Search
Monday, 12th of March 2018, 21:02:29 UTC
21:06:18
stassats
you can stop checking, "everybody read it differently"
21:06:53
kpoeck
where is the language lawyer when we need him
21:06:55
stassats
the example provided by clhs doesn't produce the correct results if the result of the previous snippet is #\Space
21:07:20
stassats
so how the hell did people read into that interpretation?
21:08:58
kpoeck
Looking at symbolics results?
21:09:01
stassats
even cltl2 is saying how it should work, and sbcl and others do not agree
21:11:35
stassats
actually, cltl2 disagrees a bi
21:13:56
specbot
The RECURSIVE-P argument: http://www.lispworks.com/reference/HyperSpec/Body/23_acb.htm
21:15:43
stassats
ok, the standard itself doesn't have its mind made up
21:55:58
drmeister
ACTION doesn’t feel so bad now
21:57:09
drmeister
Waiting to board my plane. 5 min
22:23:05
drmeister
It looks like recursiveP is not equivalent to my *preserve-whitespace-p* special variable.
22:23:26
stassats
drmeister: i have no idea how recursive-p is supposed to work
22:23:42
stassats
the standard gives conflicting information
22:24:17
stassats
but read without recursive-p clearly should consume the trailing whitespace
22:24:47
drmeister
Ok - I'm with you on the read without recursive-p.
22:25:26
drmeister
And I'm unsettled by the uncertainty in how recursive-p is supposed to work but I'll be ok.
22:25:50
stassats
everyone appears to be ok, implementing it however they like
22:26:18
stassats
i think the approach clasp is taking is valid, i.e. preserving on recursive-p being true
22:26:30
stassats
so, no need to change that
22:27:02
drmeister
Ambiguous standards lead to moral turpitude.
22:29:43
drmeister
Hmm, now I can go (read-char) <enter> <enter> <enter> <enter> as many times as I like and it doesn't return until I provide a non-whitespace character.
22:35:44
drmeister
I tried to implement read like this...
22:35:45
drmeister
https://www.irccloud.com/pastebin/d3uNkHVD/
22:36:25
drmeister
But the cl__peek_char(T,sin,eofErrorP,eofValue,_Nil<T_O>()) doesn't work as I first expected it to.
22:36:35
drmeister
I think it blocks until it hits a non-whitespace character.
22:37:28
Bike
well, one thing is that it only actually reads one character
22:37:39
Bike
i mean, one whitespace is eaten, not all of it
22:37:42
drmeister
What I think I want it to do is advance over whitespace characters and when it hits the end of file it should return.
22:38:23
drmeister
Passing T as the first argument eats all whitespace until a non-whitespace character is seen - if I understand correctly.
22:40:14
Bike
i mean read does this. one whitespace
22:40:52
drmeister
You mean read should eat just one whitespace character?
22:41:15
Bike
"read-preserving-whitespace is like read but preserves any whitespace[2] character that delimits the printed representation of the object."
22:45:38
Bike
https://github.com/sbcl/sbcl/blob/master/src/code/reader.lisp#L772-L779
23:44:33
whoman
no github activity for a year plus .. ?
0:15:12
stassats`
whoman: we're exchanging code on floppy disks
0:20:23
whoman
as long as ur feeding the sled dogs
2:38:00
drmeister
whoman: You need to look deeper - look at the dev branch
2:40:57
drmeister
I pushed a fix for READ
2:43:21
whoman
ahh, didnt realise there are branches! i dont use git enough
6:14:14
beach
Good morning everyone!
7:59:27
frgo
Good morning. Living on countryside has its disadvantages. Missed my plane due to having to go to the airport by car and traffic jam caused a 1 hour delay for just 20km.... Gave up after 1 and a half hours and drove back home. Hmpf.
8:47:12
Shinmera
Shoulda flown to the airport to bypass the traffic
8:58:44
attila_lendvai
good morning for the european (+asian?) team! :)
8:59:56
Shinmera
I don't think we have anyone from asia (yet)
Tuesday, 13th of March 2018, 9:02:29 UTC