they are both statically and strongly typed.

but I
don't really want to use it for anything

Neither do I, but not because of dynamic typing. By experience, dynamic
typing works just fine for most applications. And by experience too,
static typing doesn't magically make your code bullet-proof.

(used to use it a lot in
Classic ASP.) I believe everyone that Python is great, but some of it
doesn't make sense to me as to why.

Quite a lot of Java stopped making sens to me since I learned some other
languages. And most of the "OO" (hem) complexity, patterns madness etc
you'll typically find in Java code only exists because of the language's
arbitrary restrictions and lack of expressivity.

Learn Python, learn Ruby or (preferably) Smalltalk, learn common lisp or
(preferably) Scheme, learn OCaml or (preferably) Haskell, learn Erlang,
and you'll find out that there's much more in programming and languages
than can be dreamt of in Java's philosophy !-)

[Bjoern Schliessmann]

Re: Am I missing something with Python not having interfaces?

jmDesktop wrote:

Studying OOP and noticed that Python does not have Interfaces.

By "OOP", you mean "Java", right? 8)

Is that correct? Is my schooling for nought on these OOP concepts
if I use Python. Am I losing something if I don't use
the "typical" oop constructs found in other languages (Java, C#
come to mind.)

AFAIK, Interfaces weren't "typical" before Java. CMIIW.

BTW, Zope has interfaces. Not sure if it's exactly the same.

Regards,


Björn

--
BOFH excuse #424:

operation failed because: there is no message for this error (#1014)

[bruno.desthuilliers@gmail.com]

Re: Am I missing something with Python not having interfaces?

On 7 mai, 18:05, Bjoern Schliessmann <usenet-
mail-0306.20.chr0n.. .@spamgourmet.c omwrote:

jmDesktop wrote:

Studying OOP and noticed that Python does not have Interfaces.

>
By "OOP", you mean "Java", right? 8)
>

Is that correct? Is my schooling for nought on these OOP concepts
if I use Python. Am I losing something if I don't use
the "typical" oop constructs found in other languages (Java, C#
come to mind.)

>
AFAIK, Interfaces weren't "typical" before Java. CMIIW.
>
BTW, Zope has interfaces. Not sure if it's exactly the same.
>

It isn't.

[Daniel Marcel Eichler]

Re: Am I missing something with Python not having interfaces?

Am Dienstag 06 Mai 2008 16:07:01 schrieb Mike Driscoll:

If so, then it looks like an Interface is a generic class with method
stubs. You can do that with Python just as easily by creating empty
methods with just the "pass" keyword.

Well, no. It's a litte different. Interfaces force to implement methods.
Simple inheritance with method-stubs can't garantee that a specific
method is reimplementat in a sub-class. Interfaces work at
compile-time, while method-stubs raise at their first call, so in worst
case, never (when the developer is available).

Since it's just a construct to implement polymorphism, I don't think
you'll lose anything. However, Python does not require you to re-
implement every method of the class it is inheriting from.

That's the point. Interfaces garantee that a duck is a duck, an not only
a chicken that quack.

You can just override those that you want and leave the others alone.

Not always desirable. But often enough ;)

[Luis Zarrabeitia]

Re: Am I missing something with Python not having interfaces?

On Wednesday 07 May 2008 03:19:30 pm Daniel Marcel Eichler wrote:

Am Dienstag 06 Mai 2008 16:07:01 schrieb Mike Driscoll:
>
Well, no. It's a litte different. Interfaces force to implement methods.
Simple inheritance with method-stubs can't garantee that a specific
method is reimplementat in a sub-class. Interfaces work at
compile-time, while method-stubs raise at their first call, so in worst
case, never (when the developer is available).

<Pseudo-java code>
class MyClass implements MyInterfaceWith PassMethod {
function Pass() {
}
}
</pseudo-java code>

There you have it, interfaces are not enough to ensure that the implementors
actually implement the methods. They are useful for warning at compile time
if there is a missing method, but nothing more. I believe you could achievea
very similar warning in python using some kind of Interface metaclass (too
lazy right now to hack a proof of concept, but it looks doable).

I actually like the Interface concept... as portrayed by zope.Interfaces and
pyprotocols (not the same thing, but I like them), and the ABCs (these, not
so much). But Java interfaces are barely a replacement for multiple
inheritance+abs tract methods.

That's the point. Interfaces garantee that a duck is a duck, an not only
a chicken that quack.

Not really. (See the NotImplementedE xceptions in .NET 1.1 and window forms,
and/or their TabbedControl/TabPage background color attribute. Can't find
them now, as I don't have .NET anymore, but I consider that an example of why
the static typing are... overrated)

--
Luis Zarrabeitia (aka Kyrie)
Fac. de Matemática y Computación, UH.

http://profesores.matcom.uh.cu/~kyrie

[bruno.desthuilliers@gmail.com]

Re: Am I missing something with Python not having interfaces?

On 7 mai, 21:19, Daniel Marcel Eichler <onsen-n...@gmx.netwro te:

Am Dienstag 06 Mai 2008 16:07:01 schrieb Mike Driscoll:
>

If so, then it looks like an Interface is a generic class with method
stubs. You can do that with Python just as easily by creating empty
methods with just the "pass" keyword.

>
Well, no. It's a litte different. Interfaces force to implement methods.
Simple inheritance with method-stubs can't garantee that a specific
method is reimplementat in a sub-class.

And Java's "Interface" wart can't garantee that a specific method is
*appropriately* implemented. One often sees do-nothing methods in Java
code - useless boilerplate code that's only here to satisfy the
compiler. It's just like Java's "checked exception" mechanism : one
very often sees do-nothing catch-all try/catch blocks in Java - which
is way worse than just letting the exception propagate. I find all
this totally pointless, because there's just no way for a compiler to
check if your code is logically correct.

Interfaces work at
compile-time, while method-stubs raise at their first call, so in worst
case, never.

And then ? If it's never called, why bother implementing it ?

Here again, I don't see the point of compile-time checking.
RuntimeError is a well-known Java exception, so the fact is that Java
forces you into either rigid designs or overly complex workarounds
(best knowns as "design patterns"), without even being able to
garantee anything wrt/ correctness nor even safety. Not that static
typing by itself is necessarily bad - as usual, there's a balance
between somewhat conflicting goals -, but Java's type system
definitivly is (bad).

(snip)

That's the point. Interfaces garantee that a duck is a duck, an not only
a chicken that quack.

Who cares if it's a chicken as long as it quacks when you ask her to ?
Now *This* is the whole point of chicken^Mduck typing, isn't it ?-)

[Daniel Marcel Eichler]

Re: Am I missing something with Python not having interfaces?

Am Mittwoch 07 Mai 2008 22:39:30 schrieb Luis Zarrabeitia:

There you have it, interfaces are not enough to ensure that the
implementors actually implement the methods. They are useful for
warning at compile time if there is a missing method, but nothing
more.

It's not the fault of the enviroment, if the coder is to stupid to use
it the right way.

I believe you could achieve a very similar warning in python
using some kind of Interface metaclass (too lazy right now to hack a
proof of concept, but it looks doable).

Of course. And unlike Javas interfaces, it ensure a more dynamic
coding-style, without great tools which check all the time for correct
behaviour.

[Daniel Marcel Eichler]

Re: Am I missing something with Python not having interfaces?

Am Donnerstag 08 Mai 2008 00:12:26 schrieb
bruno.desthuill iers@gmail.com:

very often sees do-nothing catch-all try/catch blocks in Java - which
is way worse than just letting the exception propagate. I find all
this totally pointless, because there's just no way for a compiler to
check if your code is logically correct.

But it's enough if the called method exists and returns the correct
type. At least it prevents a crash.

Interfaces work at
compile-time, while method-stubs raise at their first call, so in
worst case, never.

And then ? If it's never called, why bother implementing it ?

You never can't say when it's called at least, that's the point.

That's the point. Interfaces garantee that a duck is a duck, an not
only a chicken that quack.

Who cares if it's a chicken as long as it quacks when you ask her to
? Now *This* is the whole point of chicken^Mduck typing, isn't it ?-)

Ducks can also swim and fly. And if you need a really duck, but have
onyl a chicken while the coder was to bored to make one...

Of course, in the practical world that all doesn't matter. But in the
theoretical world of the big coding farms, called business, that's one
cornerstone of success, in the tinking of managers and so.

[J. Clifford Dyer]

Re: Am I missing something with Python not having interfaces?

On Thu, 2008-05-08 at 09:12 +0200, Daniel Marcel Eichler wrote:

Am Mittwoch 07 Mai 2008 21:48:56 schrieben Sie:
>

That's the point. Interfaces garantee that a duck is a duck, an not
only a chicken that quack.

Which, in spite of your rhetorical flourish, is the exact opposite of
duck typing. Duck typing means that if you're looking for quacking
behavior, and you find a quacking chicken, it's close enough.

>
I didn't said that interfaces are a kind of duck-typing. In fact it was
the exact opposite.
>

Sometimes you need that kind of rigor, and you can get it as easily
as

>
And sometimes you need more. So what?
>

More rigor than Zope's interfaces offer? That's new information to me.
Perhaps you should stop being argumentative for a moment, and explain
exactly what it is you're looking for and why Zope interfaces don't fit
the bill.

[J. Cliff Dyer]

Re: Am I missing something with Python not having interfaces?

On Thu, 2008-05-08 at 13:25 -0400, J. Cliff Dyer wrote:

On Thu, 2008-05-08 at 19:11 +0200, Daniel Marcel Eichler wrote:

Am Donnerstag 08 Mai 2008 13:02:52 schrieb J. Clifford Dyer:

I didn't said that interfaces are a kind of duck-typing. In fact it
was the exact opposite.

Sometimes you need that kind of rigor, and you can get it as
easily as

And sometimes you need more. So what?
>
More rigor than Zope's interfaces offer?

Did i say that? I only explained why interfaces are not 'simple method
stubs'.

>
Um. Yes. I said "sometimes you need that kind of rigor," and proceeded
to explain how zope interfaces offer it. (which you snipped from your
quotation), and you said "sometimes you need more." If you were
actually responding to what I said, then "sometimes you need more" must
mean more than zope interfaces.
>

That's new information to
me. Perhaps you should stop being argumentative for a moment, and
explain exactly what it is you're looking for and why Zope interfaces
don't fit the bill.

Perhaps you should read more accurate. It's not my thread and i'm not
looking for anything here.

>
Then what were you trying to say?
>

--
Oook,
J. Cliff Dyer
Carolina Digital Library and Archives
UNC Chapel Hill

[Bruno Desthuilliers]

Re: Am I missing something with Python not having interfaces?

Daniel Marcel Eichler a écrit :

Am Donnerstag 08 Mai 2008 00:12:26 schrieb
bruno.desthuill iers@gmail.com:
>

>very often sees do-nothing catch-all try/catch blocks in Java - which
>is way worse than just letting the exception propagate. I find all
>this totally pointless, because there's just no way for a compiler to
>check if your code is logically correct.

>
But it's enough if the called method exists and returns the correct
type. At least it prevents a crash.

Then providing an appropriate default in the base class is enough too.

>>Interfaces work at
>>compile-time, while method-stubs raise at their first call, so in
>>worst case, never.

>And then ? If it's never called, why bother implementing it ?

>
You never can't say when it's called at least, that's the point.

Either the method is actually called somewhere and you'll now it pretty
soon, or it isn't and you don't care.

>>That's the point. Interfaces garantee that a duck is a duck, an not
>>only a chicken that quack.

>Who cares if it's a chicken as long as it quacks when you ask her to
>? Now *This* is the whole point of chicken^Mduck typing, isn't it ?-)

>
Ducks can also swim and fly. And if you need a really duck,

If you're code expects something that quacks, swims and flies, anything
that quacks, swims and flies is ok. You just don't care if it's a duck
or a flying whale with a quacking device tied to it.

but have
onyl a chicken while the coder was to bored to make one...

Then the coder have a nice traceback, and it's not *your* problem - as
long as the expectations are clearly documented, it's the user's (ie:
coder) duty to provide something that fullfil the expectations.

Of course, in the practical world that all doesn't matter. But in the
theoretical world of the big coding farms, called business, that's one
cornerstone of success, in the tinking of managers and so.

Sorry, I live in a very practical world - and we're by no mean running
out of business here...

[Daniel Marcel Eichler]

Re: Am I missing something with Python not having interfaces?

Am Freitag 09 Mai 2008 10:19:45 schrieb Bruno Desthuilliers:

very often sees do-nothing catch-all try/catch blocks in Java -
which is way worse than just letting the exception propagate. I
find all this totally pointless, because there's just no way for a
compiler to check if your code is logically correct.

But it's enough if the called method exists and returns the correct
type. At least it prevents a crash.

>
Then providing an appropriate default in the base class is enough
too.

Only working *if* there is a base-class, and not only convention for
should-have-methods.

>That's the point. Interfaces garantee that a duck is a duck, an
>not only a chicken that quack.
>
Who cares if it's a chicken as long as it quacks when you ask her
to ? Now *This* is the whole point of chicken^Mduck typing, isn't
it ?-)

Ducks can also swim and fly. And if you need a really duck,

>
If you're code expects something that quacks, swims and flies,
anything that quacks, swims and flies is ok. You just don't care if
it's a duck or a flying whale with a quacking device tied to it.

Not the point.

Of course, in the practical world that all doesn't matter. But in
the theoretical world of the big coding farms, called business,
that's one cornerstone of success, in the tinking of managers and
so.

>
Sorry, I live in a very practical world - and we're by no mean
running out of business here...

Like i said.

[Rhamphoryncus]

Re: Am I missing something with Python not having interfaces?

On May 9, 1:53 pm, Daniel Marcel Eichler <onsen-n...@gmx.netwro te:

Am Freitag 09 Mai 2008 10:19:45 schrieb Bruno Desthuilliers:
>

>very often sees do-nothing catch-all try/catch blocks in Java -
>which is way worse than just letting the exception propagate. I
>find all this totally pointless, because there's just no way for a
>compiler to check if your code is logically correct.

>

But it's enough if the called method exists and returns the correct
type. At least it prevents a crash.

The application has already failed. You'd prefer it silently do the
wrong thing than get an explicit error message and stop?

>>That's the point. Interfaces garantee that a duck is a duck, an
>>not only a chicken that quack.

>

>Who cares if it's a chicken as long as it quacks when you ask her
>to ? Now *This* is the whole point of chicken^Mduck typing, isn't
>it ?-)

>

Ducks can also swim and fly. And if you need a really duck,

>

If you're code expects something that quacks, swims and flies,
anything that quacks, swims and flies is ok. You just don't care if
it's a duck or a flying whale with a quacking device tied to it.

>
Not the point.

It really is. They're only going to give you something they expect to
work. They might occasionally make a mistake and give you garbage,
but it's not worth the effort of trying to catch it early.

[Pete Forman]

Re: Am I missing something with Python not having interfaces?

I would suggest that using an interface at compile time is not the
only approach. Unit tests can be run on classes to check that they do
indeed quack.
--
Pete Forman -./\.- Disclaimer: This post is originated
WesternGeco -./\.- by myself and does not represent
pete.forman@wes terngeco.com -./\.- the opinion of Schlumberger or
http://petef.22web.net -./\.- WesternGeco.