pickle vs cPickle: error is not raised at unpickle / windows vs linux (!) (repost)

[Irmen de Jong]

Tim Peters wrote:

> This is why bug reports don't belong on comp.lang.python:  the context is
> lost, and I don't have time to go searching for it.  When you don't get a
> reply to a repeated question about dubious behavior, it's probably because
> nobody reading c.l.py knows the answer.  In this case, there are maybe two
> people on Earth who understand all the pickle changes going into 2.3, and
> Guido (the one who knows best) rarely has time to read c.l.py anymore.

Yes, well, then please tell me why it's not best to first ask around and
see if anybody else has seen the same problem or even knows a solution,
before submitting an unfounded bug report in sourceforge? Which *will*
take time and resources of those involved to look at it.
Would you be happier if I submit all 'dubious behavior' as bug reports
on SF, even those that are obviously my mistake, which probably can be
pointed out quickly by several knowledgeable people here on c.l.python?

At least I had the patience to first ask around and wait a while to
see if anything interesting comes up. It didn't. What's wrong with that?

> So, if you think this is an important question, please gather all the pieces
> and open a single coherent bug report on SourceForge.  Or, if it's an idle
> question, feel encouraged to keep replying to yourself about it on c.l.py
> <wink>.

It certainly is not an idle question. How on earth could I know in advance
that nobody would reply to it?

But, to please you, and myself, I will submit a bug report on SF now.

--Irmen