From rlpowell at digitalkingdom.org  Tue Sep  2 20:37:46 2008
From: rlpowell at digitalkingdom.org (Robin Lee Powell)
Date: Tue, 2 Sep 2008 13:37:46 -0700
Subject: [heresy-devel] null/ bug.
Message-ID: <20080902203746.GF30355@digitalkingdom.org>


IMO, membership predicates should never except, but:

* (null/ 'e)

debugger invoked on a SB-KERNEL:CASE-FAILURE in thread #<THREAD
"initial thread" RUNNING {100266AC71}>: E fell through ETYPECASE
expression. Wanted one of (LAZY-LIST HERESY::LAZY-LIST-KNOWN-EMPTY
HERESY::LAZY-LIST-LIST-BASED SEQUENCE LIST).


-Robin

-- 
Lojban Reason #17: http://en.wikipedia.org/wiki/Buffalo_buffalo
Proud Supporter of the Singularity Institute - http://singinst.org/
http://www.digitalkingdom.org/~rlpowell/ *** http://www.lojban.org/


From rlpowell at digitalkingdom.org  Tue Sep  2 23:15:34 2008
From: rlpowell at digitalkingdom.org (Robin Lee Powell)
Date: Tue, 2 Sep 2008 16:15:34 -0700
Subject: [heresy-devel] null/ bug.
In-Reply-To: <20080902203746.GF30355@digitalkingdom.org>
References: <20080902203746.GF30355@digitalkingdom.org>
Message-ID: <20080902231534.GM30355@digitalkingdom.org>

On Tue, Sep 02, 2008 at 01:37:46PM -0700, Robin Lee Powell wrote:
> 
> IMO, membership predicates should never except, but:
> 
> * (null/ 'e)
> 
> debugger invoked on a SB-KERNEL:CASE-FAILURE in thread #<THREAD
> "initial thread" RUNNING {100266AC71}>: E fell through ETYPECASE
> expression. Wanted one of (LAZY-LIST HERESY::LAZY-LIST-KNOWN-EMPTY
> HERESY::LAZY-LIST-LIST-BASED SEQUENCE LIST).

This turns out to be a more serious problem than I thought;

(atom (list/ ))

returns t, so the obvious fix of

(or (atom x) (non-null/ x))

doesn't work.  On top of that, the "is a lazy list" predicate
doesn't seem to be exported from Heresy.  I had to add (atom nil) to
the etypecase for null/, but really I think that should be a
typecase with (t nil) at the end.

-Robin

-- 
Lojban Reason #17: http://en.wikipedia.org/wiki/Buffalo_buffalo
Proud Supporter of the Singularity Institute - http://singinst.org/
http://www.digitalkingdom.org/~rlpowell/ *** http://www.lojban.org/


From matthew.lamari at gmail.com  Tue Sep  2 23:38:47 2008
From: matthew.lamari at gmail.com (Matt Lamari)
Date: Tue, 02 Sep 2008 18:38:47 -0500
Subject: [heresy-devel] null/ bug.
In-Reply-To: <20080902231534.GM30355@digitalkingdom.org>
References: <20080902203746.GF30355@digitalkingdom.org>
	<20080902231534.GM30355@digitalkingdom.org>
Message-ID: <48BDCE87.4070704@gmail.com>


Would this do the job?  Basically adding null and "t" types to the 
typecase. . .   Or am I missing a problem here

(defun null/ (list)
  "Returns nil if list has contents, a value otherwise.  Will only 
traverse a single element for lazy-lists"
  (etypecase list
    (null t)
    (list (null list))
    (sequence (zerop (length list)))
    (lazy-list-list-based (null (get-list-head list)))
    (lazy-list-known-empty t)
    (lazy-list (with-traversal-result (value next) (resolved 
(get-traversal-result (get-call-for-first list))) (not (resolved next))))
    (t nil)))

Robin Lee Powell wrote:
> On Tue, Sep 02, 2008 at 01:37:46PM -0700, Robin Lee Powell wrote:
>   
>> IMO, membership predicates should never except, but:
>>
>> * (null/ 'e)
>>
>> debugger invoked on a SB-KERNEL:CASE-FAILURE in thread #<THREAD
>> "initial thread" RUNNING {100266AC71}>: E fell through ETYPECASE
>> expression. Wanted one of (LAZY-LIST HERESY::LAZY-LIST-KNOWN-EMPTY
>> HERESY::LAZY-LIST-LIST-BASED SEQUENCE LIST).
>>     
>
> This turns out to be a more serious problem than I thought;
>
> (atom (list/ ))
>
> returns t, so the obvious fix of
>
> (or (atom x) (non-null/ x))
>
> doesn't work.  On top of that, the "is a lazy list" predicate
> doesn't seem to be exported from Heresy.  I had to add (atom nil) to
> the etypecase for null/, but really I think that should be a
> typecase with (t nil) at the end.
>
> -Robin
>
>   



From rlpowell at digitalkingdom.org  Tue Sep  2 23:50:34 2008
From: rlpowell at digitalkingdom.org (Robin Lee Powell)
Date: Tue, 2 Sep 2008 16:50:34 -0700
Subject: [heresy-devel] null/ bug.
In-Reply-To: <48BDCE87.4070704@gmail.com>
References: <20080902203746.GF30355@digitalkingdom.org>
	<20080902231534.GM30355@digitalkingdom.org>
	<48BDCE87.4070704@gmail.com>
Message-ID: <20080902235034.GN30355@digitalkingdom.org>

That seems to work just fine.  Thanks!

-Robin

On Tue, Sep 02, 2008 at 06:38:47PM -0500, Matt Lamari wrote:
>
> Would this do the job?  Basically adding null and "t" types to the  
> typecase. . .   Or am I missing a problem here
>
> (defun null/ (list)
>  "Returns nil if list has contents, a value otherwise.  Will only  
> traverse a single element for lazy-lists"
>  (etypecase list
>    (null t)
>    (list (null list))
>    (sequence (zerop (length list)))
>    (lazy-list-list-based (null (get-list-head list)))
>    (lazy-list-known-empty t)
>    (lazy-list (with-traversal-result (value next) (resolved  
> (get-traversal-result (get-call-for-first list))) (not (resolved next))))
>    (t nil)))
>
> Robin Lee Powell wrote:
>> On Tue, Sep 02, 2008 at 01:37:46PM -0700, Robin Lee Powell wrote:
>>   
>>> IMO, membership predicates should never except, but:
>>>
>>> * (null/ 'e)
>>>
>>> debugger invoked on a SB-KERNEL:CASE-FAILURE in thread #<THREAD
>>> "initial thread" RUNNING {100266AC71}>: E fell through ETYPECASE
>>> expression. Wanted one of (LAZY-LIST HERESY::LAZY-LIST-KNOWN-EMPTY
>>> HERESY::LAZY-LIST-LIST-BASED SEQUENCE LIST).
>>>     
>>
>> This turns out to be a more serious problem than I thought;
>>
>> (atom (list/ ))
>>
>> returns t, so the obvious fix of
>>
>> (or (atom x) (non-null/ x))
>>
>> doesn't work.  On top of that, the "is a lazy list" predicate
>> doesn't seem to be exported from Heresy.  I had to add (atom nil) to
>> the etypecase for null/, but really I think that should be a
>> typecase with (t nil) at the end.
>>
>> -Robin
>>
>>   
>

-- 
Lojban Reason #17: http://en.wikipedia.org/wiki/Buffalo_buffalo
Proud Supporter of the Singularity Institute - http://singinst.org/
http://www.digitalkingdom.org/~rlpowell/ *** http://www.lojban.org/


From rlpowell at digitalkingdom.org  Wed Sep  3 22:17:09 2008
From: rlpowell at digitalkingdom.org (Robin Lee Powell)
Date: Wed, 3 Sep 2008 15:17:09 -0700
Subject: [heresy-devel] lazy-listp?
Message-ID: <20080903221709.GA31598@digitalkingdom.org>

Does the Heresy library export any way to test if something is a
lazy list?  If not, could it?

-Robin

-- 
Lojban Reason #17: http://en.wikipedia.org/wiki/Buffalo_buffalo
Proud Supporter of the Singularity Institute - http://singinst.org/
http://www.digitalkingdom.org/~rlpowell/ *** http://www.lojban.org/


From matthew.lamari at gmail.com  Thu Sep  4 01:44:16 2008
From: matthew.lamari at gmail.com (Matt Lamari)
Date: Wed, 03 Sep 2008 20:44:16 -0500
Subject: [heresy-devel] lazy-listp?
In-Reply-To: <20080903221709.GA31598@digitalkingdom.org>
References: <20080903221709.GA31598@digitalkingdom.org>
Message-ID: <48BF3D70.7040207@gmail.com>

I am inclined toward listp/ - and have it trigger off of listp AND 
lazy-lists.

(Given that almost all the other calls will accept sequences and lazy-lists)

Sound good?


Robin Lee Powell wrote:
> Does the Heresy library export any way to test if something is a
> lazy list?  If not, could it?
>
> -Robin
>
>   



From rlpowell at digitalkingdom.org  Thu Sep  4 05:25:04 2008
From: rlpowell at digitalkingdom.org (Robin Lee Powell)
Date: Wed, 3 Sep 2008 22:25:04 -0700
Subject: [heresy-devel] lazy-listp?
In-Reply-To: <48BF3D70.7040207@gmail.com>
References: <20080903221709.GA31598@digitalkingdom.org>
	<48BF3D70.7040207@gmail.com>
Message-ID: <20080904052504.GC31598@digitalkingdom.org>

I was thinking one of each:

(defun listp/ (list)
  (typecase list
    (lazy-list-list-based t)
    (lazy-list-pair-based t)
    (lazy-list t)
    (list t)
    (t nil)))

(defun lazy-listp (list)
  (typecase list
    (lazy-list-list-based t)
    (lazy-list-pair-based t)
    (lazy-list t)
    (t nil)))

BTW, the version you gave me seems to be quite a bit different (as
judged by diff -w -r) than the version currently on sourceforge;
which one's more recent?

-Robin

On Wed, Sep 03, 2008 at 08:44:16PM -0500, Matt Lamari wrote:
> I am inclined toward listp/ - and have it trigger off of listp AND
> lazy-lists.
>
> (Given that almost all the other calls will accept sequences and
> lazy-lists)
>
> Sound good?
>
>
> Robin Lee Powell wrote:
>> Does the Heresy library export any way to test if something is a
>> lazy list?  If not, could it?
>>
>> -Robin
>>
>>   
>

-- 
Lojban Reason #17: http://en.wikipedia.org/wiki/Buffalo_buffalo
Proud Supporter of the Singularity Institute - http://singinst.org/
http://www.digitalkingdom.org/~rlpowell/ *** http://www.lojban.org/


From matthew.lamari at gmail.com  Thu Sep  4 05:50:41 2008
From: matthew.lamari at gmail.com (Matt Lamari)
Date: Thu, 04 Sep 2008 00:50:41 -0500
Subject: [heresy-devel] lazy-listp?
In-Reply-To: <20080904052504.GC31598@digitalkingdom.org>
References: <20080903221709.GA31598@digitalkingdom.org>
	<48BF3D70.7040207@gmail.com>
	<20080904052504.GC31598@digitalkingdom.org>
Message-ID: <48BF7731.4090303@gmail.com>


I'm about to publish a new version.  I'll add lazy-listp as well.

The one on Sourceforge is a fair bit older than the one you have - what 
you have now is similar to my next deployment.



Robin Lee Powell wrote:
> I was thinking one of each:
>
> (defun listp/ (list)
>   (typecase list
>     (lazy-list-list-based t)
>     (lazy-list-pair-based t)
>     (lazy-list t)
>     (list t)
>     (t nil)))
>
> (defun lazy-listp (list)
>   (typecase list
>     (lazy-list-list-based t)
>     (lazy-list-pair-based t)
>     (lazy-list t)
>     (t nil)))
>
> BTW, the version you gave me seems to be quite a bit different (as
> judged by diff -w -r) than the version currently on sourceforge;
> which one's more recent?
>
> -Robin
>
> On Wed, Sep 03, 2008 at 08:44:16PM -0500, Matt Lamari wrote:
>   
>> I am inclined toward listp/ - and have it trigger off of listp AND
>> lazy-lists.
>>
>> (Given that almost all the other calls will accept sequences and
>> lazy-lists)
>>
>> Sound good?
>>
>>
>> Robin Lee Powell wrote:
>>     
>>> Does the Heresy library export any way to test if something is a
>>> lazy list?  If not, could it?
>>>
>>> -Robin
>>>
>>>   
>>>       
>
>   



From matthew.lamari at gmail.com  Thu Sep  4 06:30:00 2008
From: matthew.lamari at gmail.com (Matt Lamari)
Date: Thu, 04 Sep 2008 01:30:00 -0500
Subject: [heresy-devel] Heresy_0_0_4 published and ASDF-installable
Message-ID: <48BF8068.4010101@gmail.com>


I got a linux 64/SBCL and got it to work, and made it ASDF-installable 
(I'm a LW/Win32 user, hence the outstanding issues)

That the unit-tester happens to recompile (to test some compiler-macros) 
makes running the unit-test particularly painful.  Tons of style 
warnings (most are repeats of the "sink-count" one)

the CSV test seems to have failed - there will be another revision 
within the week to clean this up; but hopefully this makes the library a 
bit more accessible.





From rlpowell at digitalkingdom.org  Thu Sep  4 19:39:18 2008
From: rlpowell at digitalkingdom.org (Robin Lee Powell)
Date: Thu, 4 Sep 2008 12:39:18 -0700
Subject: [heresy-devel] Heresy_0_0_4 published and ASDF-installable
In-Reply-To: <48BF8068.4010101@gmail.com>
References: <48BF8068.4010101@gmail.com>
Message-ID: <20080904193918.GJ31598@digitalkingdom.org>

On Thu, Sep 04, 2008 at 01:30:00AM -0500, Matt Lamari wrote:
>
> I got a linux 64/SBCL and got it to work, and made it
> ASDF-installable  (I'm a LW/Win32 user, hence the outstanding
> issues)

Only one problem: the perms are *way* off.  All files have no
permissions at all, including the directory.  I don't know how to
fix this with whatever tar app you are using, but if you tell me
what it is I can try to figure it out.

Other than that, and that it still runs the unit tests on every
(require 'heresy), it's fine.  Thanks!

-Robin

-- 
Lojban Reason #17: http://en.wikipedia.org/wiki/Buffalo_buffalo
Proud Supporter of the Singularity Institute - http://singinst.org/
http://www.digitalkingdom.org/~rlpowell/ *** http://www.lojban.org/


From rlpowell at digitalkingdom.org  Fri Sep  5 00:59:24 2008
From: rlpowell at digitalkingdom.org (Robin Lee Powell)
Date: Thu, 4 Sep 2008 17:59:24 -0700
Subject: [heresy-devel] Equality testing.
Message-ID: <20080905005924.GR31598@digitalkingdom.org>


* (equal (LIST/ 'B 'A 'Q '(C 1)) (LIST/ 'B 'A 'Q '(C 1)))
NIL

So, umm, how do I do that?  Convert to a regular list first?

-Robin

-- 
Lojban Reason #17: http://en.wikipedia.org/wiki/Buffalo_buffalo
Proud Supporter of the Singularity Institute - http://singinst.org/
http://www.digitalkingdom.org/~rlpowell/ *** http://www.lojban.org/


From matthew.lamari at gmail.com  Fri Sep  5 21:08:52 2008
From: matthew.lamari at gmail.com (Matt Lamari)
Date: Fri, 05 Sep 2008 16:08:52 -0500
Subject: [heresy-devel] Equality testing.
In-Reply-To: <20080905005924.GR31598@digitalkingdom.org>
References: <20080905005924.GR31598@digitalkingdom.org>
Message-ID: <48C19FE4.2010105@gmail.com>



(note - has to be read in heresy namespace, i.e. add it to the file)
 
I believe it works correctly with sub-lists too.

Also note - consumes stack - when I get time I'll rewrite it in 
continuation-calling form.  It does not consume stack based on list 
"length"; but on list-of-list depth.  I'll also optimize it better for 
speed (expand out/not use those readpoints and their performance hit).


(defun equal/ (a b)
  (let ((infinite (iterate/ #'identity nil))
        (end (cons nil nil))
        (a-is-list (listp/ a))
        (b-is-list (listp/ b)))
    (cond
     ((and a-is-list b-is-list)
      (let ((a (read-point-built (to-lazy-list a)))
            (b (read-point-built (to-lazy-list b))))
        (unless
            (loop while not (or (read-point-at-end a) (read-point-at-end 
b)) do
                  (unless (equal/ (read-point-value a) (read-point-value b))
                    (return-from nil t))
                  (setq a (read-point-advanced a))
                  (setq b (read-point-advanced b)))
          (not (or (read-point-at-end a) (read-point-at-end b))))))
     ((or a-is-list b-is-list) nil)
     (t (equal a b)))))


Robin Lee Powell wrote:
> * (equal (LIST/ 'B 'A 'Q '(C 1)) (LIST/ 'B 'A 'Q '(C 1)))
> NIL
>
> So, umm, how do I do that?  Convert to a regular list first?
>
> -Robin
>
>