<phadthai>
I don't think removing format from the language is realistic, but I did use other things that are more like templates as well, CL nicely allowing it
<phadthai>
hello beach
<aeth>
I don't use FORMAT that much, but I don't have to. Unless you're doing FORMAT NIL you can just mix and match whatever you want to write to a stream, and I find that that leads to clearer code, personally.
Kevslinger has quit [Quit: Connection closed for inactivity]
orivej has joined #lisp
loskutak has joined #lisp
jello_pudding has quit [Quit: Quit]
nowhereman has joined #lisp
<beach>
Posterdati: That won't be a specialized array though.
<beach>
Posterdati: What is it that you are trying to do.
<beach>
So you can also write (TYPEP (MAKE-ARRAY 4 :ELEMENT-TYPE '(UNSIGNED-BYTE *) :INITIAL-CONTENTS '(1 2 3 4)) '(SIMPLE-ARRAY T))
<beach>
And it will also return T.
<beach>
And that's because (upgraded-array-element-type '(unsigned-byte *)) is T.
brown121408 has quit [Ping timeout: 250 seconds]
brown121408 has joined #lisp
scymtym has quit [Ping timeout: 268 seconds]
earl-ducaine has joined #lisp
ljavorsk has joined #lisp
isBEKaml has joined #lisp
wiselord has quit [Read error: Connection reset by peer]
raghavgururajan has joined #lisp
scymtym has joined #lisp
isBEKaml has quit [Quit: leaving]
wiselord has joined #lisp
loskutak has quit [Ping timeout: 276 seconds]
karlosz has quit [Quit: karlosz]
ljavorsk has quit [Ping timeout: 276 seconds]
Bike has joined #lisp
william1 has joined #lisp
enrio has joined #lisp
fookara has quit [Remote host closed the connection]
xkapastel has quit [Quit: Connection closed for inactivity]
raghavgururajan has quit [Remote host closed the connection]
loskutak has joined #lisp
smokeink has quit [Quit: Leaving]
<pjb>
smokeink: perhaps the author found an implementaion where apropos expected a package, instead of nil or a package designator (note that nil is a package designator for a package named "NIL"…)
clothespin_ has quit [Ping timeout: 265 seconds]
clothespin__ has quit [Ping timeout: 265 seconds]
<pjb>
APROPOS is specified so it's impossible to select only symbols in a package named "NIL".
<pjb>
So I would have implemented an AP that would have dealt better with this situation instead…
slyrus_ has joined #lisp
slyrus__ has quit [Ping timeout: 268 seconds]
wxie has joined #lisp
longshi has joined #lisp
<pjb>
Posterdati: (make-array dimensions :element-type `(unsigned-byte ,n) :initial-element 0) will make you an array of elements of n bits.
__vlgvrs has quit [Remote host closed the connection]
Necktwi has quit [Read error: Connection reset by peer]
brown121408 has quit [Read error: Connection reset by peer]
brown121408 has joined #lisp
wxie has quit [Ping timeout: 245 seconds]
Necktwi has joined #lisp
<beach>
Posterdati: Did you see what I wrote about specialized arrays?
ljavorsk has joined #lisp
<Posterdati>
beach: yes
<beach>
I don't know the reason why you are doing what you are doing, but you might have some surprises if you think that the array element type will be (integer 0) when that is the type you give to make-array.
<pjb>
it depends on the type AND on the implementation.
<pjb>
for example, (subtypep (array-element-type (make-array 4 :element-type (quote (unsigned-byte 4)))) (quote (unsigned-byte 4))) returns NIL in abcl, ccl, and ecl, but T in clisp and sbcl.
<jackdaniel>
(integer 0) is unlikely to have a specialized representation because it is unknown how many bytes one field should reserve
<beach>
Posterdati: If you insist (subtypep (make-array 4 :element-type '(integer 0 *)) '(integer 0 *)) => NIL, T
<beach>
jackdaniel: That's what I am trying to tell Posterdati, but I don't think my message is clear.
<pjb>
You never know when somebody will write an implementation where integers are better represened than other objects and where arrays can store integers in a specialized array.
<pjb>
Again that is not the point. The point is the semantics of the conforming application.
<pjb>
The rest is the problem of the implementation and implementer, and we don't care if they are not able to optimize our application specific case.
<pjb>
We'll just choose an implementation that is better able to deal with our conforming code…
Lord_of_Life_ has joined #lisp
<loke>
Is it legal for a Lisp implementation to limit the size of an integer?
<loke>
(I mean, there is always a limit in the amount of available memory, but can you have a conforming implementation where integer and fixnum are the same?
<jackdaniel>
no, I've said it is unlikely that arbitrary integer will have specialized array representation
fookara has joined #lisp
<pjb>
loke: IIRC, they must be distinct, but integer could be quite limited.
<jackdaniel>
but bignums are obligatory in conforming implementation
<loke>
Right. But there is no minimum size of a bignum?
<jackdaniel>
minimum?
<pjb>
We could have 16-bit fixnums and 64-bit bignums.
<jackdaniel>
" The type bignum is defined to be exactly (and integer (not fixnum)). "
<jackdaniel>
from the spec
<loke>
Yeah. I mean could a bignum be limited to, say, 65 bits?
Lord_of_Life has quit [Ping timeout: 240 seconds]
Lord_of_Life_ is now known as Lord_of_Life
<pjb>
System Class INTEGER: An integer is a mathematical integer. There is no limit on the magnitude of an integer.
<jackdaniel>
the only limit is the memory available. I suppose one could decide, that bignum has tops 65 bits and signal serious-condition that storage is exhausted, but that would be lame :)
<loke>
But then a conforming implementation could exist which doesn't actually have a bignum implementtation. The type would exist, but there would be no way to obtain an instance of such type.
<pjb>
So probably a hard limit such as 64-bit or 128-bit would not be acceptable. What "There is no limit on the magnitude of an integer." must mean, is that the actual limit should be a function of the memory size available.
<loke>
There is nothing that says that (1+ MAX-POSITIVE-FIXNUM) must succeed?
<pjb>
loke: only if your system only has 16 bits of memory.
<pjb>
Now, the maximum integer size can be a small or very small fraction of the memory. But it must still be more or less proportional to it.
<loke>
pjb: I agree that that is clearly the spirit of the standard.
<pjb>
If you use generational GC, (/ memory-size (* ngen 2)) at most.
* loke
is getting flashbacks from my argument on the SBCL list about the guarantees of SORT, where SBCL's position can only be maintined with a _very_ literal reading of the spec but was clearly against its spirit
<pjb>
But nothing prevents the implementation to have more overhead. (/ (- memory-size some-gigabyes) (* ngen 2 some-factor)) such as the maximum integer size could be in the 10,000 bits… But I would expect it to be above 2Kbit at least.
<pjb>
Or for example, that integer size must increment by a small amount, such as one word. bigint sizes could be exponents of 2.
<Bike>
really? what's ambiguous about sort?
<loke>
Bike: look up the spec and tell me if SORT on a vector is guaranteed to modify the array in-place, or if the implementation is allowed to allocate a new array and return it.
<Bike>
oh, the latter.
<Bike>
doesn't sbcl destroy it?
<loke>
Yes, I agree. And there are even examples in the spec that explains that literally says so.
<Bike>
but i don't understand what you mean by sbcl's position - it's certainly allowed for sort to always destroy its argument.
<loke>
But the SBCL guys insist it's guaranteed, and the argument about the example is waved away with the argument that the notes section are not normative
<Bike>
but i mean, even if it's allowed to not destroy it, sbcl can just decide not to make use of that latitude
<loke>
Bike: Allowed, yes. But SBCL argues that it's gunarteed (SBCL used to emit a warning when you didn't read the return value of SORT, but they removed it with the argument that it's gunarteed by the spec)
<Bike>
oh, i see.
<loke>
I found that to be odd, since the spec explicitly says that it's allowed to copy.
kritixilithos has quit [Read error: Connection reset by peer]
oxford has quit [Read error: Connection reset by peer]
zooey_ has quit [Remote host closed the connection]
<jackdaniel>
where does it say that it is allowed to copy?
<Bike>
well it's not that explicit, it just says sorting "can be" destructive, which kind of implies it doesn't have to be.
<jackdaniel>
(examples are not normative!)
kritixilithos has joined #lisp
<loke>
jackdaniel: yeah, but nothing else says it it _must_ be destructive.
<Bike>
oh,wait, the notes also say the result might not be identical to the argument.
<loke>
Interesting... "the usual sense". SO I guess an implementation that hard crashes when overflowing a fixnum woul dbe technically compliant (albeit a bit useless)
<Bike>
this doesn't rule out signalling an error on (1+ most-positive-fixnum), i don't think
<jackdaniel>
I don't think that hard crashes are conforming ,)
<jackdaniel>
I'd expect a storage-condition being signalled
aindilis has quit [Read error: Connection reset by peer]
<pjb>
Yes, it says that you can get an out-of-memory condition, but not an overflow-error or integer-too-big (unless integer-too-big is a subclass of out-of-memory).
<_death>
were any implementations submitted to storage stress tests
<pjb>
_death: somebody tried it a few year ago, with results in cll.
<pjb>
It would have to be included as an annex to ansi-test…
<pjb>
Most implementations just crash, so it's not something that can be tested entirely from within the lisp image.
aindilis has joined #lisp
<pjb>
Also, it's dependent on the memory size, so to test it well, you need both a lot of memory, and a way to restrict the memory allocated to the implementation.
<pjb>
ie. you may have to use VMs.
<_death>
well sbcl can often run out of dynamic space and end up in the debugger
<pjb>
That's the worse: the process doesn't exit, but is hung in the debugger. How a controlling process can detect it?
<pjb>
s/e/t
<_death>
if that's important then you should --disable-debugger
EvW1 has joined #lisp
<_death>
but I didn't check if a storage-condition is signaled
<_death>
(I mean "often" as in, there's a cycle of developing, getting hit by it, tweaking dynamic space size.. in production the gc seems to do ok if enough RAM is available)
brown121408 has quit [Ping timeout: 240 seconds]
brown121408 has joined #lisp
FreeBirdLjj has joined #lisp
clothespin has joined #lisp
clothespin_ has joined #lisp
dddddd has joined #lisp
gareppa has quit [Quit: Leaving]
brown121408 has quit [Read error: Connection reset by peer]
raghavgururajan has quit [Read error: Connection reset by peer]
longshi has joined #lisp
tourjin has joined #lisp
brown121407 has joined #lisp
Oddity has quit [Ping timeout: 240 seconds]
ljavorsk_ has joined #lisp
ljavorsk has quit [Ping timeout: 240 seconds]
clothespin_ has quit [Ping timeout: 252 seconds]
clothespin has quit [Ping timeout: 276 seconds]
Oddity has joined #lisp
grewal_ has quit [Ping timeout: 265 seconds]
grewal_ has joined #lisp
william1 has quit [Ping timeout: 250 seconds]
phlim has quit [Quit: WeeChat 2.4]
vseloved has quit [Ping timeout: 265 seconds]
cl-arthur has quit [Read error: Connection reset by peer]
EvW1 has quit [Ping timeout: 245 seconds]
Inline has quit [Quit: Leaving]
brown121407 has quit [Read error: Connection reset by peer]
iovec has joined #lisp
davepdotorg has quit []
phlim has joined #lisp
Lycurgus has joined #lisp
raghavgururajan has joined #lisp
Inline has joined #lisp
noobineer has quit [Read error: Connection reset by peer]
william1 has joined #lisp
FreeBird_ has joined #lisp
FreeBirdLjj has quit [Ping timeout: 240 seconds]
Inline has quit [Quit: Leaving]
noobineer has joined #lisp
raghavgururajan has quit [Read error: Connection reset by peer]
stepnem has quit [Ping timeout: 268 seconds]
william1 has quit [Quit: WeeChat 1.9.1]
william1 has joined #lisp
davepdotorg has joined #lisp
noobineer has quit [Read error: Connection reset by peer]
stepnem has joined #lisp
kslt1 has joined #lisp
kslt1 has quit [Remote host closed the connection]
william1 has quit [Ping timeout: 240 seconds]
Inline has joined #lisp
kslt1 has joined #lisp
efm has quit [Ping timeout: 268 seconds]
efm has joined #lisp
kslt1 has quit [Remote host closed the connection]
kslt1 has joined #lisp
Krystof has quit [Ping timeout: 252 seconds]
efm has quit [Read error: Connection reset by peer]
davepdotorg has quit [Remote host closed the connection]
Kevslinger has joined #lisp
ebzzry has joined #lisp
tourjin has quit [Ping timeout: 268 seconds]
efm has joined #lisp
kslt1` has joined #lisp
xkapastel has joined #lisp
kslt1`` has joined #lisp
kslt1 has quit [Ping timeout: 268 seconds]
stepnem has quit [Ping timeout: 240 seconds]
pfdietz27 has joined #lisp
kslt1` has quit [Ping timeout: 240 seconds]
stepnem has joined #lisp
cosimone has joined #lisp
defaultxr has quit [Ping timeout: 265 seconds]
zmt00 has quit [Quit: Leaving]
tarod16 has joined #lisp
<tarod16>
hi
<tarod16>
Do you know vms14?
<beach>
vms14 comes here regularly. Why?
<tarod16>
I want to contact him
<boeg>
If I have a function `(defun test (alst) (let ((name (assoc 'name alst))) (print name)))`, why can't I do `(test '((name . "Bob")))` and have it print out "(NAME . "Bob")" when I can do `(assoc 'name '((name . "Bob")))` and it returns "(NAME . "Bob")" as expected?
<boeg>
The function just prints out "NIL"
<tarod16>
Can you guys tell to vms14 that I was here?
<beach>
minion: memo for vms14: tarod16 was here looking for you.
<minion>
Remembered. I'll tell vms14 when he/she/it next speaks.
<tarod16>
btw, I'm interested in Lisp programming because of him haha
<beach>
Great!
<tarod16>
Thanks :)
FreeBird_ has quit [Remote host closed the connection]
tarod16 has left #lisp ["Killed buffer"]
<beach>
boeg: Should work.
<boeg>
beach: I thought so too
<boeg>
but it doesn't
<beach>
What is the exact form you type that doesn't work?
<boeg>
exactly that
<beach>
Forget the `s
<boeg>
I copy pasted those things and added the ` and " around
<boeg>
I load the function from a buffer into sly
<boeg>
and then I call the functions in the repl
<boeg>
well, actually, i'm mistaken
gareppa has joined #lisp
<boeg>
the actual function call is to http::test, not test. It's in the system land-of-lisp.http - the http function has the nickname http so I do (http::test ...)
<boeg>
does that make a difference?
rotucer has quit [Ping timeout: 265 seconds]
<beach>
Then try (test '((http::name . "Bob"))) instead.
<beach>
You very likely have two symbols named NAME.
<boeg>
that works. It returns "(LAND-OF-LISP.HTTP::NAME . "Bob")"
<beach>
There you go.
stepnem has quit [Ping timeout: 250 seconds]
<boeg>
so that is because there are two symbols named "name"?
<beach>
Yes.
<beach>
One in the package that was current when the TEST function was defined.
<beach>
And the other one in the package that is current in your REPL.
<boeg>
So when I'm working inside a package and referencing symbols, I should preface them with, in this example, http:: to make sure it's the correct one i'm targeting?
<boeg>
because symbols are global
<beach>
The word "global" does not characterize symbols. Symbols are just objects. Each symbol typically has a name and a package.
<boeg>
alright
<beach>
In your case, the two names are the same, but the packages are different.
<boeg>
so to not run into that problem again, one would "namespace" symbols in practice?
brown121407 has joined #lisp
Dibejzer has joined #lisp
<beach>
It is hard to give a general answer to that.
Dibejzer has quit [Client Quit]
<beach>
I typically have many packages, and each one :USEs the CL package but no others.
<beach>
So if I need to refer to a symbol that is neither in the current package nor in the CL package, I use an explicit package prefix.
<beach>
Other people use :USE more frequently, and then you don't need the package prefix for symbols in the :USEd packages.
<boeg>
alright thanks
<beach>
Sure.
chipolux has quit [Ping timeout: 250 seconds]
stepnem has joined #lisp
chipolux has joined #lisp
smokeink has joined #lisp
__vlgvrs has quit [Quit: Leaving]
gabiruh has joined #lisp
kritixilithos has quit [Quit: quit]
rotucer has joined #lisp
<boeg>
Anybody here know about usocket? When I read-line my socket stream, I get an utf-8 decoding error. Do I have to tell usocket to stream in utf-8 or something? I've tried googling but with no luck
efm has quit [Ping timeout: 240 seconds]
enzuru has joined #lisp
efm has joined #lisp
fanta1 has joined #lisp
hiroaki has joined #lisp
scymtym_ has joined #lisp
kslt1`` has quit [Ping timeout: 240 seconds]
scymtym has quit [Ping timeout: 252 seconds]
earl-ducaine has quit [Ping timeout: 276 seconds]
brown121407 has quit [Read error: Connection reset by peer]
doublex has quit [Read error: Connection reset by peer]
doublex has joined #lisp
<boeg>
smokeink: thanks, ill check it out
<smokeink>
also try these search-strings on google: "usocket filetype:lisp" "usocket inurl:github"
earl-ducaine has joined #lisp
doublex has quit [Read error: Connection reset by peer]
doublex has joined #lisp
<boeg>
smokeink: yes, thanks
kslt1`` has joined #lisp
oni-on-ion has quit [Remote host closed the connection]
oni-on-ion has joined #lisp
fanta1 has quit [Quit: fanta1]
gareppa has quit [Quit: Leaving]
_whitelogger has joined #lisp
earl-ducaine has quit [Ping timeout: 276 seconds]
fookara has quit [Remote host closed the connection]
longshi has quit [Ping timeout: 276 seconds]
oni-on-ion has quit [Ping timeout: 245 seconds]
cosimone has quit [Quit: Terminated!]
oni-on-ion has joined #lisp
stepnem has quit [Ping timeout: 265 seconds]
stepnem has joined #lisp
scymtym has joined #lisp
buffergn0me has joined #lisp
brown121407 has joined #lisp
<boeg>
smokeink: Omg ... the problem was that i have the https everywhere extension in the browser i use so it calls the server with ssl over http so yeah
xkapastel has quit [Quit: Connection closed for inactivity]
<boeg>
forcing http and everything worked
<smokeink>
:)
ggole has quit [Quit: Leaving]
izh_ has joined #lisp
Krystof has joined #lisp
<boeg>
if I have (in-package :my-package) at the top of my file, why do I still have to designate :my-package when I use `intern` or else it doesn't work?
<boeg>
shouldn't in-package set current package correctly so `intern` uses it?
cosimone has joined #lisp
stepnem_ has joined #lisp
ljavorsk_ has quit [Ping timeout: 250 seconds]
bitmapper has joined #lisp
stepnem has quit [Ping timeout: 250 seconds]
enrioog has joined #lisp
<boeg>
Seems if I set the current-package with in-package in the repl, it works as intended
enrio has quit [Ping timeout: 240 seconds]
<Bike>
depends on where you're using intern
<Bike>
if you have like (in-package foo) (defun bar () (intern ...)), the package used by intern will be the one in place when bar is CALLED
<Bike>
rather than defined, which you may be expecting
rozenglass has joined #lisp
<boeg>
yeah, so the repl was in cl-user and intern was called in a function in a file with (in-package...) at the top, and so I thought when intern was called, the package set with in-package would be current-package, but it was cl-user. That was the mistake
<boeg>
I guess thats exactly what you just said so yeah, he :P
Arcaelyx has joined #lisp
loskutak has quit [Ping timeout: 250 seconds]
smokeink has quit [Remote host closed the connection]
loskutak has joined #lisp
smokeink has joined #lisp
enzuru has quit [Ping timeout: 240 seconds]
enzuru has joined #lisp
stepnem_ has quit [Ping timeout: 276 seconds]
stepnem has joined #lisp
phlim has quit [Quit: WeeChat 2.2]
phlim has joined #lisp
notzmv has quit [Remote host closed the connection]
buffergn0me has quit [Ping timeout: 245 seconds]
oxum_ has joined #lisp
oxum has quit [Read error: Connection reset by peer]