00:50:57 -!- vi1 [~vi1@93.92.216.186] has quit [Ping timeout: 268 seconds] 00:52:25 -!- drmeister [~drmeister@mobile-198-228-197-121.mycingular.net] has quit [Remote host closed the connection] 01:15:19 -!- prxq [~mommer@mnhm-590c0e6d.pool.mediaWays.net] has quit [Read error: Operation timed out] 01:15:40 prxq [~mommer@mnhm-5f75c3e4.pool.mediaWays.net] has joined #sbcl 02:39:42 -!- christoph_debian [~christoph@ppp-93-104-161-143.dynamic.mnet-online.de] has quit [Ping timeout: 264 seconds] 02:46:57 drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has joined #sbcl 02:52:46 christoph_debian [~christoph@ppp-93-104-186-190.dynamic.mnet-online.de] has joined #sbcl 03:34:57 -!- drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 03:38:16 -!- redline6561 [~redline65@li69-162.members.linode.com] has quit [Ping timeout: 256 seconds] 03:42:43 redline6561 [~redline65@li69-162.members.linode.com] has joined #sbcl 04:55:02 sdemarre [~serge@91.180.84.159] has joined #sbcl 04:57:36 -!- joshe [~joshe@opal.elsasser.org] has quit [Ping timeout: 260 seconds] 04:58:05 joshe [~joshe@opal.elsasser.org] has joined #sbcl 05:09:09 -!- yacks [~py@180.151.36.168] has quit [Quit: Leaving] 06:03:02 ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has joined #sbcl 06:19:18 -!- Vivitron [~Vivitron@pool-98-110-213-33.bstnma.fios.verizon.net] has quit [Ping timeout: 264 seconds] 06:21:27 -!- prxq [~mommer@mnhm-5f75c3e4.pool.mediaWays.net] has quit [Quit: Leaving] 06:26:26 prxq [~mommer@mnhm-5f75c3e4.pool.mediaWays.net] has joined #sbcl 06:36:31 teggi [~teggi@123.20.107.133] has joined #sbcl 06:45:06 -!- ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has quit [Ping timeout: 264 seconds] 06:55:28 -!- teggi [~teggi@123.20.107.133] has quit [Remote host closed the connection] 07:21:48 Quadrescence [180405b0@gateway/web/freenode/ip.24.4.5.176] has joined #sbcl 07:30:26 Vivitron [~Vivitron@pool-98-110-213-33.bstnma.fios.verizon.net] has joined #sbcl 07:36:28 -!- sdemarre [~serge@91.180.84.159] has quit [Ping timeout: 252 seconds] 07:42:53 -!- wbooze [~wbooze@xdsl-78-35-144-234.netcologne.de] has quit [Ping timeout: 240 seconds] 07:45:39 attila_lendvai [~attila_le@92.46.9.24] has joined #sbcl 07:45:39 -!- attila_lendvai [~attila_le@92.46.9.24] has quit [Changing host] 07:45:39 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 07:47:38 edgar-rft [~GOD@HSI-KBW-149-172-63-75.hsi13.kabel-badenwuerttemberg.de] has joined #sbcl 07:49:43 -!- Bike [~Glossina@67-5-201-64.ptld.qwest.net] has quit [Ping timeout: 264 seconds] 07:57:14 Hello, (pprint '`(do () () ,x)) prints SB-IMPL stuff, which I consider a bug. What do you guys think? 07:57:23 Same with LET. 08:02:55 Odyessus [~odyessus@213.47.71.36] has joined #sbcl 08:05:53 angavrilov [~angavrilo@217.71.227.190] has joined #sbcl 08:14:13 Odyessus_ [~odyessus@089144192152.atnat0001.highway.a1.net] has joined #sbcl 08:14:14 -!- Odyessus_ [~odyessus@089144192152.atnat0001.highway.a1.net] has quit [Client Quit] 08:16:23 -!- Odyessus [~odyessus@213.47.71.36] has quit [Ping timeout: 260 seconds] 08:18:10 -!- Blkt [~user@82.84.173.138] has quit [Read error: Operation timed out] 08:18:17 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.] 08:45:38 -!- ASau` is now known as ASau 09:00:43 -!- foom [jknight@nat/google/x-kbpfcyibqjrpgbhn] has quit [Ping timeout: 260 seconds] 09:01:34 foom [jknight@nat/google/x-zojxhvssgixsjzey] has joined #sbcl 09:02:49 Sagane [~Sagane@177.100-226-89.dsl.completel.net] has joined #sbcl 09:36:51 stassats [~stassats@wikipedia/stassats] has joined #sbcl 09:37:34 benkard [~benkard@mnch-5d856f2e.pool.mediaWays.net] has joined #sbcl 09:38:03 Quadrescence: sure, looks like a bug to me 09:38:40 I hate to ask this since it's volunteer work, but is anyone willing to look into it? I can try, but I (like everyone else probably) am a little busy with other coding things. 09:39:16 is it an actual problem for your work? 09:39:24 yes 09:39:48 i am trying to serialize code by storing it as consed data, not strings 09:40:43 the fastest and surest route is almost certainly fixing it yourself. A good second best might be a launchpad bug report and a patch to the test suite with failing cases 09:42:07 it's generally not portable to store as read data 09:42:27 what with (declare ((integer 0 #.most-positive-fixnum) x)) etc. 09:42:45 store code, that is 09:44:41 stassats: i think a pretty good subset can though, don't you agree? 09:45:34 backqoute can't be used, as you can see 09:45:55 stassats: it can't be used because SBCL is broken 09:46:21 if SBCL was not broken, it would print it back out nicely 09:47:00 no, sbcl isn't broken 09:48:32 stassats: are you suggesting the diff between the printed representation of '`(do () (,x)) and '`(du () (,x)) should be more than just that single character o -> u change? 09:48:44 whitespace ignored 09:49:01 broken doesn't mean "doesn't do what i want it to do" 09:49:34 broken means disagreement with the standard 09:49:42 which is not present here 09:49:54 sorry, i disagree with that definition 09:50:59 the word i use for conforming is "conforming", the word i use for non-conforming is "non-conforming". 09:51:32 that's the only thing you can expect 09:51:32 non-conforming implies broken, but the converse (to me) is not true. 09:52:19 stassats: if that were true we would all just use a simple minimal interpreter and go home 09:52:22 it is suboptimal 09:53:20 Krystof: we could, but it won't be broken, just not very good 09:53:46 I don't think it's that controversial to say that, in sbcl, printing SB-IMPL::BACKQ-COMMA in the output of printing simple backquoted do forms is a bug 09:54:03 and I think that trying to manufacture that controversy is not particularly helpful to anyone 09:54:11 i'm talking about what you can expect from portable code 09:54:31 yes, it'll be good for sbcl to be excellent, but if you're writing portable code, you have to expect anything 09:54:49 when this is fixed, you'll still have to deal with old versions of sbcl, with cmucl, etc. 09:55:15 nevertheless 10:00:42 -!- angavrilov [~angavrilo@217.71.227.190] has quit [Ping timeout: 264 seconds] 10:02:46 -!- benkard [~benkard@mnch-5d856f2e.pool.mediaWays.net] has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz] 10:04:07 benkard [~benkard@mnch-5d856f2e.pool.mediaWays.net] has joined #sbcl 10:04:48 Krystof: I'll start by looking at it 10:12:09 wbooze [~wbooze@xdsl-87-79-252-90.netcologne.de] has joined #sbcl 10:13:18 (funcall (formatter ...)) huh? 10:25:59 stassats: a perhaps old way to optimize format strings into functions? 10:37:01 sdemarre [~serge@91.180.84.159] has joined #sbcl 10:41:48 heh. A four-year-old branch merges with only one trivial conflict 10:42:40 I wonder if I can remember why we gave up on it 10:46:29 foreignFunction [~niksaak@ip-4761.sunline.net.ua] has joined #sbcl 11:01:54 -!- Vivitron [~Vivitron@pool-98-110-213-33.bstnma.fios.verizon.net] has quit [Quit: trivial-irc-0.0.4] 11:25:30 pranavrc [~pranavrc@unaffiliated/pranavrc] has joined #sbcl 11:39:01 -!- Quadrescence [180405b0@gateway/web/freenode/ip.24.4.5.176] has quit [Ping timeout: 250 seconds] 12:01:48 well, Quadrescence is gone, but the fix is at http://paste.lisp.org/display/137611 12:06:34 -!- wbooze [~wbooze@xdsl-87-79-252-90.netcologne.de] has quit [Quit: none] 12:09:46 wbooze [~wbooze@xdsl-87-79-252-90.netcologne.de] has joined #sbcl 12:15:27 -!- benkard [~benkard@mnch-5d856f2e.pool.mediaWays.net] has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz] 12:35:07 drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has joined #sbcl 12:38:41 benkard [~benkard@mnch-5d856f2e.pool.mediaWays.net] has joined #sbcl 12:55:41 -!- pranavrc [~pranavrc@unaffiliated/pranavrc] has quit [Quit: Ping timeout: ] 13:05:08 -!- drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 13:13:26 -!- wbooze [~wbooze@xdsl-87-79-252-90.netcologne.de] has quit [Quit: none] 13:13:47 wbooze [~wbooze@xdsl-87-79-252-90.netcologne.de] has joined #sbcl 13:23:07 i love it when the debugger goes # # # # # # # # 13:25:32 drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has joined #sbcl 13:32:28 gah, what do you mean "failure-p was set compiling src/compiler/node.lisp" stupid cross-compiler 13:33:29 -!- drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 13:35:15 -!- sdemarre [~serge@91.180.84.159] has quit [Ping timeout: 256 seconds] 13:37:31 *stassats* finds some code with triple backqoute 13:39:08 and on which my backqoute printer fails 13:40:06 the old printer didn't work that great on it either 13:41:41 '```(,,,x ,y) is printed as ``,,,X 13:41:58 funnily, that only happens with triple quotes, not with double 13:47:45 2: (CSUBTYPEP # #) 13:47:45 2: CSUBTYPEP returned NIL T 13:47:58 oh boy 13:49:26 unknown-type sset 13:49:52 I mean, I could ask "why why why?" but I actually know why probably 13:50:18 I think the moral here is that 10 years' experience with sbcl's type system still doesn't convert this into an hour's hack :-( 13:52:51 though, curses, maybe I don't know why 14:12:14 (still not quite sure why but I can make it go away with liberal application of exclamation marks) 14:23:00 -!- Sagane [~Sagane@177.100-226-89.dsl.completel.net] has quit [Read error: Connection reset by peer] 14:28:16 TRACE prints the arguments simply, but pretty prints the results 14:38:23 Krystof: build on !sbcl and see where it chokes? 14:44:30 pkhuong: no, I've just added a maybe-reparse-specifier! where there was none before 14:45:06 it is odd that I need it on this branch -- I can't see anything that would push things down a different code path -- but the maybe-reparse-specifier! is not wrong 14:46:32 the cl-test-grid discussion will continue; should continue CCing sbcl-devel? 14:46:59 *should i 14:47:00 hlavaty`` [~user@friedrichstrasse.knowledgetools.de] has joined #sbcl 14:48:51 -!- hlavaty` [~user@friedrichstrasse.knowledgetools.de] has quit [Ping timeout: 260 seconds] 14:51:41 -!- wbooze [~wbooze@xdsl-87-79-252-90.netcologne.de] has quit [Read error: Connection reset by peer] 14:52:16 I'm interested, but I can just go to google groups. 14:53:15 pkhuong: ok 14:54:53 fisxoj [~fisxoj@c-24-12-190-29.hsd1.il.comcast.net] has joined #sbcl 15:23:37 Bike [~Glossina@67-5-201-64.ptld.qwest.net] has joined #sbcl 15:35:08 '``('',y ,,x) is pprinted as ``(,,'`'',Y ,,X) 15:35:13 such a joy 15:35:55 wbooze [~wbooze@xdsl-87-79-252-90.netcologne.de] has joined #sbcl 15:36:01 haha 15:36:15 well, at least it got the ,,x right 15:37:59 writing code which uses triple backqoutes is a pretty tough exercise, imagine writing code to pprint them 15:38:27 *stassats* gets one case right, another one breaks 15:40:10 drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has joined #sbcl 15:50:09 hm, so my mental implementation founders on (at least) one subtlety 15:50:21 array types? 15:50:24 what is the union of (vector (eql 2)) and (vector (eql 3))? 15:51:08 neither choice makes sense without co/contra variance information 15:51:33 ah but the spec has made some choices 15:51:42 I'm not sure yet whether they are all the right choices 15:51:47 oh, yeah. with upgraded element types, it works 15:51:53 ah but wait 15:52:13 (or (vector (eql 2)) (vector (eql 3)))? 15:52:17 I was thinking you were trying to propagate declared element types through 15:52:30 when you say (declare (type (vector (eql 2)) x)), you say that both x is a (vector (mod 4)) and that (aref x 0) is (eql 2) 15:52:44 so, it gets more fun 15:53:03 but for our purposes, only the first part is essential; the second is a bonus, right? 15:53:21 if you say (declare (type (or (vector (eql 2)) (vector (eql 3))) x)), you say that both x is a (vector (mod 4)), twice, and that (aref x 0) is (eql 2) or (eql 3) 15:53:23 so here's the question 15:53:42 if you make that declaration, and discover that (aref x 0) is 2 15:53:47 what does that say about (aref x 1)? 15:53:50 you also say that you can store a 2 or a 3 in vector 15:57:21 upgraded element types can be handled with a finite set representation. declared element types, aren't as nice, but I don't know that complicated declared element types are used in the wild. 15:59:12 pragmatist! 16:01:05 I agree, and if I make the simplifying assumption that (or (vector (eql 2)) (vector (eql 3))) is equivalent to (vector (or (eql 2) (eql 3))) [ because they have the same UAET, and because as you say the rest is a bunch of THE forms which we can pretend aren't there in any case ] then I can carry on being merry 16:01:52 I'd nuke the declared element type on join ;) 16:02:21 no fun at all 16:09:24 -!- drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 16:10:55 as long you do the same for intersections... 16:11:29 that is (and (vector (eql 2)) (vector (eql 3))) => (vector (or (eql 2) (eql 3))). 16:12:41 unless there's a hidden gem in the spec that says we can assume arrays are immutable when playing with expressed element types. 16:17:07 doesn't the intersection actually hit the empty type? 16:17:21 if we assume arrays are immutable. 16:17:21 that is, any access must be of an (and (eql 2) (eql 3)) 16:17:36 yes, but what does it say about writes? 16:17:55 any legal write in the lexical scope of the declaration would have to be of an (and 2 3) 16:18:02 It says I can write 2 into it, and I can write 3 into it. 16:23:39 '``('',y ,,x `(,a)) is printed as ``(,,'`'',Y ,,X ,','`(,A)), and '```('',y ,,,x) as ``,,','`'',Y 16:23:42 *stassats* cries 16:24:50 a backquote pretty printing fuzzer would be interesting. 16:24:59 "let's stuff `,' at random" 16:25:19 (i meant that not for a fuzzer, but describing how the printer works) 16:25:54 pkhuong: I don't think it does. Any access to the array must be of the appropriate type 16:25:54 just a check that (print (read-from-string (print X))) doesn't error and is the same as (print X) would already be nice. 16:26:40 "each access were surrounded by an appropriate /the/ form" is what CLHS says 16:26:43 i just need to come up with a general recursive algorithm, otherwise it's a game of whack-a-mole 16:26:58 Krystof: ok. so we're assuming read-only arrays. nice. 16:27:08 wait. "appropriate" what's that? 16:27:24 so (setf (aref x 0) y) => (the (and (eql 2) (eql 3)) (setf (aref x 0) y)) I think 16:27:43 this is clhs type 16:28:17 if I have function A that works with a (vector X), and B with (vector Y), should (and (vector X) (vector Y)) be acceptable for both A and B? 16:29:34 Sagane [~Sagane@177.100-226-89.dsl.completel.net] has joined #sbcl 16:29:39 AND? 16:30:23 yes. OR would be appropriate for one of them, but not necessarily both. 16:33:31 davazp [~user@165.Red-79-144-63.dynamicIP.rima-tde.net] has joined #sbcl 16:36:43 perhaps the reader itself should be changed to produce more regular structures for backqoutes 16:37:07 Odyessus [~odyessus@chello080109062130.15.14.vie.surfer.at] has joined #sbcl 16:37:31 yes. I think many of the reader-side optimisations for backquote can be removed now that the compiler is smarter about things. 16:38:04 clisp seems to do well on backqoute printing 16:38:11 but i can't look at its source code 16:38:29 yes. relatedly, it also prints (let ((function 42)) ...) as (let (#'42) ...) 16:45:18 -!- Bike [~Glossina@67-5-201-64.ptld.qwest.net] has quit [Quit: abandoning elliott, wish me luck~] 16:45:42 what about translating directly to backqoute, backq-comma, backq-comma-at, etc? 16:47:04 '`(,y `(,x)) => (SB-IMPL::BACKQ-CONS Y (QUOTE ((SB-IMPL::BACKQ-LIST X)))), i don't really see how you can reconstruct the original 16:47:22 and that's printed as `(,Y ,'`(,X)) 16:48:25 because '`(,Y ,'`(,X)) translates to (SB-IMPL::BACKQ-CONS Y (QUOTE ((SB-IMPL::BACKQ-LIST X)))) as well 16:51:46 WFM. I think we want to avoid atrocious performance regressions, but, like I said, ISTM most of the reader side cleverness is subsumed by what we now do in the compiler. 16:55:40 -!- Sagane [~Sagane@177.100-226-89.dsl.completel.net] has quit [Read error: Connection reset by peer] 16:56:27 LiamH [~none@pool-173-73-123-20.washdc.east.verizon.net] has joined #sbcl 17:06:23 -!- fisxoj [~fisxoj@c-24-12-190-29.hsd1.il.comcast.net] has quit [Ping timeout: 245 seconds] 17:11:37 -!- xymox [lechuck@unaffiliated/contempt] has quit [Quit: Lost terminal] 17:15:33 xymox [lechuck@unaffiliated/contempt] has joined #sbcl 17:22:17 attila_lendvai [~attila_le@92.46.11.104] has joined #sbcl 17:22:17 -!- attila_lendvai [~attila_le@92.46.11.104] has quit [Changing host] 17:22:17 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 17:24:48 *stassats* finally convinced to go the reader-macro changing route 17:24:58 printing just can't be done with the current scheme 17:27:04 and that's a task for another day 17:27:07 -!- Odyessus [~odyessus@chello080109062130.15.14.vie.surfer.at] has quit [Quit: Colloquy for iPad - http://colloquy.mobi] 17:31:53 -!- davazp [~user@165.Red-79-144-63.dynamicIP.rima-tde.net] has quit [Ping timeout: 240 seconds] 17:41:14 drmeister [~drmeister@pool-71-185-168-200.phlapa.fios.verizon.net] has joined #sbcl 17:41:48 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Read error: Connection reset by peer] 17:41:54 attila_lendvai [~attila_le@92.46.11.104] has joined #sbcl 17:41:55 -!- attila_lendvai [~attila_le@92.46.11.104] has quit [Changing host] 17:41:55 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 17:42:49 leuler [~user@p548FA47D.dip0.t-ipconnect.de] has joined #sbcl 17:46:13 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Ping timeout: 248 seconds] 17:54:34 attila_lendvai [~attila_le@92.46.11.104] has joined #sbcl 17:54:34 -!- attila_lendvai [~attila_le@92.46.11.104] has quit [Changing host] 17:54:34 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 17:55:06 pkhuong: but expressed array type declarations are lexical in scope 17:55:08 anyway 17:55:13 I have concluded that no-one knows :-) 18:01:51 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.] 18:20:21 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 248 seconds] 18:21:25 -!- benkard [~benkard@mnch-5d856f2e.pool.mediaWays.net] has quit [Ping timeout: 256 seconds] 18:27:13 benkard [~benkard@mnch-5d856f2e.pool.mediaWays.net] has joined #sbcl 18:30:39 yacks [~py@180.151.36.168] has joined #sbcl 18:32:46 ASau` [~user@p4FF969EC.dip0.t-ipconnect.de] has joined #sbcl 18:36:58 -!- ASau [~user@p5797EB0A.dip0.t-ipconnect.de] has quit [Ping timeout: 276 seconds] 18:43:08 Quadrescence [180405b0@gateway/web/freenode/ip.24.4.5.176] has joined #sbcl 18:44:14 robgssp [~user@cpe-24-93-28-218.rochester.res.rr.com] has joined #sbcl 18:44:21 Krystof: s/function/piece of code/ 18:47:14 oh, I see. If one piece of code is contained within the other, then yes; if they are sequential, then it's more like the union 18:47:17 that's horrible 18:48:09 stassats wrote a patch for the backq printing, cool! let's try it out 18:48:10 Quadrescence, memo from Bike: as a last ditch measure you can just blindly subst sb-impl::backq-cons with cons etc. It won't print with pretty backquotes but it'll be portable code, and those functions never appear anywhere else so a codewalker is overkill. 18:48:12 -!- ASau` is now known as ASau 18:49:30 Krystof: mutability and subtyping make an awful combo. 18:50:39 I wonder if the answer is to have these special set-of-representation types only for those arrays whose expressed and actual element types are the same 18:55:44 SBCL committers: if Stassats' patch seems to be okay, is there any reason why it should not be sent upstream? http://paste.lisp.org/display/137611 18:55:51 look at logs 18:56:03 stassats has been discovering all sorts of horribleness 18:56:06 Sorry, I don't have IRC logs :( 18:56:22 they are publically available; google for "sbcl irc logs" 18:56:22 tunes and clozure do. 18:56:37 ok, i will look 19:05:50 Krystof: (sb-c::specifier-type `(and (vector (member t 0 1)) (vector (member nil 0 1)))) 19:05:56 => # 19:06:00 :| 19:08:00 It should be a T vector with I don't know what expressed element type. But definitely not a bit vector. 19:10:31 -!- benkard [~benkard@mnch-5d856f2e.pool.mediaWays.net] has quit [Ping timeout: 256 seconds] 19:10:51 Also, interesting discussion on array types. 19:10:58 My argument is something like (defun a (vector) (declare (type (vector (member t 0 1)) vector)) ...) (defun b (vector) (declare (type (vector (member nil 0 1)) vector)) ...) 19:12:14 (defun c (vector) (declare (type (and (vector (member t 0 1)) (vector (member nil 0 1))) vector)) (a vector) (b vector)) should work. 19:13:28 benkard [~benkard@mnch-5d856f2e.pool.mediaWays.net] has joined #sbcl 19:14:31 pkhuong: should that only work because the "UAET" of those two vectors is (probably) T? 19:14:35 Clearly, if the two vector types don't have the same UAET, the intersection is NIL. Otherwise, the intersection has the same UAET. 19:15:07 s/is (probably)/should (probably) be/ 19:16:13 pkhuong: that's definitely wrong 19:16:14 However, what the expressed element type isn't clear to me. If we assume that A and B only read from the vector, then intersecting the expressed element types works. if they only write, then taking the union works. If they do both, nothing works. 19:16:31 I mean, the specifier-type of the intersection is definitely wrong 19:16:36 right 19:17:35 on the plus side, this means that I don't have to get exactly the Right Answer first time round 19:17:45 given that what we have now is definitely incorrect 19:18:18 There's a similar issue on union. (defun c (vector switch) (declare (type (or ...) vector)) (if switch (a vector) (b vector))), and assume that switch and vector are correctly associated. 19:19:01 or no, maybe unions are actually fine. 19:28:05 Krystof: oh, i think it's just the printing logic that fails here. 19:30:51 yeah, type-specifier is bugged, but I don't know how the intersection can be unparsed sanely. 19:31:57 hm, that's a fair point actually 19:32:39 -!- Quadrescence [180405b0@gateway/web/freenode/ip.24.4.5.176] has quit [Ping timeout: 250 seconds] 19:39:32 -!- prxq [~mommer@mnhm-5f75c3e4.pool.mediaWays.net] has quit [Remote host closed the connection] 19:40:41 -!- benkard [~benkard@mnch-5d856f2e.pool.mediaWays.net] has quit [Ping timeout: 268 seconds] 19:41:57 prxq [~mommer@mnhm-5f75c3e4.pool.mediaWays.net] has joined #sbcl 19:41:58 -!- LiamH [~none@pool-173-73-123-20.washdc.east.verizon.net] has quit [Quit: Leaving.] 19:42:54 benkard [~benkard@mnch-5d8548d0.pool.mediaWays.net] has joined #sbcl 19:52:53 -!- wbooze [~wbooze@xdsl-87-79-252-90.netcologne.de] has quit [Ping timeout: 240 seconds] 20:03:43 wbooze [~wbooze@xdsl-78-35-144-170.netcologne.de] has joined #sbcl 20:16:03 slyrus [~chatzilla@adsl-108-80-229-231.dsl.pltn13.sbcglobal.net] has joined #sbcl 20:21:43 -!- leuler [~user@p548FA47D.dip0.t-ipconnect.de] has quit [Quit: ERC Version 5.1.2 $Revision: 1.796.2.6 $ (IRC client for Emacs)] 20:58:10 -!- benkard [~benkard@mnch-5d8548d0.pool.mediaWays.net] has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz] 21:37:15 Hydan [~hydan@ip-89-103-110-5.net.upcbroadband.cz] has joined #sbcl 21:37:19 -!- prxq [~mommer@mnhm-5f75c3e4.pool.mediaWays.net] has quit [Remote host closed the connection] 21:39:35 -!- slyrus [~chatzilla@adsl-108-80-229-231.dsl.pltn13.sbcglobal.net] has quit [Ping timeout: 240 seconds] 21:46:48 benkard [~benkard@mnch-5d8548d0.pool.mediaWays.net] has joined #sbcl 21:49:07 slyrus [~chatzilla@adsl-108-80-229-231.dsl.pltn13.sbcglobal.net] has joined #sbcl 21:51:37 LiamH [~none@pool-173-73-123-20.washdc.east.verizon.net] has joined #sbcl 21:59:11 Vivitron [~Vivitron@pool-98-110-213-33.bstnma.fios.verizon.net] has joined #sbcl 21:59:53 -!- benkard [~benkard@mnch-5d8548d0.pool.mediaWays.net] has quit [Read error: Connection reset by peer] 22:06:13 benkard [~benkard@mnch-5d8548d0.pool.mediaWays.net] has joined #sbcl 22:09:42 -!- wbooze [~wbooze@xdsl-78-35-144-170.netcologne.de] has quit [Ping timeout: 264 seconds] 22:39:02 -!- benkard [~benkard@mnch-5d8548d0.pool.mediaWays.net] has quit [Quit: Textual IRC Client: www.textualapp.com] 23:18:45 wbooze [~wbooze@xdsl-78-35-144-170.netcologne.de] has joined #sbcl 23:48:57 -!- wbooze [~wbooze@xdsl-78-35-144-170.netcologne.de] has quit [Quit: none] 23:49:15 wbooze [~wbooze@xdsl-78-35-144-170.netcologne.de] has joined #sbcl