Python 1.6 alpha 1 released

James Logajan JamesL at Lugoj.Com
Sat Apr 1 00:46:21 EST 2000


Fredrik Lundh wrote:
> 
> Bjorn Pettersen <bjorn at roguewave.com> wrote:
> > Just read the announcement page, and found that socket.connect() no
> > longer takes two arguments as was previously documented.
> 
> note that the actual reference page says that connect takes
> just one argument, and it has done so since at least 1.4.
> 
> > A quick look shows that this breaks all the network scripts I have
> > installed (at least the ones that I found, undoubtedly there are many
> > more).  Because of this I will put any upgrade plans on hold.
> 
> were you really planning to *upgrade* to a rough alpha?
> 
> </F>

Sorry, but Bjorn Pettersen has a perfectly valid point. The copy of the
reference page that I have on hand at the moment (1.5b1) uses the word
"pair", not "tuple" and the example clearly shows a two argument call, not a
one argument call using a tuple. In the face of conflicting information in
the Python library reference manual (which has been significantly less than
perfect), I have tended to use examples as a guide. I should also note that
a quick check of some old code of mine indicates use of the two argument
form for connect. So you have at least two idiots who read the library
reference and still managed to do the wrong thing (and made it work; I
wonder how...). Worse, the idiot implementor of the connect interface
managed to support the same thing! I wonder who the third idiot is? I've
noticed the world is full of idiots like me and I bet there is a lot more
code out there that will break due to this.

Also, how good are you at math? You seem to feel that 5 minutes of "Guido
the god's" time is of much more value than that of 1000 knuckle-dragging
Neanderthal programmers like myself spending a mere 10 minutes each to make
sure their code doesn't break. Of course, if we happen to take more than 10
minutes its our own damn fault I'm sure you'll agree.

Now how did I get so pissed over this exchange of posts?



More information about the Python-list mailing list