2014-10-02T00:02:47Z Bicyclidine quit (Ping timeout: 258 seconds)
2014-10-02T00:05:45Z slyrus quit (Ping timeout: 272 seconds)
2014-10-02T00:42:36Z scymtym_ quit (Ping timeout: 260 seconds)
2014-10-02T00:55:35Z karswell` joined #sbcl
2014-10-02T01:34:56Z davazp quit (Remote host closed the connection)
2014-10-02T02:30:43Z White_Flame quit (Remote host closed the connection)
2014-10-02T02:31:06Z White_Flame joined #sbcl
2014-10-02T02:38:32Z christoph_debian quit (Ping timeout: 245 seconds)
2014-10-02T02:51:54Z christoph_debian joined #sbcl
2014-10-02T03:41:43Z slyrus joined #sbcl
2014-10-02T04:00:14Z nyef quit (Quit: G'night all.)
2014-10-02T04:04:42Z gingerale joined #sbcl
2014-10-02T04:50:05Z edgar-rft joined #sbcl
2014-10-02T04:56:18Z leo2007 quit (Quit: rcirc on GNU Emacs 24.4.50.1)
2014-10-02T04:57:01Z gingerale quit (Ping timeout: 260 seconds)
2014-10-02T05:20:33Z oleo__ quit (Quit: Verlassend)
2014-10-02T06:49:22Z fridim__ joined #sbcl
2014-10-02T07:23:14Z fikusz quit (Ping timeout: 258 seconds)
2014-10-02T07:36:56Z fikusz joined #sbcl
2014-10-02T07:37:40Z leo2007 joined #sbcl
2014-10-02T07:37:41Z leo2007 quit (Remote host closed the connection)
2014-10-02T07:43:09Z leo2007 joined #sbcl
2014-10-02T08:02:13Z slyrus quit (Ping timeout: 272 seconds)
2014-10-02T08:05:06Z angavrilov joined #sbcl
2014-10-02T08:26:28Z akkad quit (Quit: Emacs must have died)
2014-10-02T08:29:10Z alchemis7 left #sbcl
2014-10-02T08:29:14Z alchemis7 joined #sbcl
2014-10-02T08:36:20Z minion quit (Remote host closed the connection)
2014-10-02T08:36:20Z specbot quit (Remote host closed the connection)
2014-10-02T08:37:27Z minion joined #sbcl
2014-10-02T08:37:46Z specbot joined #sbcl
2014-10-02T12:37:41Z nyef joined #sbcl
2014-10-02T12:50:15Z attila_lendvai joined #sbcl
2014-10-02T12:56:08Z psilord quit (Quit: Leaving.)
2014-10-02T13:20:34Z flip216 is now known as flip214
2014-10-02T13:33:48Z loke_ joined #sbcl
2014-10-02T13:41:57Z eudoxia joined #sbcl
2014-10-02T13:51:46Z oleo joined #sbcl
2014-10-02T14:22:51Z stassats joined #sbcl
2014-10-02T14:35:04Z psilord joined #sbcl
2014-10-02T14:53:07Z gingerale joined #sbcl
2014-10-02T15:51:49Z edgar-rft quit (Quit: you are not expected to understand this message)
2014-10-02T15:55:57Z slyrus joined #sbcl
2014-10-02T16:28:12Z slyrus quit (Ping timeout: 245 seconds)
2014-10-02T16:39:22Z slyrus joined #sbcl
2014-10-02T16:51:04Z attila_lendvai quit (Quit: Leaving.)
2014-10-02T17:00:21Z slyrus quit (Ping timeout: 260 seconds)
2014-10-02T17:13:56Z hlavaty` quit (Read error: Connection reset by peer)
2014-10-02T17:14:27Z hlavaty` joined #sbcl
2014-10-02T17:24:48Z stassats quit (Ping timeout: 246 seconds)
2014-10-02T18:05:19Z akkad joined #sbcl
2014-10-02T18:16:38Z slyrus joined #sbcl
2014-10-02T18:32:22Z Blkt: is there a way to debug sbcl's type inference?
2014-10-02T18:33:08Z nyef: The inference process itself, or derive-type method that's doing something wrong?
2014-10-02T18:33:29Z Blkt: ehrm
2014-10-02T18:33:32Z Blkt: let me try to rephrase
2014-10-02T18:34:05Z Blkt: executing some code I get
2014-10-02T18:34:06Z Blkt: The value -9
2014-10-02T18:34:06Z Blkt: is not of type
2014-10-02T18:34:06Z Blkt:   (OR (SINGLE-FLOAT (0.0)) (DOUBLE-FLOAT (0.0d0))
2014-10-02T18:34:06Z Blkt:       (RATIONAL (0))).
2014-10-02T18:34:06Z Blkt:    [Condition of type TYPE-ERROR]
2014-10-02T18:34:30Z Blkt: but with (declaim (optimize speed)) the code works correctly
2014-10-02T18:34:41Z eudoxia quit (Quit: Lost terminal)
2014-10-02T18:34:45Z Blkt: (it's Norvig's PAIP code from Chapter 18)
2014-10-02T18:34:52Z Blkt: (Othello bots)
2014-10-02T18:35:02Z Blkt: (and SBCL is 1.2.1 on Windows)
2014-10-02T18:35:22Z nyef: So the question is either "where does the -9 come from" or "why isn't it some form of zero"?
2014-10-02T18:35:36Z Blkt: no, the -9 is expected to be there
2014-10-02T18:35:38Z nyef: Err... "why is the compiler expecting it to be some form of zero".
2014-10-02T18:35:46Z Blkt: exactly
2014-10-02T18:36:43Z Blkt: are you using SBCL 1.2.4?
2014-10-02T18:37:04Z nyef: My copy of PAIP is unavailable to me for about another 75 hours or so.
2014-10-02T18:38:25Z nyef: I don't even know which version of SBCL I'm using where at this point.
2014-10-02T18:38:27Z Blkt: http://paste.lisp.org/display/143917
2014-10-02T18:39:02Z nyef: Okay, so the -9 is in all-directions?
2014-10-02T18:39:16Z nyef: What function is the error in?
2014-10-02T18:39:35Z Blkt: The value -1
2014-10-02T18:39:35Z Blkt: is not of type
2014-10-02T18:39:35Z Blkt:   (OR (SINGLE-FLOAT (0.0)) (DOUBLE-FLOAT (0.0d0))
2014-10-02T18:39:35Z Blkt:       (RATIONAL (0))).
2014-10-02T18:39:35Z Blkt:    [Condition of type TYPE-ERROR]
2014-10-02T18:39:36Z Blkt: Restarts:
2014-10-02T18:39:38Z Blkt:  0: [RETRY] Retry SLIME REPL evaluation request.
2014-10-02T18:39:40Z Blkt:  1: [*ABORT] Return to SLIME's top level.
2014-10-02T18:39:42Z Blkt:  2: [ABORT] Abort thread (#<THREAD "repl-thread" RUNNING {1005660083}>)
2014-10-02T18:39:44Z Blkt: Backtrace:
2014-10-02T18:39:46Z nyef: No! No pasting into the channel.
2014-10-02T18:39:46Z Blkt:   0: (MAKE-FLIPS 56 1 #(3 3 3 3 3 3 ...) -1)
2014-10-02T18:39:49Z Blkt:   1: (MAKE-MOVE 56 1 #(3 3 3 3 3 3 ...))
2014-10-02T18:39:51Z Blkt:   2: (OTHELLO #<FUNCTION RANDOM-STRATEGY> #<FUNCTION RANDOM-STRATEGY> T)
2014-10-02T18:39:53Z Blkt:   3: (SB-INT:SIMPLE-EVAL-IN-LEXENV (OTHELLO (FUNCTION RANDOM-STRATEGY) (FUNCTION RANDOM-STRATEGY)) #<NULL-LEXENV>)
2014-10-02T18:39:55Z Blkt:   4: (EVAL (OTHELLO (FUNCTION RANDOM-STRATEGY) (FUNCTION RANDOM-STRATEGY)))
2014-10-02T18:40:00Z Blkt: ok sorry
2014-10-02T18:40:30Z Blkt: http://paste.lisp.org/display/143917#1
2014-10-02T18:42:07Z nyef: Hrm, no inlines...
2014-10-02T18:42:46Z Blkt: what do you mean?
2014-10-02T18:43:15Z nyef: There are no INLINE declarations, which affects the scope of what's going on.
2014-10-02T18:43:56Z nyef: clhs 6.1.2.1.1
2014-10-02T18:43:56Z specbot: The for-as-arithmetic subclause: http://www.lispworks.com/reference/HyperSpec/Body/06_abaa.htm
2014-10-02T18:44:09Z nyef: Have a look at the constraints on the BY preposition.
2014-10-02T18:44:18Z Blkt: I just copied the code from the book
2014-10-02T18:44:26Z nyef: ... So?
2014-10-02T18:44:38Z pkhuong: Blkt: dir must be strictly positive.
2014-10-02T18:45:02Z nyef: Essentially, the code in the book is wrong.
2014-10-02T18:45:15Z Blkt: no, dir should not be strictly positive
2014-10-02T18:45:39Z nyef: Read the CLHS link, it needs to be positive.
2014-10-02T18:45:42Z pkhuong: optimising for speed weakens the type check (while assuming that dir is strictly positive), because you asked for speed.
2014-10-02T18:46:06Z Blkt: pkhuong: yes, I did it exactly for that
2014-10-02T18:46:24Z pkhuong: so that's why you don't get a type error when optimising for speed.
2014-10-02T18:46:32Z pkhuong: the code is still wrong, but silently so.
2014-10-02T18:47:18Z nyef: You might be better off with FOR C = (+ MOVE DIR) THEN (+ C DIR) UNTIL (EQL C BRACKETER).
2014-10-02T18:47:57Z nyef: minion: Advice on compiler?
2014-10-02T18:47:57Z minion: #12000: Looking for a compiler bug is the second-to-last resort. The last resort is blaming bad RAM. It's never the correct hypothesis.
2014-10-02T18:48:20Z Blkt: :D
2014-10-02T18:48:25Z Blkt: I got it now
2014-10-02T18:48:49Z Blkt: thanks
2014-10-02T18:48:56Z nyef: Good luck.
2014-10-02T18:50:16Z Blkt: should I consider it mandatory for a compiler to signal an error on dir under (declaim (optimize safety))?
2014-10-02T18:51:43Z nyef: Depends on how good the compiler is.
2014-10-02T18:52:12Z nyef: And what the rest of the optimization settings are.
2014-10-02T18:56:41Z jsnell: you should not consider it mandatory, the places where an implementation must signal an error for safe code are explicitly marked as such in the clhs
2014-10-02T18:59:27Z slyrus quit (Ping timeout: 245 seconds)
2014-10-02T19:01:10Z Blkt: I see
2014-10-02T19:01:14Z Blkt: thanks
2014-10-02T19:05:48Z slyrus joined #sbcl
2014-10-02T19:29:58Z stassats joined #sbcl
2014-10-02T19:52:28Z slyrus quit (Ping timeout: 244 seconds)
2014-10-02T19:54:32Z slyrus joined #sbcl
2014-10-02T20:01:46Z angavrilov quit (Remote host closed the connection)
2014-10-02T20:45:17Z gingerale quit (Ping timeout: 245 seconds)
2014-10-02T20:53:46Z slyrus_ joined #sbcl
2014-10-02T20:55:32Z slyrus quit (Ping timeout: 250 seconds)
2014-10-02T20:55:36Z slyrus_ is now known as slyrus
2014-10-02T21:03:24Z irsol quit (Ping timeout: 260 seconds)
2014-10-02T21:06:39Z irsol joined #sbcl
2014-10-02T21:15:47Z stassats quit (Ping timeout: 272 seconds)
2014-10-02T21:45:21Z fridim__ quit (Ping timeout: 258 seconds)
2014-10-02T21:51:22Z scymtym_ joined #sbcl
2014-10-02T22:04:19Z slyrus quit (Ping timeout: 272 seconds)
2014-10-02T22:09:34Z Bicyclidine joined #sbcl
2014-10-02T22:14:09Z oleo is now known as Guest6399
2014-10-02T22:15:01Z oleo__ joined #sbcl
2014-10-02T22:17:41Z Guest6399 quit (Ping timeout: 272 seconds)
2014-10-02T22:58:24Z psilord quit (Quit: Leaving.)
2014-10-02T23:12:10Z krzysz00 joined #sbcl
2014-10-02T23:12:25Z krzysz00: Krystof: I've pushed a change that fixes your concerns about Unicode 1 names to my branch
2014-10-02T23:42:16Z oleo__ is now known as oleo
2014-10-02T23:58:52Z karswell` quit (Read error: Connection reset by peer)