[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gnu-arch-users] is there demand for itla?

From: Tom Lord
Subject: Re: [Gnu-arch-users] is there demand for itla?
Date: Sun, 16 Nov 2003 14:29:14 -0800 (PST)

    > From: address@hidden

    > > The question I'm stuck on is how much "demand" there is for itla.
    > > Opinions?  Ideas? Rants?  Tips?   I think it's a pretty exciting idea 
    > > but I'm not sure what priority to give it relative to other projects.

    > If you feel you're gonna be busy for 3 months with this; I'd like to
    > see tla itself get the finishing touches first..

    > You can't build on a leaky foundation...
    > Many error messages are not readble for end users, how is itla gonny
    > handle them?

Many (most even?) of the error messages you're talking about are given
for invalid arguments provided to a program -- not enough input

It was a safe tactic to use to rapidly implement tla commands because
of the invariants in the lower level functions.

A huge part of what itla would do is (rather sophisticated) input
validation.   In other words, itla mostly won't run tla in a way that
leads to those "botched invariant"-style error messages.

    > The fact that ',,' dirs are left in my tree after a fatal is also quite
    > telling on error handling.

The innuendo of "quite telling" is just goofy.   This is (1) a fairly
minor issue, (2) hardly a "barrier to adoption", (3) easilly addressed
by a contributor.   I don't disagree that it's worth fixing -- just
not so sure that that should be a high priority for me.

    > Return codes are not entirely logical, I get '2' when everything went
    > just fine..

That is rather unlikely.   Perhaps you have found an exceptional case
but overall, exit codes tend to be reasonable.

    > I get '1' most of the time, when 0 would be unix default.

It is not.   tla mostly follows the unix conventions using 0 vs. non-0
for commands that only either succeed or fail, 0 vs. 1 vs. 2 for
diff-like and patch-like commands.   I am sure that there are glitches
in that area (that would likely be detected and fixed while working
with itla) but not the glitches you suggest.


That's a "small feature idea" bug report that isn't about error codes
at all but is about a request to permit abbreviating command names.

It is not a good small feature idea because it constrains the
namespace of commands-added-in-the-future in an undesirable way:
adding any new command which has a prefix which is an old command can
break scripts.

    > OTOH. I'm probably the last one that should tell you what to spent
    > your time on, but you asked for our opinions. :-)


    > ps. 3 months is pretty darn long; my guess would be 1-2 weeks..

"If a thing is worth doing, it is worth doing well."

I've implemented systems like itla twice before and worked with
similar systems written by others for around 20 years.   There's
plenty of subtleties to get right at the foundation.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]