Re: missing methods
Christopher Stacy <[email protected]> Fri, 3 Apr 2026 16:44:49 -0400
| Newsgroups | gmane.lisp.steel-bank.general |
|---|---|
| Message-ID | <[email protected]> |
This is a multi-part message in MIME format.
--===============8899224833824024175==
Content-Type: multipart/alternative;
boundary="------------sQlMaoACKm3m6EAZe3PvnSOs"
Content-Language: en-US
This is a multi-part message in MIME format.
--------------sQlMaoACKm3m6EAZe3PvnSOs
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
On 4/3/26 11:48 AM, Stas Boukarev wrote:
/paraphrasing: maybe turn on debugging?/
Ladies and Gentlemen, We Have A Winner.
Stas Boukarev!
/(* In the unlikely event you didn't already know he was a winner....)/
I was about to go see if ASDF had a keyword that turns on those flags,
but then his suggestion came in and I just went and set them like Stas
suggested. Probably more reliable that way, anyway. Thank You, Stas. The
problem was then solved and fixed in the source code in under 3 seconds
after seeing the verbose compiler output.
I can't believe how dumb this one is. And now that I think back, I
/have/ seen something like it before, although I can't remember the
exact details. (It might have even been this same thing. Too long ago,
but rings a bell.)
The lesson here is to think of simple things, when very fundamental
things seem inexplicably borked. This is a personal vulnerability of
mine, and not the first time I have fallen victim to myself this way. A
CLOS method doesn't work and my first thought is: "Did I forget
something about CLOS?" Followed by: "No, it's supposed to work like I
think. So... is CLOS broken?" And I begin imagining all the things that
could go wrong based on every CLOS implementation I ever looked at. Most
of which are based on PCL, and none of which I've truly studied in depth
(ie. not to the point I could re-create them just off the top of my
head). The mind whirls. Then it quickly turns to contamination of the
compile-time environment, and I go looking for anything I (or any
libraries) might have possibly done to mess that up. Or maybe there is a
bug in the compiler itself. Down all the rabbit holes of disbelief and
panic.
Even though the whole time, I am thinking, "Chris, you idiot, this has
to be something you did, and it's probably simple." But there's too much
bias about it being a bug, and I just don't SEE anything wrong with my
code. I've looked at this code a hundred times and haven't seen anything
wrong with it! Ahem.
In my lame defense, the last time I reported a bug in a CLOS
implementation -- a case similar to this -- it was in fact the CLOS
implementation that was quite broken. But don't worry, that was in
emacs-lisp. Back when I was originally hoping to have
cross-compatibility. And it was part of this same project. I got pretty
far there but eventually punted to CL-only. And I guess that experience
was a factor in my bias this time.
Now we know the bug is here. That it has taken human form. Somehow,
cstacy must convince a disbelieving world that the nightmare has already
begun.
Have you guessed what it was, yet?
If not, here's the big hint: I am on a standard American QWERTY keyboard.
Second huge hint: It is SOooo obvious, and I did actually already look
for a much more complex, but even more obvious, similar mistake. Not
finding that, I didn't think of the most trivial version. Because I was
looking for something intentional I might have done wrong. Rather than
something I never remotely intended.
READ this far, and still don't know what it was? I am ambivalent about
eliciting an educational list of responses on this mailing list, so I
will spill the beans here. (Too bad there's no reasonable way to cover
this up and uncover it with a click.) I'll just leave you with the
following quote:
;; This next piece of code is tricky.
'; Here it comes.
;; You'll love this.
(wtf)
Thanks again!
PS. Would a text-based language's compiler (or linter) have noticed this
mistake, where Lisp doesn't see it because of READ?
--------------sQlMaoACKm3m6EAZe3PvnSOs
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: 7bit
<!DOCTYPE html>
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<div class="moz-cite-prefix">On 4/3/26 11:48 AM, Stas Boukarev
wrote:<br>
<i>paraphrasing: maybe turn on debugging?</i><br>
<br>
Ladies and Gentlemen, We Have A Winner.<br>
Stas Boukarev!<br>
<font size="1"><i>(* In the unlikely event you didn't already know
he was a winner....)</i></font><br>
<br>
I was about to go see if ASDF had a keyword that turns on those
flags, but then his suggestion came in and I just went and set
them like Stas suggested. Probably more reliable that way, anyway.
Thank You, Stas. The problem was then solved and fixed in the
source code in under 3 seconds after seeing the verbose compiler
output.<br>
<br>
I can't believe how dumb this one is. And now that I think back, I
<i>have</i> seen something like it before, although I can't
remember the exact details. (It might have even been this same
thing. Too long ago, but rings a bell.)<br>
<br>
The lesson here is to think of simple things, when very
fundamental things seem inexplicably borked. This is a personal
vulnerability of mine, and not the first time I have fallen victim
to myself this way. A CLOS method doesn't work and my first
thought is: "Did I forget something about CLOS?" Followed by: "No,
it's supposed to work like I think. So... is CLOS broken?" And I
begin imagining all the things that could go wrong based on every
CLOS implementation I ever looked at. Most of which are based on
PCL, and none of which I've truly studied in depth (ie. not to the
point I could re-create them just off the top of my head). The
mind whirls. Then it quickly turns to contamination of the
compile-time environment, and I go looking for anything I (or any
libraries) might have possibly done to mess that up. Or maybe
there is a bug in the compiler itself. Down all the rabbit holes
of disbelief and panic.<br>
<br>
Even though the whole time, I am thinking, "Chris, you idiot, this
has to be something you did, and it's probably simple." But
there's too much bias about it being a bug, and I just don't SEE
anything wrong with my code. I've looked at this code a hundred
times and haven't seen anything wrong with it! Ahem.<br>
<br>
In my lame defense, the last time I reported a bug in a CLOS
implementation -- a case similar to this -- it was in fact the
CLOS implementation that was quite broken. But don't worry, that
was in emacs-lisp. Back when I was originally hoping to have
cross-compatibility. And it was part of this same project. I got
pretty far there but eventually punted to CL-only. And I guess
that experience was a factor in my bias this time.<br>
<br>
</div>
<div class="moz-cite-prefix">Now we know the bug is here. That it
has taken human form. Somehow, cstacy must convince a disbelieving
world that the nightmare has already begun.<br>
<br>
Have you guessed what it was, yet?<br>
<br>
If not, here's the big hint: I am on a standard American QWERTY
keyboard.<br>
<br>
Second huge hint: It is SOooo obvious, and I did actually already
look for a much more complex, but even more obvious, similar
mistake. Not finding that, I didn't think of the most trivial
version. Because I was looking for something intentional I might
have done wrong. Rather than something I never remotely intended.<br>
<br>
READ this far, and still don't know what it was? I am ambivalent
about eliciting an educational list of responses on this mailing
list, so I will spill the beans here. (Too bad there's no
reasonable way to cover this up and uncover it with a click.) I'll
just leave you with the following quote:</div>
<div class="moz-cite-prefix"><br>
</div>
<div class="moz-cite-prefix"><font face="Monaco" size="1">;; This
next piece of code is tricky.</font><br>
<font face="Monaco" size="1">'; Here it comes.</font><br>
<font face="Monaco" size="1">;; You'll love this.</font><br>
<br>
<font face="Monaco" size="1">(wtf)</font><br>
<br>
Thanks again!<br>
<br>
PS. Would a text-based language's compiler (or linter) have
noticed this mistake, where Lisp doesn't see it because of READ?<br>
<br>
<br>
<br>
<br>
</div>
<div class="moz-cite-prefix"><br>
</div>
</body>
</html>
--------------sQlMaoACKm3m6EAZe3PvnSOs--
--===============8899224833824024175==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
--===============8899224833824024175==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
_______________________________________________
Sbcl-help mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/sbcl-help
--===============8899224833824024175==--