<aw->
Regenaxer: when I call (close *Sock) on a TCP socket, it returns the FD number to confirm it was closed, but it still appears in /prod/<pid>/fd/; what could cause that?
orivej has quit [Ping timeout: 260 seconds]
orivej has joined #picolisp
<aw->
they appear to be in CLOSE_WAIT 19328/picolisp status
orivej has quit [Quit: No Ping reply in 180 seconds.]
orivej has joined #picolisp
<Regenaxer>
Hi aw-! Hmm, no idea atm
<Regenaxer>
Is the other side still connected
<aw->
hmm indeed, other side disconnected
<Regenaxer>
and still in /proc ?
<aw->
i do (close *Port) and then (close *Sock)
<Regenaxer>
yes, close *Port can be earlier
<Regenaxer>
it is only for listening I think
<Regenaxer>
If you connect again, do you get the same socket fd again? I would expect so
<Regenaxer>
You could also call shutdown()
<Regenaxer>
I don't call it in pil, but in httpGate I do iirc
<Regenaxer>
Still strange, I would expect the fd to disappear from /proc after closing it
<aw->
no, different socket fd each time
<Regenaxer>
I see
<Regenaxer>
yeah, must be, if still in /proc
<beneroth>
Nistur, haha nice one
<aw->
OK Regenaxer that gave me an idea
<aw->
thanks
<aw->
i might have fixed it
<Regenaxer>
Ah, cool! How?
<aw->
i copied the code from misc/chat
<aw->
i realized the code after the (loop) was occuring in the child, so i adjusted my code to work that way
<Regenaxer>
ok
<Regenaxer>
Does the fd disappear from /proc then?
<Regenaxer>
So it was the parent
<Regenaxer>
that still had it open?
<aw->
yes
orivej has quit [Quit: No Ping reply in 180 seconds.]
<aw->
it was the parent that had the sockets open
<Regenaxer>
Good, makes sense
<aw->
just poorly coded ;)
<Regenaxer>
Happens :)
<aw->
i copy your example and it works as intended haha
<aw->
ok maybe now I'll release this
orivej has joined #picolisp
orivej has quit [Quit: No Ping reply in 180 seconds.]
orivej has joined #picolisp
razzy has quit [Quit: Connection closed]
orivej has quit [Ping timeout: 256 seconds]
orivej_ has joined #picolisp
orivej_ has quit [Quit: No Ping reply in 180 seconds.]
orivej has joined #picolisp
orivej has quit [Ping timeout: 240 seconds]
orivej has joined #picolisp
rob_w has quit [Remote host closed the connection]
<beneroth>
aw-, awesome, very nice!
orivej has quit [Ping timeout: 256 seconds]
cyborg_ar has quit [Ping timeout: 260 seconds]
cyborg_ar has joined #picolisp
<aw->
;)
orivej has joined #picolisp
orivej has quit [Quit: No Ping reply in 180 seconds.]
orivej has joined #picolisp
orivej has quit [Ping timeout: 256 seconds]
orivej has joined #picolisp
orivej has quit [Ping timeout: 256 seconds]
orivej has joined #picolisp
mtsd has quit [Quit: Leaving]
orivej_ has joined #picolisp
orivej has quit [Ping timeout: 256 seconds]
orivej_ has quit [Ping timeout: 260 seconds]
rob_w has joined #picolisp
orivej has joined #picolisp
orivej has quit [Quit: No Ping reply in 180 seconds.]
orivej has joined #picolisp
razzy has joined #picolisp
razzy has quit [Ping timeout: 240 seconds]
orivej_ has joined #picolisp
orivej has quit [Read error: Connection reset by peer]
orivej_ has quit [Ping timeout: 260 seconds]
orivej has joined #picolisp
orivej has quit [Read error: Connection reset by peer]
orivej has joined #picolisp
orivej has quit [Ping timeout: 256 seconds]
orivej has joined #picolisp
orivej_ has joined #picolisp
orivej has quit [Read error: Connection reset by peer]
orivej_ has quit [Quit: No Ping reply in 180 seconds.]
orivej has joined #picolisp
<Regenaxer>
aw-: Thanks for publishing picolisp-kv ! :)
<beneroth>
absolutely
<beneroth>
I actually thought before that pilDB might offer itself as an alternative Redis implementation
<beneroth>
but my Redis understanding is rather superficial
razzy has joined #picolisp
<beneroth>
s/offer/lend
<razzy>
hi i have trouble makinch changes in pil21 and|or compilling lastest version. when compiling lastest version it says https://www.heypasteit.com/clip/0IUSK8
<razzy>
i will read logs to check even late answers
orivej_ has joined #picolisp
orivej has quit [Read error: Connection reset by peer]
zillion has joined #picolisp
orivej has joined #picolisp
orivej_ has quit [Ping timeout: 264 seconds]
razzy has quit [Ping timeout: 264 seconds]
orivej has quit [Ping timeout: 260 seconds]
orivej has joined #picolisp
zillion has quit [Remote host closed the connection]
zillion has joined #picolisp
zillion has quit [Remote host closed the connection]
orivej has quit [Read error: Connection reset by peer]
orivej has joined #picolisp
razzy has joined #picolisp
razzy has quit [Client Quit]
razzy has joined #picolisp
orivej has quit [Ping timeout: 246 seconds]
orivej has joined #picolisp
orivej has quit [Ping timeout: 246 seconds]
bitmapper has joined #picolisp
<bitmapper>
how is pil21 doing?
<Regenaxer>
Progressing ... :)
<Regenaxer>
razzy: No idea what this error means
<razzy>
Regenaxer i will continue to explore pil21
<Regenaxer>
ok, good
<razzy>
Regenaxer, question unrelated to posted error: I was trying to modify src and i was unable to compile even with slightest changes
<Regenaxer>
Hmm, yes, it is not simple
<Regenaxer>
You must know how the source maps to the llvm architecture
<Regenaxer>
So the above pil21 soure directly maps to the *.ll source
<razzy>
i will think about it. afk, empty battery
razzy has quit [Quit: Connection closed]
<beneroth>
Regenaxer, +Aux relations with some of the aux'ed properties being NIL is fine, correct?
<Regenaxer>
yes, the values
<Regenaxer>
They may also be set or cleared later
<Regenaxer>
The index will always be updated accordingly
<beneroth>
good
<beneroth>
because I have here some properties which would make sense to have a +Ref on, but I find an index on every property a bit much. so I +Aux them together according to their usual use (they have a strong but not strict relation)
<Regenaxer>
ok if they have a natural order
<Regenaxer>
like (date time)
<beneroth>
hmmm
<beneroth>
+Number and +Link
<Regenaxer>
So the index is sorted by number
<beneroth>
+Link: a rather static number of objects
<Regenaxer>
if you don't know the number you must traverse the whole tree
<beneroth>
the alternative would be to have +Ref indexes for each of them
<Regenaxer>
yeah
<beneroth>
the range of values for both is rather small
<Regenaxer>
but these indexes are smaller each
<beneroth>
T
<beneroth>
right
<beneroth>
well the number of objects with this properties is huge
<beneroth>
but the indexed values are just low two digit count
<Regenaxer>
So perhaps index only the link, and simply search for the number in the results?
<beneroth>
there is certainly also a business need to find all objects with a certain number irrespective of the link
<Regenaxer>
slower but smaller :)
<beneroth>
but usually the query would be for both
<Regenaxer>
with known number?
<beneroth>
yeah
<beneroth>
all with number 7
<beneroth>
or all with Link {XX}
<Regenaxer>
ok, then +Aux is perfect
<Regenaxer>
The second is not possible then
<Regenaxer>
or all with Link {XX}
<beneroth>
both number or link might be missing, but only for a very small number of records (new ones)
<beneroth>
ok
<beneroth>
so two separate indexes
<Regenaxer>
I think this is easiest
<beneroth>
and when we want to query with number and link, then we use select/3
<beneroth>
simplest for sure :)
<Regenaxer>
right
<beneroth>
thanks for your advice :)
<Regenaxer>
The best would be +Aux +Ref +Number with link PLUS just the indexed link
miskatonic has joined #picolisp
<Regenaxer>
then all 3 ways can be searched directly
<Regenaxer>
fastest but biggest
<beneroth>
Regenaxer, ah you mean (foo (+Ref +Link)) and (bar (+Aux +Ref +Number) (foo)) ?
<Regenaxer>
yes
<Regenaxer>
searches only number, number + link, and link only
<beneroth>
will do that
<beneroth>
yeah perfect
<beneroth>
each an optimal BTree
<Regenaxer>
Not so very big I think
<Regenaxer>
yeah
<beneroth>
and even so, it's okay
<beneroth>
storage and RAM should not be an issue
<Regenaxer>
perfect :)
<beneroth>
it's small values, so small entries, just many of them :)
<beneroth>
but not string indexing or such :)
<Regenaxer>
right, number indexing is fast
<Regenaxer>
i.e. number comparisons
<beneroth>
it all boils down to comparison of raw nums, technically, right?
<Regenaxer>
yes
<beneroth>
just in case of strings, pretty quickly this will be technicall bignums (chained)
<Regenaxer>
ideally short nums
<beneroth>
so short string, short values, short nums :)
<beneroth>
+Link are short anyway
<Regenaxer>
yes, still strings are compared by characters
<Regenaxer>
nums in a single piece
<beneroth>
okay
<Regenaxer>
short ones
<beneroth>
of course
<Regenaxer>
bignums are expensive too
<Regenaxer>
comparison must start at the tail (most significant words)
<beneroth>
well this one should replace a previous (badly programmed) application which was quite slow (just because of bad programming)
<Regenaxer>
good, then no problem ;)
<beneroth>
aye
<beneroth>
only bad thing is
<beneroth>
as of now I use a simple forking server without keeping session processes around, so not much of caching in pil ram
<beneroth>
this I will change asap after initial version
<Regenaxer>
ok
<beneroth>
I guess that will make a huge difference, but I guess/hope the initial version will be good enough even with one-shot forking architecture
<Regenaxer>
Perhaps use a server without forking at all?
<beneroth>
yes that is the plan
<beneroth>
fixed number of workers
<Regenaxer>
if not many acccesses in parallel
<beneroth>
not many probably, maybe a few dozen users at the same time, at least initially
<beneroth>
maybe more later
<beneroth>
one process per core :-)
<beneroth>
or a few more, in case they're waiting on IO
<Regenaxer>
I experimented with non-forking server some time age. Single process only: http://ix.io/2pmm
<beneroth>
hehe
<beneroth>
good cache usage :)
<Regenaxer>
yes
<beneroth>
yeah maybe one should start this way
<Regenaxer>
The file is from Dec 2008
<beneroth>
and then add processes whenever some slowness could be seen
<beneroth>
though IO could make it busy
<Regenaxer>
T
orivej has joined #picolisp
orivej has quit [Quit: No Ping reply in 180 seconds.]
orivej has joined #picolisp
orivej has quit [Quit: No Ping reply in 180 seconds.]
orivej has joined #picolisp
orivej_ has joined #picolisp
orivej has quit [Read error: Connection reset by peer]
orivej_ has quit [Quit: No Ping reply in 180 seconds.]
orivej has joined #picolisp
orivej has quit [Ping timeout: 264 seconds]
orivej has joined #picolisp
orivej has quit [Quit: No Ping reply in 180 seconds.]
orivej has joined #picolisp
orivej_ has joined #picolisp
orivej has quit [Ping timeout: 264 seconds]
rob_w has quit [Ping timeout: 258 seconds]
rob_w has joined #picolisp
rob_w has quit [Remote host closed the connection]
orivej_ has quit [Ping timeout: 256 seconds]
Nistur has quit [*.net *.split]
DerGuteMoritz has quit [*.net *.split]
casaca has quit [*.net *.split]
aw- has quit [*.net *.split]
elioat has quit [*.net *.split]
lodsw has quit [*.net *.split]
bitmapper has quit [*.net *.split]
masak has quit [*.net *.split]
patrixl has quit [*.net *.split]
twelvemonkeys has quit [*.net *.split]
libertas has quit [*.net *.split]
anddam has quit [*.net *.split]
f8l has quit [*.net *.split]
pointfree has quit [*.net *.split]
Regenaxer has quit [*.net *.split]
anjaa has quit [*.net *.split]
root____ has quit [*.net *.split]
emacsomancer has quit [*.net *.split]
mario-goulart has quit [*.net *.split]
ChanServ has quit [*.net *.split]
Blukunfando has quit [*.net *.split]
inara has quit [*.net *.split]
miskatonic has quit [*.net *.split]
tankf33der has quit [*.net *.split]
michelp has quit [*.net *.split]
viaken has quit [*.net *.split]
joebo has quit [*.net *.split]
twelvemonkeys has joined #picolisp
miskatonic has joined #picolisp
inara has joined #picolisp
Blukunfando has joined #picolisp
masak has joined #picolisp
bitmapper has joined #picolisp
tankf33der has joined #picolisp
anddam has joined #picolisp
libertas has joined #picolisp
patrixl has joined #picolisp
casaca has joined #picolisp
root____ has joined #picolisp
Regenaxer has joined #picolisp
anjaa has joined #picolisp
joebo has joined #picolisp
lodsw has joined #picolisp
Guest92206 has joined #picolisp
DerGuteMoritz has joined #picolisp
aw- has joined #picolisp
viaken has joined #picolisp
Nistur has joined #picolisp
ChanServ has joined #picolisp
elioat has joined #picolisp
pointfree has joined #picolisp
_whitelogger has joined #picolisp
miskatonic has quit [Quit: ERC (IRC client for Emacs 24.5.1)]