freenode/#lisp - IRC Chatlog
Search
4:11:54
rpg
Does anyone have a Jenkins recipe for working with quicklisp? I.e., having a module that is checked out of a repository, but goes into a quicklisp/local-projects/ for jenkins?
4:12:56
Fare
rpg, my apologies for merging #89 by accident :-( https://gitlab.common-lisp.net/asdf/asdf/merge_requests/89
4:13:53
rpg
Fare: No problem -- it's just run through my Jenkins tests and it seems fine. I'll take a more in-depth look, but I think the syntax control needs more concentration right now, now that I'm back from my travels
4:15:12
Fare
And it only affects upgrade --- and whoever besides me would have been deep enough to extend bundle classes, which I think is no one (or there would have been a lot of complaints as I kept refactoring it over the years)
4:15:49
Fare
on the other hand, extending the same fix to other more basic operation classes might be harder, if anyone used them (which I bet at least someone would).
4:31:12
rpg
bah! I'm going to have a glass of wine... Once more getting my bloody tests to work right is harder than getting the code under test to work right -- for reasons totally unrelated to the code under test.
6:29:47
Zhivago
Gerbil looks nice, but unexceptional, at a glance -- what motivated the new dialect?
7:24:31
lieven
at a guess, some asian language with different pronouns for people older or younger than you :)
7:26:09
lieven
Thai seems similar. Some of my Thai acquaintances want to find out a person's age when they first meet someone.
7:27:23
beach
It is an interesting situation of negotiation. Things are a bit simpler for me because I am both older than most people I deal with, and I have a higher rank.
8:02:05
beach
ông: grandfather, old man, gentleman, you (used by grandchild to grandfather), I (used by grandfather to grandchild, you (used to men), he (of men over 30), husband, man, Mr.
8:05:16
loke`
beach: Hey! I spend my time on social media. I see no traces of such fancy languageisms there.
8:05:55
Zhivago
loke: Well, you have similar rules regarding the use of words like "mister" and "sir".
8:24:40
Ober
Error: Last argument to apply is too long:. uff. must use reduce and not apply for 6k items
8:28:20
aeth
clisp has a call-arguments-limit of 4096. Every other implementation that I've checked is *much* higher than that, at least 60k. SBCL's is practically infinite (you'll run out of memory first)
8:30:38
aeth
Someone needs to compile a list of the various actual limits and sizes because the minimums in the CLHS are much lower than any implementation, but some implementations still have surprisingly low numbers
8:32:10
aeth
jackdaniel: apply could be used in serialization because you can apply a plist, which would make it very convenient for serialization
8:35:36
aeth
very useful.. (apply #'make-array 42 '(:element-type 'single-float :initial-element 42f0))
8:35:40
jackdaniel
it might be that after nuclear disaster, 2000 years from now, people will try to recreate machines capable of running program of our age
8:36:31
aeth
jackdaniel: Someone should make a revised CLHS and only bump up the minimums for things. Or, actually, define separate 32-bit and 64-bit minimums. Iirc, a fixnum is only 16 bit as a minimum, which is tiny.
8:40:09
Ober
what is the proper way to replace if seen progn with cond? (cond ((seen ... in this case sees it as a function
8:46:17
jackdaniel
each clause cond is (test forms*), where test is arbitrary expression returning value
8:47:42
Ober
ok so under cond a single or double open paren before it should not confuse it. ok. must be error. lisp-critic cleanup
9:07:01
jasom
You can have 31 bit fixnums on a 32 bit machine; if you require all allocations be 8-bit aligned, that still gives you two tag bits for non-fixnums
13:35:19
myrkraverk
Is my assumption correct, that :below (foo) is only eval'd once? In (:loop :for bar :from 0 :below (foo) ... )
13:38:07
myrkraverk
I tried "simplifying" my code, and got rid of a variable from (sb-posix:read ...)
13:38:40
myrkraverk
And while I'm doing SBCL specfic stuff, I try not to let my guesses and tests rule the world.
13:39:24
specbot
The for-as-arithmetic subclause: http://www.lispworks.com/reference/HyperSpec/Body/06_abaa.htm
13:46:47
myrkraverk
(loop :for i :below (sb-posix:read fd (sb-sys:vector-sap buffer) (length buffer)) ... )
13:50:21
myrkraverk
Later, I have :with v := 10 ... and then :finally (return v) ; do I need the (return v) or is there another way to specify the return value of LOOP?
13:51:40
mfiano
certain clauses imply a return value, such as the last `collect`. You can either do it as you are now, or wrap it in a let variable.
13:54:57
beach
myrkraverk: No need to force it if it doesn't fit your needs. I was just answering your question.
15:30:36
warweasle
Can someone point me to a simple cl-autowrap project? Where they create the json files and then use them in their package.
15:36:57
jmercouris
I'm trying to figure out the regex to get "a string of stops salmon", and I want to find given input "st", the position of "stops" and "string"
15:40:04
jmercouris
I also tried \st\b but the syntax is confusing me, I got so used to emacs regex, and it keeps changing for every language
15:41:26
Bicyclidine
it kind of goes badly with the string syntax. for ppcre to see a backslash b, the lisp string would be "\\b"
15:43:13
loke`
jmercouris: I'm not sure I understand what you want... Do you want to match any word beginning with "st"?
15:45:34
warweasle
In cl-autowrap, Am I supposed to export it's functions directly or are they just thin wrappers around the library?
15:47:05
rumbler31
^ talks about adding emacs settings that make some of the pain go away, and explain what's going on
15:50:27
jdz
jmercouris: you can also do (ppcre:parse-string "\\W(st\w+)") to see if it does what you expect.
15:52:13
jdz
Back in the day Edi had the regexp-coach (that used cl-ppcre) GUI application. Not sure it is still ovailable.