00:15:26 -!- wbooze [~wbooze@xdsl-78-35-160-98.netcologne.de] has quit [Ping timeout: 245 seconds] 00:19:31 do you think the draft at http://www.techfak.uni-bielefeld.de/~jmoringe/sbcl.html#Extensible-Sequences is going in the right direction? if not, what (major) changes should I make? 00:19:33 this is 00:19:34 lp 994528 00:19:34 https://bugs.launchpad.net/bugs/994528 00:19:36 btw. 01:30:29 zulu_inuoe [~quassel@184.89.111.53] has joined #sbcl 01:30:51 -!- zulu_inuoe [~quassel@184.89.111.53] has quit [Read error: Connection reset by peer] 01:32:05 zulu_inuoe [~quassel@184.89.111.53] has joined #sbcl 01:48:27 yacks [~py@180.151.36.168] has joined #sbcl 02:02:06 -!- edgar-rft [~GOD@HSI-KBW-149-172-63-75.hsi13.kabel-badenwuerttemberg.de] has quit [Quit: bleeding] 02:50:13 -!- easye [~user@213.33.70.157] has quit [Remote host closed the connection] 02:53:19 easye [~user@213.33.70.157] has joined #sbcl 03:05:34 -!- jsnell [~jsnell@ash.snellman.net] has quit [Ping timeout: 240 seconds] 03:06:30 jsnell [~jsnell@ash.snellman.net] has joined #sbcl 03:48:14 edgar-rft [~GOD@HSI-KBW-149-172-63-75.hsi13.kabel-badenwuerttemberg.de] has joined #sbcl 03:54:54 ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has joined #sbcl 03:55:19 -!- drmeister [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 04:03:38 -!- jsnell [~jsnell@ash.snellman.net] has quit [Ping timeout: 252 seconds] 04:03:45 jsnell [~jsnell@ash.snellman.net] has joined #sbcl 04:05:25 -!- danlentz [~danlentz@c-68-37-70-235.hsd1.nj.comcast.net] has quit [Ping timeout: 248 seconds] 04:12:18 Strigoides [~owen@60-234-213-126.bitstream.orcon.net.nz] has joined #sbcl 04:26:49 pranavrc [~pranavrc@122.164.71.164] has joined #sbcl 04:26:50 -!- pranavrc [~pranavrc@122.164.71.164] has quit [Changing host] 04:26:50 pranavrc [~pranavrc@unaffiliated/pranavrc] has joined #sbcl 04:52:05 sdemarre [~serge@175.87-64-87.adsl-dyn.isp.belgacom.be] has joined #sbcl 05:00:54 attila_lendvai [~attila_le@92.47.252.144] has joined #sbcl 05:00:54 -!- attila_lendvai [~attila_le@92.47.252.144] has quit [Changing host] 05:00:54 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 05:03:43 -!- yacks [~py@180.151.36.168] has quit [Read error: Operation timed out] 05:05:31 prxq [~mommer@mnhm-590c0240.pool.mediaWays.net] has joined #sbcl 05:12:16 danlentz [~danlentz@c-68-37-70-235.hsd1.nj.comcast.net] has joined #sbcl 05:15:30 Bike_ [~Glossina@67-5-250-200.ptld.qwest.net] has joined #sbcl 05:17:53 -!- Bike [~Glossina@67-5-250-200.ptld.qwest.net] has quit [Ping timeout: 240 seconds] 05:18:17 -!- Bike_ is now known as Bike 05:26:04 -!- sdemarre [~serge@175.87-64-87.adsl-dyn.isp.belgacom.be] has quit [Ping timeout: 256 seconds] 05:31:24 -!- Bike [~Glossina@67-5-250-200.ptld.qwest.net] has quit [Ping timeout: 264 seconds] 05:34:06 Bike [~Glossina@67-5-250-200.ptld.qwest.net] has joined #sbcl 05:39:32 yacks [~py@180.151.36.168] has joined #sbcl 05:40:34 Bike_ [~Glossina@67-5-250-200.ptld.qwest.net] has joined #sbcl 05:43:39 -!- Bike [~Glossina@67-5-250-200.ptld.qwest.net] has quit [Ping timeout: 256 seconds] 05:43:50 -!- Bike_ is now known as Bike 05:47:07 -!- psilord [~psilord@c-69-180-173-249.hsd1.mn.comcast.net] has quit [Quit: Leaving.] 05:49:22 echo-area [~user@182.92.247.2] has joined #sbcl 06:29:09 tcr1 [~tcr@46-126-110-164.dynamic.hispeed.ch] has joined #sbcl 06:51:11 -!- easye [~user@213.33.70.157] has quit [Ping timeout: 256 seconds] 06:56:44 easye [~user@213.33.70.157] has joined #sbcl 07:00:31 -!- Munksgaard [~philip@80-71-135-72.u.parknet.dk] has quit [Ping timeout: 264 seconds] 07:01:59 Munksgaard [~philip@80-71-135-72.u.parknet.dk] has joined #sbcl 07:15:38 davazp [~user@178.167.155.218.threembb.ie] has joined #sbcl 07:28:01 -!- ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has quit [Read error: Connection reset by peer] 07:28:31 ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has joined #sbcl 07:31:07 -!- easye [~user@213.33.70.157] has quit [Remote host closed the connection] 07:34:54 easye [~user@213.33.70.157] has joined #sbcl 07:48:47 -!- davazp [~user@178.167.155.218.threembb.ie] has quit [Ping timeout: 245 seconds] 07:49:34 -!- easye [~user@213.33.70.157] has quit [Ping timeout: 256 seconds] 07:53:35 -!- tcr1 [~tcr@46-126-110-164.dynamic.hispeed.ch] has quit [Quit: Leaving.] 07:54:22 -!- ehaliewicz [~user@50-0-51-11.dsl.static.sonic.net] has left #sbcl 07:56:04 -!- pranavrc [~pranavrc@unaffiliated/pranavrc] has quit [Remote host closed the connection] 07:56:11 easye [~user@213.33.70.157] has joined #sbcl 07:58:53 -!- Bike [~Glossina@67-5-250-200.ptld.qwest.net] has quit [Ping timeout: 240 seconds] 08:01:29 davazp [~user@92.251.203.229.threembb.ie] has joined #sbcl 08:26:00 stassats [~stassats@wikipedia/stassats] has joined #sbcl 08:27:52 kanru` [~kanru@31.4.212.249] has joined #sbcl 08:47:23 angavrilov [~angavrilo@217.71.227.190] has joined #sbcl 08:47:57 foreignFunction [~niksaak@94.27.88.206] has joined #sbcl 08:55:53 -!- easye [~user@213.33.70.157] has quit [Ping timeout: 256 seconds] 08:56:55 -!- kanru` [~kanru@31.4.212.249] has quit [Ping timeout: 264 seconds] 09:00:42 easye [~user@213.33.70.157] has joined #sbcl 09:27:01 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 248 seconds] 09:33:01 stassats [~stassats@wikipedia/stassats] has joined #sbcl 09:35:58 currently feeling: stupid. 09:38:24 if only I had a thorough grounding in theoretical computer science 09:40:13 -!- easye [~user@213.33.70.157] has quit [Remote host closed the connection] 09:40:16 I don't think that exists ;) 09:40:25 I have: a graph, fairly densely but not fully-connected. I want: to assign integers to nodes such that (a) all sums of node integers connected by edges in the graph are distinct (b) the maximal such sum is minimal 09:40:52 (actually I don't have such a graph but I think I can probably cope with constructing one) 09:41:02 eesh, that feels NP 09:41:09 NP is fine 09:41:27 -!- scymtym_ [~user@ip-5-147-116-166.unitymediagroup.de] has quit [Ping timeout: 276 seconds] 09:41:31 I am also fine with a reasonable heuristic for (b) 09:41:55 okay, so um... by sums you mean sums of two connected integers? or sums of the entire connected subgraphs? 09:42:01 of two connected integers 09:42:46 easye [~user@213.33.70.157] has joined #sbcl 09:43:18 in case it helps to see where I'm coming from, or perhaps to reformulate the problem: what I'm trying to do is assign indices in a lookup table to go from pairs of unicode code points to their primary composition 09:43:28 e.g. #\e + #\acute -> #\é 09:43:35 could it maybe be phrased as a graph-coloring problem? 09:43:50 it sounds kind of graph-coloring-esque 09:43:55 the graph is two-coloured 09:44:07 Krystof: so, why sums? Couldn't it be any hash function? 09:44:18 and the links always join nodes of opposite colours 09:44:22 Krystof: The Unicode tables already contains that information 09:44:37 loke: this is an attempt to create a highly compact encoding of the unicode table 09:44:40 two-colorede... make red nodes '1', make blue nodes even numbers that increment starting at 0? 09:44:48 would that guarantee uniqueness? 09:45:06 oh... no it wouldn't, since a blue node can connect to multiple red ones. 09:45:22 pkhuong: yes, it can be any hash function 09:45:47 Krystof: what is the use case? 09:45:56 easy-iPad [~easyipad@chello080109062130.15.14.vie.surfer.at] has joined #sbcl 09:46:11 implementing the function which performs Unicode primary composition 09:46:57 Krystof: Ah, implementing Unicode features to SBCL? Thank you! :-) 09:47:13 Krystof: How is ICU doing it? 09:47:15 Krystof: I'd try something like a static cuckoo hash table. 09:47:59 loke: that's cheating! I am trying to prove to myself that I can do something myself :-) (well, with this channel's help ;-) 09:48:32 -!- easy-iPad is now known as strideeer 09:49:02 Krystof: Not really cheating. After all, ICU is written in C, so it's not like you can just steal their solution. 09:49:31 well, for me the solution is the design. Implementing it is a job for minions 09:49:35 *Krystof* is his own minion 09:49:46 pkhuong: OK, that is something for me to look at. Thanks 09:49:50 *thiiinks* assign 0 to the first blue node, assign 1..N to the red nodes it neighbors. then find the next blue node, assign N+1 to it, and assign 1..M to its neighbors, where M is max(number of neighbors, highest already-tagged neighbor red node)? 09:49:58 Krystof: working on this as part of gsoc? 09:50:15 *loke* has longed for good Unicode support 09:50:52 Krystof: how were you planning on mapping code points to their weights? 09:51:10 *Fiora* hasn't done graph theory in ageeeeeessssss 09:52:21 -!- easye [~user@213.33.70.157] has quit [Remote host closed the connection] 09:53:05 Fiora: so I strongly suspect that is a long way from minimal 09:53:19 that's basically equivalent to having one array per blue node of size N 09:53:31 yeah... especially if there's like, widely far apart sections of the graph that aren't connected 09:53:51 pkhuong: I have plenty space in the character database (one memory lookup) 09:54:04 easye [~user@213.33.70.157] has joined #sbcl 09:54:06 sorry, two memory lookups 09:54:30 what I was hoping to avoid (and cuckoo hashing doesn't seem to let me) was to store the characters or equivalent in the lookup table for collision resolution 09:54:32 Krystof: ok, so looking up two random values per codepoint is doable. 09:55:13 I don't have infinite space, but I do probably have 16 bits or so. The total number of entries in the table is ~1000 09:55:36 Krystof: why is space limited? 09:55:40 because 09:56:08 I mean I should probably stop worrying about it and just use the 70*457 words to encode 1000 words of data, but... 09:56:27 Krystof: Yes you should. 09:56:29 :-) 09:56:46 but every so often leuler or froydnj come along and triumphantly shave 10k off core sizes 09:57:03 so I feel bad about adding 100k to store 4k's worth of information :/ 09:57:43 wait, there are only 1k entries... How many pairs are there for each codepoint? 09:57:49 I saw on sbcl-devel that CR had a branch that implements normalisation. Did you talk to him also? 09:57:58 loke: Krystof is CR. 09:58:04 Oh 09:58:17 -!- foreignFunction [~niksaak@94.27.88.206] has quit [Quit: Leaving.] 09:58:23 Good to know 09:58:32 foreignFunction [~niksaak@94.27.88.206] has joined #sbcl 09:58:40 So is the real name Krystof or Christophe? 09:59:30 which one doesn't fit into the RFC limit for IRC nicks? 10:00:06 pkhuong: right, that is a potential simplification. My guess is that latin is the densest bit -- you can put an acute accent on more or less anything 10:00:17 as I say, I haven't actually constructed the graph yet :) 10:00:27 It's been many years (a decade and a half) since I implemented the IRC protocol. I don't remember the limits. :-) 10:00:48 it's 9-character 10:00:57 or at least it was when I discovered IRC 10:01:58 easy-iPad [~easyipad@089144192152.atnat0001.highway.a1.net] has joined #sbcl 10:04:26 -!- strideeer [~easyipad@chello080109062130.15.14.vie.surfer.at] has quit [Ping timeout: 252 seconds] 10:04:46 Krystof: You could have a sequence of [second code point, composition] for each codepoint. Concatenate in a single vector, and remember where the codepoint's sequence begins 10:06:09 -!- foreignFunction [~niksaak@94.27.88.206] has quit [Quit: Leaving.] 10:06:52 foreignFunction [~niksaak@94.27.88.206] has joined #sbcl 10:07:36 linear search for the second codepoint? 10:08:27 yup. 10:08:31 or bsearch. 10:11:24 hm, means knowing where the sequence ends too 10:11:42 it ends where the next codepoint's sequence begins... 10:11:56 true, but for a given lookup I don't have the "next" codepoint 10:12:26 anyway. I have been given some ideas. Maybe I should construct the graph and see what the simplest solution looks like 10:12:35 ah, ok. 10:12:39 (or maybe I should do some Real Work) 10:13:24 We don't have a table of data indexed by char code... 10:14:36 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 245 seconds] 10:18:45 pkhuong: wait, we do 10:18:54 or what do you mean? 10:19:12 drmeister [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has joined #sbcl 10:20:20 Krystof: well then, there's your next codepoint. Increment the codepoint, look where its sequence begins. 10:20:55 -!- easye [~user@213.33.70.157] has quit [Ping timeout: 256 seconds] 10:21:23 ah, unfortunately most codepoints don't take part in composition at all 10:22:06 *p_l* meanwhile looked at very, very fast internal<->utf8 conversions 10:22:18 most sequence will begin and end at the same index ;) .. or concatenate [sequence length] [second codepoint, composition]*. 10:23:09 yeah, that would be sensible 10:23:14 ok, will think 10:26:04 easye [~user@213.33.70.157] has joined #sbcl 10:26:56 -!- scymtym [~user@2001:638:504:2093:226:b9ff:fe7d:3e1f] has quit [Remote host closed the connection] 10:28:38 scymtym [~user@2001:638:504:2093:226:b9ff:fe7d:3e1f] has joined #sbcl 10:34:04 pranavrc [~pranavrc@122.164.10.98] has joined #sbcl 10:34:07 -!- pranavrc [~pranavrc@122.164.10.98] has quit [Changing host] 10:34:07 pranavrc [~pranavrc@unaffiliated/pranavrc] has joined #sbcl 10:37:40 -!- easye [~user@213.33.70.157] has quit [Remote host closed the connection] 10:39:36 easye [~user@213.33.70.157] has joined #sbcl 10:55:59 hi 11:02:48 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Read error: Operation timed out] 11:04:27 easy-iPad_ [~easyipad@chello080109062130.15.14.vie.surfer.at] has joined #sbcl 11:04:39 -!- easy-iPad [~easyipad@089144192152.atnat0001.highway.a1.net] has quit [Ping timeout: 276 seconds] 11:04:39 -!- easy-iPad_ is now known as easy-iPad 11:09:06 -!- easye [~user@213.33.70.157] has quit [Ping timeout: 256 seconds] 11:12:38 -!- easy-iPad [~easyipad@chello080109062130.15.14.vie.surfer.at] has quit [Quit: Outta here?] 11:17:49 attila_lendvai [~attila_le@92.46.3.91] has joined #sbcl 11:17:49 -!- attila_lendvai [~attila_le@92.46.3.91] has quit [Changing host] 11:17:49 attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has joined #sbcl 11:23:05 easye [~user@213.33.70.157] has joined #sbcl 11:25:35 HATE 11:25:44 that assertion that I made, that the graph was 2-coloured? 11:25:46 HA HA HA 11:26:06 oh dear :< 11:26:22 COMBINING GREEK DIALYTIKA TONOS, what misbegotten planet do you come from? 11:28:13 oh, but wait! COMBINING GREEK DIALYTIKA TONOS is in the composition exclusions! Yay! 11:29:51 ok, 2-colouring is back. 11:48:15 kanru` [~kanru@106.Red-83-46-193.dynamicIP.rima-tde.net] has joined #sbcl 12:26:26 -!- drmeister [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 12:29:13 -!- zulu_inuoe [~quassel@184.89.111.53] has quit [Remote host closed the connection] 12:29:56 drmeister [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has joined #sbcl 12:40:33 -!- drmeister [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 12:45:15 Odyessus [~odyessus@089144192063.atnat0001.highway.a1.net] has joined #sbcl 12:47:13 -!- kanru` [~kanru@106.Red-83-46-193.dynamicIP.rima-tde.net] has quit [Ping timeout: 245 seconds] 12:51:07 -!- easye [~user@213.33.70.157] has quit [Ping timeout: 256 seconds] 12:51:20 kanru` [~kanru@106.Red-83-46-193.dynamicIP.rima-tde.net] has joined #sbcl 12:52:58 segv- [~mb@95-91-241-83-dynip.superkabel.de] has joined #sbcl 12:57:24 easye [~user@213.33.70.157] has joined #sbcl 12:58:33 scymtym_ [~user@ubi-1-232-140.dhcp.uni-bielefeld.de] has joined #sbcl 13:00:35 -!- Odyessus [~odyessus@089144192063.atnat0001.highway.a1.net] has quit [Quit: Colloquy for iPad - http://colloquy.mobi] 13:04:09 -!- echo-area [~user@182.92.247.2] has quit [Remote host closed the connection] 13:22:27 tsuru`` [~charlie@adsl-98-87-25-181.bna.bellsouth.net] has joined #sbcl 13:24:02 -!- tsuru` [~charlie@adsl-98-87-25-187.bna.bellsouth.net] has quit [Ping timeout: 252 seconds] 13:27:13 -!- kanru` [~kanru@106.Red-83-46-193.dynamicIP.rima-tde.net] has quit [Ping timeout: 245 seconds] 13:32:56 kanru` [~kanru@106.Red-83-46-193.dynamicIP.rima-tde.net] has joined #sbcl 13:39:55 drmeister [~drmeister@farnsworth.chem.temple.edu] has joined #sbcl 13:42:15 nyef [~nyef@c-76-119-183-159.hsd1.ma.comcast.net] has joined #sbcl 13:48:06 psilord [~psilord@23-25-144-217-static.hfc.comcastbusiness.net] has joined #sbcl 14:00:13 wbooze [~wbooze@xdsl-84-44-152-119.netcologne.de] has joined #sbcl 14:14:18 -!- kanru` [~kanru@106.Red-83-46-193.dynamicIP.rima-tde.net] has quit [Ping timeout: 245 seconds] 14:18:31 -!- Munksgaard [~philip@80-71-135-72.u.parknet.dk] has quit [Ping timeout: 264 seconds] 14:46:38 tsuru``` [~charlie@adsl-98-87-44-21.bna.bellsouth.net] has joined #sbcl 14:47:19 I have a macro roughly of the form (defmacro with-foo ((&rest foos) &body body) (let ((*foos* (list* ,@foos *foos*))) (declare (dynamic-extent *foos*)) ,@body)). Invoking a function that uses it conses. Why? 14:48:13 list! 14:48:14 -!- tsuru`` [~charlie@adsl-98-87-25-181.bna.bellsouth.net] has quit [Ping timeout: 252 seconds] 14:49:05 *FEATURES* contains :stack-allocatable-lists or whatever it's called. 14:50:42 why you want to copy the global into the local one ? 14:51:15 What local one? 14:51:29 nyef: works here (although, I don't think that declaration is right, given a strict reading of the spec) 14:51:41 Hrm... Really? 14:52:29 So, might well be platform specific, then... 14:53:20 Okay, next stop is probably a trace-file. Thanks. 14:56:48 Oh, and what would be the "right" way to do this? 15:04:27 -!- tsuru``` is now known as tsuru` 15:09:39 -!- segv- [~mb@95-91-241-83-dynip.superkabel.de] has quit [Remote host closed the connection] 15:14:16 Damn. A disassembly clearly shows heap allocation for the LIST* call. 15:14:42 yes, list/list* is supposed to cons freshly.... 15:15:18 and you create a lexical var of the same name like the global one and copy therein, i don't get what you want..... 15:15:56 I want to have a new binding on the global variable, with the new CONS cells allocated on the stack. 15:15:57 you cons anew just to throw all away ? 15:15:59 kanru` [~kanru@201.Red-88-12-18.staticIP.rima-tde.net] has joined #sbcl 15:16:06 Over the duration of the body. 15:16:58 The macro behaves "correctly", but I'd vastly prefer that it didn't cause consing on the heap. 15:17:35 -!- wbooze [~wbooze@xdsl-84-44-152-119.netcologne.de] has quit [Remote host closed the connection] 15:18:20 wbooze [~wbooze@xdsl-84-44-152-119.netcologne.de] has joined #sbcl 15:19:06 have you read that here http://www.sbcl.org/manual/Dynamic_002dextent-allocation.html 15:19:07 -!- scymtym_ [~user@ubi-1-232-140.dhcp.uni-bielefeld.de] has quit [Ping timeout: 264 seconds] 15:19:35 -!- bege [~bege@S0106001d7e5132b0.ed.shawcable.net] has quit [Quit: leaving] 15:19:54 <|3b|> nyef: none of the FOOS can escape the body? 15:19:58 the line cons, list, list*, and vector when the result is bound to a variable declared dynamic-extent. 15:20:15 you bind it before the declaration.... 15:20:20 not after it 15:20:36 |3b|: Right. They just have to be live over the extent of the body. 15:21:03 wbooze: It's a "bound declaration", isn't it? 15:21:10 *|3b|* was asking more about the possibility than the need 15:21:40 It's possible, if someone is doing something stupid. 15:21:53 when is declarations taking effect ? 15:22:02 I'll come clean, this is the PPC/SPARC gencgc version of WITH-PINNED-OBJECTS. 15:22:53 maybe you need really-truly-dynamic-extent? 15:23:00 ... Maybe. 15:23:09 It's been far too long since I've messed with the d-x logic. 15:23:39 I wish I could remember why we decided we needed truly-dynamic-extent as well as dynamic-extent 15:24:04 Probably because there's a switch in sb-ext to turn off dynamic-extent at compile-time? 15:24:17 yeah, but why? 15:24:51 maybe it was that all naive uses of dynamic-extent were in fact wrong, but I can't remember why. Something to do with transitive reachability? 15:25:10 at least partially because we only do d-x on functions at low safety levels? 15:25:39 Krystof: right... which is why I think the declaration is actually incorrect. 15:26:14 Building now with t-d-x. Maybe that'll work. 15:27:29 does the declaration actually imply that everything that was in *foos* before is also unreachable at the end? 15:27:36 this rings a vague bell 15:27:38 <|3b|> not the way i read it 15:27:50 i would put the decl before the let 15:27:53 <|3b|> previous *foos* is not otherwise inaccessible 15:28:00 within a progn or so..... 15:28:23 |3b|: how else is it accessible? 15:28:40 wbooze: you are not helping. 15:28:51 sorry 15:29:51 <|3b|> pkhuong: i'd hope the special var binding/unbinding stuff keeps the old value around somewhere 15:30:15 so maybe (let ((new-foos (list ,@stuff))) (declare (dynamic-extent new-foos)) (setf (cdr (last new-foos) *foos*)) (let ((*foos* new-foos)) ...))? 15:30:30 |3b|: how are those accessible *when *foos* is bound to the result of list* *? 15:30:40 <|3b|> though it is sort of confusing in that it isn't /currently/ accessible but might become accessible again at some point in the future 15:30:47 pkhuong: Close over the old value, surely? 15:31:00 nyef: did you? :) 15:31:35 No. I just made sure that it was on the list for the new value of *PINNED-OBJECTS*. It's not like anyone other than the GC needs to grovel over it, anyway. 15:31:39 my understanding is you did already cons.....decl too late.....(and even misleading maybe....) 15:31:52 you are not understanding correctly 15:32:00 <|3b|> wbooze: you need to read more closely, particularly about bound declarations as mentioned earlier 15:32:08 ok then explain me pleas what happens there 15:32:26 wbooze: Seriously. Have a look for "bound declaration" in CLHS. 15:32:44 this is like d-x on &rest lists, which is never an issue, to the extent that it's an example for d-x 15:33:17 No, an escaping d-x &rest list is bad mojo, but that's not what's at issue here. 15:34:37 nyef: that's not what I'm referring to. It's that a really clever compiler could exploit d-x on &rest to dxify values in the &rest list. 15:34:50 Ah. 15:34:52 Eesh. 15:35:00 -!- cic_ [~connolly@Catnip.AI.SRI.COM] has quit [Ping timeout: 240 seconds] 15:35:02 easy-iPad [~easyipad@chello080109062130.15.14.vie.surfer.at] has joined #sbcl 15:35:19 It'd have to be doing some sort of interprocedural analysis, though... 15:35:28 or inlining. 15:35:41 I think that that might count. 15:36:28 So, WITH-GENSYMS or similar, set up local bindings for each of the objects to pin, THEN do the dx-allocation of the pin list? 15:36:44 cic_ [~connolly@Catnip.AI.SRI.COM] has joined #sbcl 15:37:24 wouldn't it be slightly simpler to have DX vectors with a next slot? 15:38:08 More complicated code on the GC side, and I had enough context in my head at the time that I really just wanted something simple. 15:38:12 k. 15:38:28 <|3b|> pkhuong: best i can come up with is that old value of *foos* is accessible from the old binding of *foos* at the point where the new binding is created, even if it isn't while the new binding is in scope, so it isn't otherwise inaccessible (assuming some other first element in the list* call) 15:38:29 I mean, you remember the patch series for PPC threading, right? 15:38:38 nyef: sure :) 15:39:41 Making the runtime actually somewhat close to thread-safe took quite a bit of doing, plus sorting out the GC requirements, plus finding bugs in the GC, plus... Well, yeah... 15:41:04 wtf 15:41:15 what was not so clear about bound declarations ? 15:41:16 <|3b|> though i guess even if it is otherwise inaccessible, it is still an otherwise inaccessible part of the current value 15:41:20 can't you all read or what ? 15:42:09 "within the special form that establishes the variable or function, but before the body of that special form...." 15:42:34 Right, and the initial value is NIL, and the only bindings are from this macro. 15:42:43 "specifically at the head of that forms body" 15:43:00 wbooze: Are you trying to say that the LET special form doesn't establish *FOOS*? 15:43:33 wbooze: what's more likely -- pkhuong, nyef, |3b| and I all can't read, or there's something that you've not read carefully enough? 15:43:43 it does, but the decl is hinging after it..... 15:43:45 you get ? 15:43:56 should come before.... 15:44:36 The *BODY* is the first non-declaration form *AFTER* the binding specifiers. 15:44:36 <|3b|> wbooze: 'within the special form' ... ' but before the body' 15:44:39 -!- easy-iPad [~easyipad@chello080109062130.15.14.vie.surfer.at] has quit [Quit: Outta here?] 15:46:14 read on the last part 15:46:26 if it were lexical rather then dynamic..... 15:48:18 you form is a let form and it's body is your (list* ....thing..... 15:48:47 <|3b|> no, the body was ,@body 15:49:07 <|3b|> (list* ...) was a init-form 15:49:18 hmmmmm 15:52:12 so you mean the compiler already changes the order accordingly ? 15:52:20 when reading that out ? 15:52:28 or does it rather fail ? 15:52:59 <|3b|> compiler reads declarations as part of evaluating special forms (and some macros) and applies them to the bindings created by that special form 15:53:11 <|3b|> s/compiler/evaluator/ 15:54:06 easy-iPad [~easyipad@chello080109062130.15.14.vie.surfer.at] has joined #sbcl 15:55:08 so not all macros ? and which would that be then ? 15:55:12 <|3b|> (also, none of this is sbcl specific, so if you need further clarification, #lisp might be a better place for it) 15:55:23 what's the criteria for some there ? 15:55:29 <|3b|> the ones specified to allow declarations 15:55:31 ok 15:55:46 <|3b|> things like DEFUN 15:57:13 hmmm, init-form should consist rather of constants then ? 15:59:37 maybe cons does not fall into const-folding then ? 16:05:01 -!- easy-iPad [~easyipad@chello080109062130.15.14.vie.surfer.at] has quit [Quit: Outta here?] 16:06:39 easy-iPad [~easyipad@chello080109062130.15.14.vie.surfer.at] has joined #sbcl 16:13:15 Posterdati [~antani@host163-225-dynamic.10-87-r.retail.telecomitalia.it] has joined #sbcl 16:16:04 -!- easy-iPad [~easyipad@chello080109062130.15.14.vie.surfer.at] has quit [Quit: Outta here?] 16:25:04 When using with-pinned-objects, and some of the object forms have side-effects, are those side-effects supposed to occur on all platforms, or only on gencgc platforms? 16:27:59 Bike [~Glossina@67-5-250-200.ptld.qwest.net] has joined #sbcl 16:29:54 Hrm. TRULY-D-X doesn't seem to be helping. 16:30:20 Yeah, still seeing references to boxed_region. 16:33:19 -!- Bike [~Glossina@67-5-250-200.ptld.qwest.net] has quit [Ping timeout: 245 seconds] 16:33:57 Bike [~Glossina@67-5-250-200.ptld.qwest.net] has joined #sbcl 16:37:27 -!- pranavrc [~pranavrc@unaffiliated/pranavrc] has quit [Quit: Ping timeout: ] 16:39:52 Bike_ [~Glossina@67-5-250-200.ptld.qwest.net] has joined #sbcl 16:41:34 rpg [~rpg@23-25-144-220-static.hfc.comcastbusiness.net] has joined #sbcl 16:42:42 -!- Bike [~Glossina@67-5-250-200.ptld.qwest.net] has quit [Ping timeout: 264 seconds] 16:43:20 -!- Bike_ is now known as Bike 16:48:50 -!- rpg [~rpg@23-25-144-220-static.hfc.comcastbusiness.net] has quit [Remote host closed the connection] 16:48:58 rpg [~rpg@23-25-144-217-static.hfc.comcastbusiness.net] has joined #sbcl 16:50:32 -!- kanru` [~kanru@201.Red-88-12-18.staticIP.rima-tde.net] has quit [Ping timeout: 272 seconds] 16:54:40 LiamH [~none@pdp8.nrl.navy.mil] has joined #sbcl 17:00:04 side effects in the with-pinned-objects forms? That seems poor. :P 17:01:15 Yes, it'd be poor form, but what's the actual correct thing to do with them? 17:11:30 -!- davazp [~user@92.251.203.229.threembb.ie] has quit [Ping timeout: 264 seconds] 17:13:30 definitely call them. 17:30:16 Can we dx-allocate specialized vectors to the number stack? 17:59:13 sdemarre [~serge@109.134.133.167] has joined #sbcl 18:49:46 davazp [~user@178.167.210.13.threembb.ie] has joined #sbcl 19:00:31 -!- attila_lendvai [~attila_le@unaffiliated/attila-lendvai/x-3126965] has quit [Quit: Leaving.] 19:07:27 stassats [~stassats@wikipedia/stassats] has joined #sbcl 19:10:29 Krystof: do you per chance overwrite lvar-use with a constant, in the modarith thing? 19:12:17 Munksgaard [~philip@0803ds2-bi.0.fullrate.dk] has joined #sbcl 19:14:03 Krystof: srctran.lisp:3063, set lvar-%derived-type to NIL instead of (make-values-type ...). Does that sound right? 19:16:00 -!- LiamH [~none@pdp8.nrl.navy.mil] has quit [Ping timeout: 256 seconds] 19:17:34 yes, good idea! 19:17:55 definitely worth a try anyway... 19:17:56 -!- psilord [~psilord@23-25-144-217-static.hfc.comcastbusiness.net] has quit [Ping timeout: 256 seconds] 19:19:20 it works here. 19:19:33 I'm sending patch for consideration to the ML 19:20:02 LiamH [~none@pdp8.nrl.navy.mil] has joined #sbcl 19:21:40 -!- easye [~user@213.33.70.157] has quit [Ping timeout: 256 seconds] 19:23:56 -!- foreignFunction [~niksaak@94.27.88.206] has quit [Quit: Leaving.] 19:30:13 -!- sdemarre [~serge@109.134.133.167] has quit [Ping timeout: 258 seconds] 19:32:55 -!- angavrilov [~angavrilo@217.71.227.190] has quit [Ping timeout: 264 seconds] 19:54:58 // Running /home/csr21/src/lisp/sbcl/tests/unicode-normalization.impure.lisp 19:55:05 All tests succeeded 19:55:46 (I'm not admitting how many tests are disabled. But I'm now fairly confident about NFD and NFKD) 20:06:16 -!- davazp [~user@178.167.210.13.threembb.ie] has quit [Ping timeout: 245 seconds] 20:11:31 easye [~user@213.33.70.157] has joined #sbcl 20:13:00 -!- Posterdati [~antani@host163-225-dynamic.10-87-r.retail.telecomitalia.it] has quit [Quit: KVIrc 4.1.3 Equilibrium http://www.kvirc.net/] 20:14:21 foreignFunction [~niksaak@94.27.88.232] has joined #sbcl 20:15:43 psilord [~psilord@c-69-180-173-249.hsd1.mn.comcast.net] has joined #sbcl 20:22:18 davazp [~user@92.251.204.173.threembb.ie] has joined #sbcl 20:37:23 -!- rpg [~rpg@23-25-144-217-static.hfc.comcastbusiness.net] has quit [Quit: rpg] 20:39:11 -!- Munksgaard [~philip@0803ds2-bi.0.fullrate.dk] has quit [Ping timeout: 255 seconds] 20:55:10 Posterdati [~antani@host163-225-dynamic.10-87-r.retail.telecomitalia.it] has joined #sbcl 21:00:18 -!- easye [~user@213.33.70.157] has quit [Ping timeout: 257 seconds] 21:03:26 easye [~user@213.33.70.157] has joined #sbcl 21:14:58 -!- nyef [~nyef@c-76-119-183-159.hsd1.ma.comcast.net] has quit [Quit: G'night all.] 21:17:21 tsuru`` [~charlie@adsl-74-179-31-86.bna.bellsouth.net] has joined #sbcl 21:18:42 -!- drmeister [~drmeister@farnsworth.chem.temple.edu] has quit [Remote host closed the connection] 21:19:24 -!- tsuru` [~charlie@adsl-98-87-44-21.bna.bellsouth.net] has quit [Ping timeout: 264 seconds] 21:23:21 hi, does anybody know if sbcl records in the fasl a reference to the original source file, and in that case, how to get it at load-time? 21:23:50 davazp: what are you trying to do? 21:24:14 stassats: I am fixing https://bugs.launchpad.net/sbcl/+bug/504492 21:24:54 I have managed to store xref information of toplevel forms, but they are anonymous, so I have not a source-location, just a toplevel form number 21:25:38 I would like to complete it at load-time with the original pathname 21:28:29 specifically, I would need to access to the original from a fop 21:29:54 -!- foreignFunction [~niksaak@94.27.88.232] has quit [Quit: Leaving.] 21:34:30 -!- stassats [~stassats@wikipedia/stassats] has quit [Ping timeout: 256 seconds] 21:35:53 -!- Bike [~Glossina@67-5-250-200.ptld.qwest.net] has quit [Ping timeout: 255 seconds] 21:36:37 -!- prxq [~mommer@mnhm-590c0240.pool.mediaWays.net] has quit [Quit: Leaving] 21:37:55 Bike [~Glossina@67-5-194-226.ptld.qwest.net] has joined #sbcl 21:43:48 Munksgaard [~philip@80-71-135-72.u.parknet.dk] has joined #sbcl 22:06:10 -!- davazp [~user@92.251.204.173.threembb.ie] has quit [Ping timeout: 246 seconds] 22:08:22 drmeister [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has joined #sbcl 22:20:42 -!- easye [~user@213.33.70.157] has quit [Remote host closed the connection] 22:24:49 easye [~user@213.33.70.157] has joined #sbcl 22:40:43 -!- Munksgaard [~philip@80-71-135-72.u.parknet.dk] has quit [Read error: Connection reset by peer] 22:41:53 -!- drmeister [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 22:46:37 davazp [~user@92.251.204.173.threembb.ie] has joined #sbcl 22:52:54 drmeister [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has joined #sbcl 22:58:19 -!- drmeister [~drmeister@pool-173-59-25-70.phlapa.fios.verizon.net] has quit [Remote host closed the connection] 22:58:45 -!- davazp [~user@92.251.204.173.threembb.ie] has quit [Remote host closed the connection] 23:08:43 drmeister [~drmeister@166.137.107.202] has joined #sbcl 23:18:49 -!- drmeister [~drmeister@166.137.107.202] has quit [Remote host closed the connection] 23:26:30 -!- LiamH [~none@pdp8.nrl.navy.mil] has quit [Quit: Leaving.] 23:38:03 drmeister [~drmeister@166.137.107.202] has joined #sbcl 23:42:12 -!- wbooze [~wbooze@xdsl-84-44-152-119.netcologne.de] has quit [Ping timeout: 272 seconds] 23:45:40 -!- drmeister [~drmeister@166.137.107.202] has quit [Remote host closed the connection] 23:47:53 ASau`` [~user@p5797F361.dip0.t-ipconnect.de] has joined #sbcl 23:51:12 -!- ASau` [~user@p5797F5CF.dip.t-dialin.net] has quit [Ping timeout: 264 seconds]