Re: [gentoo-pms] [PATCH] Make it clear that PMs are allowed to handle 'invalid' names.

[Brian Harring]

On Sun, Sep 23, 2012 at 09:35:19AM +0200, Micha?? G??rny wrote:
> On Sat, 22 Sep 2012 18:24:40 -0700
> Brian Harring <ferringb@×××××.com> wrote:
>
> > On Sun, Sep 23, 2012 at 12:41:06AM +0200, Micha?? G??rny wrote:
> > > ---
> > >  names.tex | 6 ++++--
> > >  1 file changed, 4 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/names.tex b/names.tex
> > > index decc8f4..5a3866f 100644
> > > --- a/names.tex
> > > +++ b/names.tex
> > > @@ -2,8 +2,10 @@
> > >
> > >  \section{Restrictions upon Names}
> > >
> > > -No name may be empty. Package managers must not impose fixed upper
> > > boundaries upon the length of any -name. A package manager should
> > > indicate or reject any name that is invalid according to these
> > > rules. +No name may be empty. Package managers must not impose
> > > fixed upper +boundaries upon the length of any name. A package
> > > manager should +indicate or reject any name that is invalid
> > > according to these rules. +Package managers are allowed to accept
> > > names not following those rules.
> >
> > "Here's some rules.  You must abide by them.  Or you can ignore them."
>
> That's wording ulm suggested, mine was more liberal.
>
> The idea is that it is enough to *warn* about them but if package
> manager *can* handle them, it should.

Jesus, do you folks even get compatibility?

This breaks existing PMs.  You can't retroactively make every version 
that's out there convert it into a warning.  That's the fucking point
of this.

Solve that, this issue goes away, and I/other shut up.


> > Offhand, and frankly I'm kind of pissed I had to even do this legwork
> > because your noisy fucking ass couldn't be bothered to do the
> > the necessary legwork looking at sources (god forbid rather
> > than chattering like a poop-flinging monkey, someone fire up a term
> > and go grep'ing and reading sources); right now, changing the rule
> > that way means that a package name of 'diffball-1' would break on
> > portage, pkgcore, and paludis.  In other words, as has been said,
> > all current versions, and likely versions from years ago, *will*
> > go boom if they encounter this.
>
> You could've just asked.

I did.

Also, I shouldn't have to ask "You're completely invaliding a rule.  
You *did* make sure this didn't break shit, right?".  Just the same as 
I shouldn't have to tell folk "please don't aim the shotgun at your 
foot."

Same thing here, skated right on past the point of breakage.


> > Minimally, pkgcore and paludis, current versions see this anywhere,
> > they *will* throw an exception because its broken data- to make it
> > into the vdb is either corruption of the disk, or corruption of the
> > VDB by a broken PM implementation.  Both things have to be flagged,
> > even though it sucks, and that's what we do.
>
> No.

Dude, you cannot say 'no' when I state how the PMs work /now/.

Seriously, emerge -p dev-util/diffball-1a alone for christ sake.


> User is free to use 'invalid' package names whenever he likes
> and 'break' the system whenever he likes. PMS should not encourage to
> lift this freedom for the sake of compatibility.

A user is free to break shit all they want on their system.  It /is/ 
their system after all.

You aren't trying to mandate that; were it limited to people pulling 
shit like EAPI=4-python, doing it outside PMS spec, ok, annoying, but 
ok.

You want PMS EAPI's to allow this, to retroactively allow behaviour 
that breaks all compliant PMS.

Do you even understand what compatibility means?


> Much like POSIX doesn't prohibit systems from accepting any characters
> outside the portable character set for filenames. If you use them, you
> don't expect 100% portability.

Posix doesn't go changing standards that actively break compliant 
versions (and in the case where they, which is stupidly rare and 
corner case tightening of rules, they sure as hell don't break all 
implementations out there).  The fact you're pointing at posix, and 
missing this very, *very* core rule understanding of posix is frankly 
kind of scary.


> 'Be conservative in what you send, and liberal in what you accept'

How I wish you'd actually take that quote to heart when it came to 
your emails discussions.

I also wish you understand what you referenced; the point there is to 
write software up front to be lenient if possible when meaning is 
clear, in the name of having things work.  Key words "having things 
work"; you point at the quote while proposing we break shit.

People who bust out quotes rarely actually understand them, let alone 
would they be doing so if they were right.

-Harring