freenode/#clasp - IRC Chatlog
Search
20:14:25
DVSSA
(where I found build-cando.asd and setup-cando.asd) but when I try start a clasp image and run (ql:quickload "build-cando") I get nothing
20:29:38
DVSSA
So I can now load (within slime & jupyter) (ql:quickload "build-cando") and it seems to work.
20:34:29
DVSSA
I can also load "build-clasp-jupyter", from what I can see the differences between the two is when i run (list-all-packages) the dif between the two is #<CANDO-USER> #<CLEXT.QUEUE> #<NGLV> #<PYTHREAD>
20:35:07
DVSSA
But when I try run an example I got from drmeister's docker images I get the following
20:39:14
Colleen
attila_lendvai: drmeister said 7 hours, 10 minutes ago: Thank you - I'll be able to take a look at it tomorrow.
20:42:10
DVSSA
attila_lendvai: I tried that but I'm not sure that is what it's supposed to be, the code is a copy-paste from drmeisters docker and he uses it like that
20:42:43
attila_lendvai
DVSSA: probably in his context it didn't make a difference, but it does in your package
20:43:30
attila_lendvai
DVSSA: you seem to be new to lisp, right? read up on packages, http://www.gigamonkeys.com/book/programming-in-the-large-packages-and-symbols.html
20:44:20
attila_lendvai
what happens is that symbols are read in the current package. a package is a collection of symbols, possibly imported from other packages. basically a namespace...
20:44:56
attila_lendvai
unprefixed symbols are read in the current package -- that is. if you prefix it, then the current package is irrelevant
21:00:14
Bike
probably drmeister's code is in some other package. maybe there's in-package in the files somewhere, dunno.
22:00:20
Bike
drmeister: a hinta bout what "The dispatch function has two selectors with different outcomes" would be helpful, because i'm pretty clueless
22:13:52
drmeister
two selectors with different outcomes is bad - I think it means the generic function dispatch is building the dispatch tree and it got to a point where it was going to add an outcome and there was one already there.
22:15:37
drmeister
But I think it means that the call history has two entries with the same key but different outcomes (really bad) - or the same outcome (and then the two entries are only redundant - but still bad). The keys in the call history should always be unique.
22:21:08
drmeister
While I was on the plane I started working on getting Maxima to work in Clasp - wow - it goes to town on symbol properties.
23:09:30
Bike
the underlying issue is probably that i messed up whatever magical process gives boot classes stamps
23:09:30
Colleen
Bike: drmeister said 51 minutes, 38 seconds ago: - I just put some stuff in the log for you.
23:19:24
Bike
the last problem i fixed as that nothing in lisp actually initialized the specializer call history slot
23:25:57
drmeister
I’m at some life sciences dinner - there’s 1300 people here and they are making speeches and giving out awards.