2017-06-16T00:02:28Z Merv_ quit (Ping timeout: 240 seconds)
2017-06-16T00:20:17Z whiteline_ is now known as whiteline
2017-06-16T00:50:38Z stassats quit (Ping timeout: 260 seconds)
2017-06-16T00:59:01Z Merv_ joined #sbcl
2017-06-16T01:01:29Z DGASAU quit (Read error: Connection reset by peer)
2017-06-16T01:02:13Z cromachina joined #sbcl
2017-06-16T01:33:35Z Merv_ quit (Ping timeout: 240 seconds)
2017-06-16T02:29:38Z Merv_ joined #sbcl
2017-06-16T02:51:05Z DGASAU joined #sbcl
2017-06-16T03:03:05Z Merv_ quit (Ping timeout: 240 seconds)
2017-06-16T03:03:24Z whiteline_ joined #sbcl
2017-06-16T03:06:58Z whiteline quit (Ping timeout: 264 seconds)
2017-06-16T03:31:56Z whiteline_ is now known as whiteline
2017-06-16T03:33:10Z Bicyclidine quit (Ping timeout: 240 seconds)
2017-06-16T03:59:53Z Merv_ joined #sbcl
2017-06-16T04:10:53Z karswell` joined #sbcl
2017-06-16T04:34:16Z Merv_ quit (Ping timeout: 255 seconds)
2017-06-16T05:39:49Z angavrilov joined #sbcl
2017-06-16T05:53:43Z shka_ joined #sbcl
2017-06-16T05:59:08Z gingerale joined #sbcl
2017-06-16T05:59:11Z slyrus_ quit (Ping timeout: 246 seconds)
2017-06-16T06:01:55Z oleo quit (Quit: irc client terminated!)
2017-06-16T06:10:10Z slyrus_ joined #sbcl
2017-06-16T06:26:06Z shka_ quit (Quit: Konversation terminated!)
2017-06-16T06:31:43Z shka joined #sbcl
2017-06-16T06:46:33Z slyrus_ quit (Read error: Connection reset by peer)
2017-06-16T06:46:58Z salv0 joined #sbcl
2017-06-16T06:47:08Z slyrus_ joined #sbcl
2017-06-16T06:59:28Z Merv_ joined #sbcl
2017-06-16T07:03:38Z Merv__ joined #sbcl
2017-06-16T07:06:35Z Merv_ quit (Ping timeout: 240 seconds)
2017-06-16T08:07:32Z scymtym joined #sbcl
2017-06-16T08:58:05Z attila_lendvai joined #sbcl
2017-06-16T09:03:52Z attila_lendvai quit (Ping timeout: 255 seconds)
2017-06-16T10:16:13Z m00natic joined #sbcl
2017-06-16T10:16:47Z karswell` quit (Ping timeout: 246 seconds)
2017-06-16T10:25:54Z Bike joined #sbcl
2017-06-16T11:06:08Z shka quit (Ping timeout: 240 seconds)
2017-06-16T12:15:05Z Bike quit (Ping timeout: 240 seconds)
2017-06-16T12:17:27Z slyrus_ quit (Ping timeout: 240 seconds)
2017-06-16T12:18:51Z slyrus quit (Ping timeout: 268 seconds)
2017-06-16T12:22:55Z rumbler31 joined #sbcl
2017-06-16T12:48:53Z oleo joined #sbcl
2017-06-16T13:00:50Z Bike joined #sbcl
2017-06-16T13:03:50Z rumbler31 quit (Remote host closed the connection)
2017-06-16T13:14:10Z gingerale quit (Read error: Connection reset by peer)
2017-06-16T13:16:00Z gingerale joined #sbcl
2017-06-16T13:28:21Z milanj joined #sbcl
2017-06-16T13:45:07Z slyrus joined #sbcl
2017-06-16T13:55:09Z cromachina quit (Read error: Connection reset by peer)
2017-06-16T14:01:52Z slyrus: this is perhaps less helpful than it could be, but I've dropped into ldb while compiling mcclim and all it's dependencies a couple of times in the last day or so. don't remember this being so frequent before.
2017-06-16T14:04:10Z slyrus: and of course it's never repeatable...
2017-06-16T14:04:24Z rumbler31 joined #sbcl
2017-06-16T14:04:29Z slyrus: s/it's/its/
2017-06-16T14:08:57Z rumbler31 quit (Ping timeout: 240 seconds)
2017-06-16T14:13:11Z slyrus_ joined #sbcl
2017-06-16T14:16:38Z oleo quit (Read error: Connection reset by peer)
2017-06-16T14:29:20Z oleo joined #sbcl
2017-06-16T14:30:28Z dougk_: nyef`: how is rebinding gensym-counter around forms any different from around the file as a whole? the symbols are guaranteed fresh
2017-06-16T14:33:31Z dougk_: and I dislike seeing #:cleanup-fun-478932 when it conveys no more information than #:cleanup-fun-77 other than possibly that it came from a later form in the file
2017-06-16T14:47:23Z nyef`: dougk_: IMPORT them. They're fresh, they have a particular (otherwise unused) prefix, and are numbered in an ascending sequence: Clearly, they cannot collide until someone monkeys the gensym counter.
2017-06-16T14:48:34Z nyef`: Now, if this is just for the XC, that's fine, but if it's applied to user code, it's wrong.
2017-06-16T14:51:27Z nyef`: As far as the cleanup-fun symbols go, you could make it (format nil "CLEANUP-FUN-~A" <source-path-for-unwind-protect>), at which point there's some actual information there.
2017-06-16T14:59:31Z fe[nl]ix: nyef`: the purpose of the gensym counter is to make the macroexpansion readable for humans
2017-06-16T15:00:06Z fe[nl]ix: the symbols are already unique even because they're uninterned, even if they had the same name
2017-06-16T15:00:59Z nyef`: No, the purpose of *GENSYM-COUNTER* is to give the programmer control over the numeric aspect of the behavior of GENSYM.
2017-06-16T15:01:20Z nyef`: As soon as you IMPORT the result of GENSYM, you introduce the possibility of name collisions.
2017-06-16T15:01:45Z nyef`: And this is *allowed*.
2017-06-16T15:01:47Z gko quit (Quit: ZNC - http://znc.in)
2017-06-16T15:24:48Z gko joined #sbcl
2017-06-16T16:09:45Z nyef`: Essentially, resetting (or rebinding) *GENSYM-COUNTER* at the TLF level or at the COMPILE-FILE level is not indicated by the specification *at all*, destroys most-to-all of the utility of *GENSYM-COUNTER*, and certainly destroys the rationale for introducing it.
2017-06-16T16:20:42Z dougk_: nyef`: i disagree for a variety of reasons. (1) the side effect of gensym is that it "might" be that it increments gensym-counter, and (2) http://www.lispworks.com/documentation/HyperSpec/Issues/iss182_w.htm says that it may be rebound at any time
2017-06-16T16:21:30Z gingerale quit (Ping timeout: 240 seconds)
2017-06-16T16:23:09Z gingerale joined #sbcl
2017-06-16T16:29:41Z nyef`: dougk_: (1) The circumstances when GENSYM increments the counter are well-defined. (2) It may be rebound *by the user* at any time, not by the implementation.
2017-06-16T16:30:56Z nyef`: Additionally, this behavioral change requires anyone who wants the sane-and-specified behavior to reimplement GENSYM for themselves.
2017-06-16T16:47:23Z slyrus quit (Ping timeout: 246 seconds)
2017-06-16T16:51:00Z dougk_: nyef`: i'll revert to per-file binding. You'll have to argue that with xof - https://sourceforge.net/p/sbcl/sbcl/ci/08d05510b5 did that. Moreover, i'll make sure that the compiler (both xc and normal) never use GENSYM per se, but instead use a version that is per-tlf rebound to 0.
2017-06-16T16:52:31Z dougk_: in other words, to it's more preposterous that compiling the same file twice should be nondeterministic with respect to dumped debug symbols
2017-06-16T16:53:48Z m00natic quit (Remote host closed the connection)
2017-06-16T17:03:13Z fe[nl]ix: nyef`: ccl has had this behavior for a long time and nobody complained about it
2017-06-16T17:03:30Z fe[nl]ix: so I get you don't like it, but it's not against the standard and it's not harmful
2017-06-16T17:18:10Z slyrus_ quit (Ping timeout: 240 seconds)
2017-06-16T17:24:11Z slyrus joined #sbcl
2017-06-16T17:24:39Z nyef`: I disagree with your interpretation of the spec, and I'm not convinced that the difference is harmless, but I can see that I'm in the minority here.
2017-06-16T17:48:41Z Merv__ quit (Remote host closed the connection)
2017-06-16T17:51:23Z nyef`: The thought occurs that it is necessarily the responsibility of the build system to rebind *GENSYM-COUNTER* around COMPILE-FILE in order to have builds be "repeatable" in this sense. And while this is not necessarily an ideal circumstance, it *is* what is required by the specification.
2017-06-16T18:21:25Z slyrus_ joined #sbcl
2017-06-16T18:27:57Z Posterdati quit (Ping timeout: 245 seconds)
2017-06-16T18:38:31Z jcloud is now known as jyc
2017-06-16T18:45:20Z slyrus_ quit (Ping timeout: 246 seconds)
2017-06-16T19:12:25Z aeth_ joined #sbcl
2017-06-16T19:13:08Z aeth quit (Ping timeout: 260 seconds)
2017-06-16T19:13:38Z aeth_ is now known as aeth
2017-06-16T19:38:47Z rumbler31 joined #sbcl
2017-06-16T19:45:08Z milanj quit (Quit: This computer has gone to sleep)
2017-06-16T19:47:59Z salv0 quit (Ping timeout: 246 seconds)
2017-06-16T19:57:47Z scymtym quit (Ping timeout: 246 seconds)
2017-06-16T19:57:56Z salv0 joined #sbcl
2017-06-16T20:06:15Z Posterdati joined #sbcl
2017-06-16T20:07:04Z milanj joined #sbcl
2017-06-16T20:54:56Z minion quit (Disconnected by services)
2017-06-16T20:55:08Z minion joined #sbcl
2017-06-16T20:56:10Z specbot quit (Disconnected by services)
2017-06-16T20:56:22Z specbot joined #sbcl
2017-06-16T20:56:25Z specbot quit (Disconnected by services)
2017-06-16T20:56:34Z gingerale quit (Remote host closed the connection)
2017-06-16T20:56:42Z specbot joined #sbcl
2017-06-16T21:06:28Z Merv_ joined #sbcl
2017-06-16T21:15:20Z slyrus_ joined #sbcl
2017-06-16T21:32:32Z Bike quit (Ping timeout: 255 seconds)
2017-06-16T21:41:05Z Merv_ quit (Ping timeout: 255 seconds)
2017-06-16T21:43:07Z angavrilov quit (Remote host closed the connection)
2017-06-16T22:01:54Z salv0 quit (Remote host closed the connection)
2017-06-16T22:14:47Z Bike joined #sbcl
2017-06-16T22:21:50Z slyrus_ quit (Ping timeout: 240 seconds)
2017-06-16T22:37:06Z Merv_ joined #sbcl
2017-06-16T23:11:54Z Merv_ quit (Ping timeout: 268 seconds)