[Asdf-devel] ASDF does not signal deferred non-style warnings as an error

[Robert P. Goldman]

Faré wrote:
> On Wed, Oct 29, 2014 at 11:40 PM, Robert P. Goldman <rpgoldman at sift.info> wrote:
>> #P"/Users/rpg/lisp/asdf/build/fasls/sbcl-1.2.0-macosx-x64/asdf/test/fun-with-undefined-locals.fasl"
>>
>> ; file: /Users/rpg/lisp/asdf/test/fun-with-undefined-locals.lisp
>> ; in: DEFUN FUN-WITH-UNDEFINED-LOCALS-PACKAGE::FOO
>> ;     (+ FUN-WITH-UNDEFINED-LOCALS-PACKAGE::A
>> FUN-WITH-UNDEFINED-LOCALS-PACKAGE::B)
>> ;
>> ; caught WARNING:
>> ;   undefined variable: FUN-WITH-UNDEFINED-LOCALS-PACKAGE::A
>> ;
>> ; caught WARNING:
>> ;   undefined variable: FUN-WITH-UNDEFINED-LOCALS-PACKAGE::B
>> ;
>> ; compilation unit finished
>> ;   Undefined variables:
>> ;     FUN-WITH-UNDEFINED-LOCALS-PACKAGE::A
>> FUN-WITH-UNDEFINED-LOCALS-PACKAGE::B
>> ;   caught 2 WARNING conditions
>> ;
>> ; compilation unit aborted
>> ;   caught 2 fatal ERROR conditions
>> Expected error type: COMPILE-FAILED-ERROR
>> Script succeeded
>> Using sbcl, test-deferred-warnings.script passed
>>
>>
>> How do you distinguish between the build failing and succeeding but
>> being followed by an error signal?
>>
> Nothing in ASDF should cause an error signal at the end of the W-C-U.

I think I'm misunderstanding the following from your previous message:

 "if deferred warnings are disabled (the default), then the
compilation succeeds, and I get an undefined variable warning at the
end of the build but that doesn't cause a build failure, because
operate's with-compilation-unit doesn't catch these warnings (and
can't replay those from previous sessions, anyway)."

OK, so there's no error signal because the warning is not caught in the
W-C-U.  TBH, I don't understand this.  When you say "caught" here, do
you mean "handled"?  And why should OPERATE need to catch these to avoid
a spurious compilation success? Shouldn't this warning cause
COMPILE-FILE's to return a non-nil warningsp?

> You're discussing the behavior of a file that isn't checked in;

No, this file is checked in, and you should see these results if you run
the test suite after pulling master from cl.net.


> does it have interesting handler-bind's?

No, it's just Ilya's example.

> Does your SBCL and/or some configuration file include
> some interesting patch with respect to W-C-U and/or compile conditions?

Not that I know of.

> (Which version of SBCL is it?)

SBCL 1.2.0

> I don't understand what's going on for you.
> 
> 
>>> On the other hand, differences between compilers easily explain
>>> variations between condition reported by ASDF, since they depend both
>>> on what conditions the compiler reports, that are not portable, and on
>>> the *compile-file-warnings-behaviour* and
>>> *compile-file-failure-behaviour* values, that are also unportable.
>>> Note that unless you :force t, non-fatal warnings can disappear the
>>> second time around.
>> It is the "not portable" part of this that makes it seem like a
>> maintenance nightmare, and not ASDF's job. If SBCL wants to signal
>> warnings at oddball times, that's too bad, but I don't see that it's
>> ASDF's job to fix that.  We have to leave some fun for the SBCL
>> maintainers ;-)
>>
> ASDF is a portability layer. It has to handle some non-portability
> in underlying implementations and provide portable abstractions.
> In this case, how to interpret the return values of COMPILE-FILE
> is de facto non-portable, with different implementations having
> wildly different interpretations of the CLHS regarding secondary values.
> ASDF thus provides the variables *compile-file-warnings-behaviour* and
> *compile-file-failure-behaviour* to interpret the warnings-p and failure-p
> return values of compile-file. It is *already* a nightmare,
> and ASDF has been *solving* that nightmare for a decade. Thanks, danb.
> 
> As for fine control of compiler conditions, ASDF doesn't do it,
> but it provides hooks in the form of uninteresting-conditions
> and compiler-hook. A colleague has written a portability layer
> for compiler conditions. I'm hoping he will clean it up and
> opensource it at some point.
> 
> 
>>> NB: Back when I first pre-released ASDF3, I wanted to enable
>>> deferred-warnings by default, but this turned out to be a
>>> community-wide coordination nightmare, and I had to bail out. If
>>> someone wants to make it happen, he'll have to make sure that it
>>> doesn't break quicklisp and then still have to convince the community
>>> and the ASDF maintainer that the change is worth it and that enough
>>> publicity is made that random users don't get badly surprised.
>> My experience with the deferred warnings on Allegro left me deeply
>> dissatisfied. ACL seems to deal with deferring warnings to the right
>> place simply through use of WITH-COMPILATION-UNIT.  In particular, it
>> correctly waits on UNDEFINED-FUNCTION warnings until the whole
>> compilation unit has been processed, yet I don't recall delays in seeing
>> UNDEFINED-VARIABLE references (I should check to confirm this, but I
>> feel sure I would have noticed if undefined lexical variable warnings
>> were deferred till the end of the build).
>>
> One reason why I didn't include a test for undefined-variable behavior
> in the existing ASDF3 test/test-deferred-warnings.script is precisely
> that the way this is treated is highly implentation-dependent, and
> I didn't have the time and energy to track down the behavior on 16
> implementations and figure out a portability layer or at least a
> conditional test.
> 
>> This suggested some conjectures:
>>
>> 1.  AFAICT, using ASDF's deferred-warnings would have required first
>> breaking the logic that ACL already provides, and then recreating it in
>> ASDF.
>>
> No. It doesn't require anything but suitable defaults.
> It might "just" be that check-deferred-warnings needs be made more clever,
> and/or might need to have its own separate variants of
> *compile-file-warnings-behaviour* and *compile-file-failure-behaviour*,
> and maybe also *compile-check*.

This still means "take code that works, break it, and then add more code
to fix the breakage."
> 
>> 2.  If ACL can handle this sensibly w/o requiring any outside assistance
>> from ASDF, then I don't believe that this should be ASDF's job.
>>
> By definition, it's ASDF's job to put the correct W-C-U, and to emulate
> a system-wide W-C-U when warnings were deferred in a previous session
> (i.e. file "a" was previously compiled with deferred warnings, and least
> we forgo all intra-system incrementality, we don't want to recompile it).
> The implementation just can't make that stuff up out of thin air.
> Hopefully, we don't have to write clever handler-bind's
> as part of ASDF itself, but there's a possibility we might,
> and indeed someone should look at the undefined-variable behavior on
> each and every of the 16 implementations.

I'd be willing to meet the implementations halfway on this.  If we can
specify the information we need from the implementors to do this right,
then we can ask them to provide an API that will provide that
information. If they don't, then ASDF won't perform as well on that
implementation.

I'm open to a volunteer to look at the undefined-variable behavior on
each and every of the 16 implementations, but as Mr. Zimmerman once said
"it ain't me you're lookin' for, babe."

I'm interested in shifting a lot of this "grovel over implementation
internals" work to the place where it can best be done -- and more
importantly, where it can best be maintained: to the implementors.
> 
>> 3.  The job of deferring warnings properly seems to involve thinking
>> hard about which warnings should and should not be deferred. That's
>> something that the implementation's maintainers will always be able to
>> do better than we can, since the condition classes are poorly
>> standardized.
>>
> No. The implementation is already doing the deferring or not.
> ASDF's job is just to capture the relevant warnings, and either
> (1) effectively discard them, which it currently does with its W-C-U,
> unless some implementations convert them to errors at the end of W-C-U
> and that's a huge incrementality and portability issue that needs be solved
> or (2) replay them deterministically at the end of the system's compilation,
> which ASDF does when you enable-deferred-warnings-check.
> 
>> 4.  It seems like the implementation needs to think carefully about how
>> to defer warnings in WITH-COMPILATION-UNIT anyway.
>>
> No. We only need worry if some implementations convert warnings to errors
> at the end of W-C-U, in which case it becomes much more important to
> implement and deploy deferred-warnings.

Why would anyone do this conversion? Is it because there's a warning
because locally the compiler can't tell that code is messed up, but once
the W-C-U has exited, the compiler assumes some sort of closure?  So if
I don't see some form of declaration locally, I signal a warning, but it
is held until the compilation-unit ends, and if it can't be handled
then, we treat it as an error?  In that case why signal a warning and
turn it into an error, instead of just signaling an error?  Is there a
concrete example of this behavior?

thanks,
r