Go Language Resources Go, golang, go... NOTE: This page ceased updating in October, 2012

--- Log opened Thu Oct 21 00:00:11 2010
00:08 -!- Wiz126 [Wiz126@h62.126.232.68.ip.windstream.net] has quit [Ping timeout:
240 seconds]
00:13 -!- felipe [~felipe@my.nada.kth.se] has quit [Quit: felipe]
00:13 -!- MaksimBurnin [~max@87.224.188.44] has quit [Read error: No route to
host]
00:15 -!- kanru [~kanru@118-160-163-40.dynamic.hinet.net] has joined #go-nuts
00:21 -!- rboyd [~rboyd@72-161-203-70.dyn.centurytel.net] has joined #go-nuts
00:24 -!- b00m_chef [~watr@66.183.100.197] has joined #go-nuts
00:25 -!- slashus2 [~slashus2@173-16-55-129.client.mchsi.com] has quit [Quit:
slashus2]
00:32 -!- pvarga [~pvarga@pool-72-88-229-188.nwrknj.east.verizon.net] has joined
#go-nuts
00:35 -!- iant [~iant@67.218.106.211] has joined #go-nuts
00:35 -!- mode/#go-nuts [+v iant] by ChanServ
00:35 -!- Tv [~tv@cpe-76-168-227-45.socal.res.rr.com] has joined #go-nuts
00:37 -!- iant1 [~iant@166.190.205.44] has quit [Ping timeout: 240 seconds]
00:39 -!- dacc [~Adium@D-128-95-10-139.dhcp4.washington.edu] has left #go-nuts []
00:41 -!- nsf [~nsf@jiss.convex.ru] has quit [Quit: WeeChat 0.3.3]
00:42 < plexdev> http://is.gd/gaqkF by [Andrew Gerrand] in 2 subdirs of go/
-- release.2010-10-20
00:42 < plexdev> http://is.gd/gaqkH by [Andrew Gerrand] in go/ -- tag
release.2010-10-20
00:42 < plexdev> http://is.gd/gaqkP by [Rob Pike] in go/test/bench/ --
test/bench: update numbers
00:52 -!- pvarga_ [~pvarga@pool-72-88-229-188.nwrknj.east.verizon.net] has joined
#go-nuts
00:56 -!- pvarga [~pvarga@pool-72-88-229-188.nwrknj.east.verizon.net] has quit
[Ping timeout: 240 seconds]
00:56 -!- Zoopee [alsbergt@zoopee.org] has quit [Read error: No route to host]
00:57 -!- Wiz126 [Wiz126@h62.126.232.68.ip.windstream.net] has joined #go-nuts
00:59 -!- rboyd [~rboyd@72-161-203-70.dyn.centurytel.net] has quit [Quit: rboyd]
01:00 -!- iant [~iant@67.218.106.211] has quit [Quit: Leaving.]
01:01 -!- kanru [~kanru@118-160-163-40.dynamic.hinet.net] has quit [Ping timeout:
265 seconds]
01:12 -!- dj2 [~dj2@CPE001f5b35feb4-CM0014048e0344.cpe.net.cable.rogers.com] has
joined #go-nuts
01:12 -!- b00m_chef [~watr@66.183.100.197] has quit [Ping timeout: 240 seconds]
01:16 -!- jcao219 [~jcao219@pool-96-226-238-248.dllstx.fios.verizon.net] has quit
[Ping timeout: 240 seconds]
01:24 -!- artefon [~thiago@189.59.182.83.dynamic.adsl.gvt.net.br] has quit [Quit:
bye]
01:31 -!- res99 [~anonymous@201.237.130.70] has quit [Quit: res99]
01:31 -!- powerje [~powerje@adsl-75-60-217-33.dsl.wotnoh.sbcglobal.net] has quit
[Quit: powerje]
01:32 -!- slashus2 [~slashus2@74-141-107-118.dhcp.insightbb.com] has joined
#go-nuts
01:33 -!- res99 [~anonymous@201.237.130.70] has joined #go-nuts
01:47 -!- jcao219 [~jcao219@pool-96-226-238-248.dllstx.fios.verizon.net] has
joined #go-nuts
01:53 -!- kanru [~kanru@61-30-10-70.static.tfn.net.tw] has joined #go-nuts
02:00 -!- niemeyer [~niemeyer@201-40-152-123.pltce701.dsl.brasiltelecom.net.br]
has quit [Ping timeout: 245 seconds]
02:06 -!- nsf [~nsf@jiss.convex.ru] has joined #go-nuts
02:10 -!- rboyd [~rboyd@72-161-203-70.dyn.centurytel.net] has joined #go-nuts
02:40 -!- fenicks [~christian@log77-3-82-243-254-112.fbx.proxad.net] has quit
[Quit: Leaving.]
02:40 -!- scyth [~scyth@rots.in.rs] has quit [Ping timeout: 245 seconds]
02:42 -!- noktoborus_ [debian-tor@gateway/tor-sasl/noktoborus] has quit [Ping
timeout: 245 seconds]
02:45 < plexdev> http://is.gd/gaxOR by [Fumitoshi Ukai] in
go/src/pkg/websocket/ -- web socket: fix short Read
02:45 -!- scyth [~scyth@rots.in.rs] has joined #go-nuts
02:57 -!- lmoura [~lauromour@189.26.29.175.dynamic.adsl.gvt.net.br] has quit
[Remote host closed the connection]
03:30 < enferex> I am debugging a gc compiled program.  It seems that the gc
compiler does not toss in the debugging tags.  And I didn't see any 'strip' calls.
Is there a way to get a compiled gc program to have the debugging tags
implemented, or is this functionality not currently implemented?
03:38 -!- rejb [~rejb@unaffiliated/rejb] has quit [Ping timeout: 276 seconds]
03:45 -!- cco3 [~conley@c-69-181-138-209.hsd1.ca.comcast.net] has joined #go-nuts
04:02 < plexdev> http://is.gd/gaDgo by [Andrew Gerrand] in go/doc/ --
go_tutorial: change wording slightly and sync .txt and .html post-gofmt
04:09 -!- pvarga_ [~pvarga@pool-72-88-229-188.nwrknj.east.verizon.net] has quit
[Quit: pvarga_]
04:20 -!- slashus2 [~slashus2@74-141-107-118.dhcp.insightbb.com] has quit [Quit:
slashus2]
04:21 -!- zozoR [~zozoR@5634798d.rev.stofanet.dk] has joined #go-nuts
04:23 -!- dj2 [~dj2@CPE001f5b35feb4-CM0014048e0344.cpe.net.cable.rogers.com] has
quit [Remote host closed the connection]
04:30 -!- iant [~iant@adsl-71-133-8-30.dsl.pltn13.pacbell.net] has joined #go-nuts
04:30 -!- mode/#go-nuts [+v iant] by ChanServ
04:34 -!- noktoborus_ [debian-tor@gateway/tor-sasl/noktoborus] has joined #go-nuts
04:34 < plexdev> http://is.gd/gaF2W by [Andrew Gerrand] in
go/misc/dashboard/builder/ -- gobuilder: write build and benchmarking logs to disk
04:34 -!- tensorpudding [~user@99.148.202.191] has quit [Read error: Connection
reset by peer]
04:36 -!- devrim [~Adium@ip-95-223-189-66.unitymediagroup.de] has quit [Quit:
Leaving.]
04:39 -!- rbraley [~rbraley@ip72-222-128-78.ph.ph.cox.net] has quit [Ping timeout:
265 seconds]
04:41 -!- devrim [~Adium@ip-95-223-189-66.unitymediagroup.de] has joined #go-nuts
04:44 -!- jcao219 [~jcao219@pool-96-226-238-248.dllstx.fios.verizon.net] has quit
[Ping timeout: 252 seconds]
04:46 -!- devrim [~Adium@ip-95-223-189-66.unitymediagroup.de] has quit [Ping
timeout: 252 seconds]
04:46 -!- araujo [~araujo@gentoo/developer/araujo] has quit [Quit: Leaving]
04:51 < plexdev> http://is.gd/gaFSt by [Andrew Gerrand] in 2 subdirs of go/
-- misc: update python scripts to specify python2 or nothing
04:52 -!- itrekkie [~itrekkie@ip72-200-114-232.tc.ph.cox.net] has joined #go-nuts
04:54 -!- devrim [~Adium@ip-95-223-189-66.unitymediagroup.de] has joined #go-nuts
04:55 -!- devrim1 [~Adium@ip-95-223-189-66.unitymediagroup.de] has joined #go-nuts
04:58 -!- devrim [~Adium@ip-95-223-189-66.unitymediagroup.de] has quit [Ping
timeout: 265 seconds]
04:59 -!- scm [scm@d018187.adsl.hansenet.de] has quit [Ping timeout: 240 seconds]
05:00 -!- devrim1 [~Adium@ip-95-223-189-66.unitymediagroup.de] has quit [Ping
timeout: 276 seconds]
05:01 -!- scm [scm@d038144.adsl.hansenet.de] has joined #go-nuts
05:03 -!- aho [~nya@fuld-4d00d680.pool.mediaWays.net] has quit [Quit:
EXEC_over.METHOD_SUBLIMATION]
05:05 -!- Rennex [rennex@giraf.fi] has quit [Remote host closed the connection]
05:05 -!- Eridius [~kevin@unaffiliated/eridius] has quit [Read error: Operation
timed out]
05:08 < plexdev> http://is.gd/gaGIy by [Russ Cox] in 6 subdirs of go/src/ --
arm: prop up software floating point
05:10 -!- Archwyrm [~archwyrm@archwyrm.net] has quit [Ping timeout: 240 seconds]
05:10 -!- Archwyrm [~archwyrm@archwyrm.net] has joined #go-nuts
05:10 -!- Soultaker [~Soultaker@hell.student.utwente.nl] has quit [Ping timeout:
276 seconds]
05:16 -!- scarabx [~scarabx@c-76-19-43-200.hsd1.ma.comcast.net] has quit [Quit:
This computer has gone to sleep]
05:17 -!- nsf [~nsf@jiss.convex.ru] has quit [Quit: WeeChat 0.3.3]
05:36 -!- slashus2 [~slashus2@74-141-107-118.dhcp.insightbb.com] has joined
#go-nuts
05:50 -!- zozoR [~zozoR@5634798d.rev.stofanet.dk] has quit [Ping timeout: 264
seconds]
05:51 -!- zozoR [~zozoR@5634798d.rev.stofanet.dk] has joined #go-nuts
06:04 -!- ronnyy [~quassel@2001:6f8:12c6:1c86:224:1dff:fed7:9541] has joined
#go-nuts
06:25 -!- dju_ [dju@fsf/member/dju] has quit [Ping timeout: 265 seconds]
06:32 -!- ExtraSpice [~XtraSpice@88.118.33.48] has joined #go-nuts
06:42 -!- slashus2 [~slashus2@74-141-107-118.dhcp.insightbb.com] has quit [Quit:
slashus2]
06:51 -!- rlab [~Miranda@91.200.158.34] has joined #go-nuts
07:23 -!- wmohchi [~wmohchi@60.247.97.98] has quit [Ping timeout: 265 seconds]
07:29 -!- wrtp [~rog@92.17.69.32] has joined #go-nuts
07:35 -!- Fish [~Fish@86.65.182.207] has joined #go-nuts
07:38 -!- zozoR [~zozoR@5634798d.rev.stofanet.dk] has quit [Ping timeout: 276
seconds]
07:40 -!- zozoR [~zozoR@5634798d.rev.stofanet.dk] has joined #go-nuts
07:47 -!- Project_2501 [~Marvin@82.84.80.32] has joined #go-nuts
07:49 -!- rlab [~Miranda@91.200.158.34] has quit [Ping timeout: 265 seconds]
07:49 -!- vrasidakos [~lala@77.49.144.38.dsl.dyn.forthnet.gr] has joined #go-nuts
07:49 -!- Fish- [~Fish@86.65.182.207] has joined #go-nuts
07:50 -!- Netsplit *.net <-> *.split quits: thomas_b, kkress, Esmil, [Eko],
suiside, Gertm, [Pete_27], strtok, DJCapelis, impl, (+18 more, use /NETSPLIT to
show all of them)
07:52 -!- Netsplit over, joins: taruti, welterde, thomas_b, [Pete_27],
emmanueloga, [Eko], DJCapelis, zaero, Gertm, Broady (+4 more)
07:52 -!- Netsplit over, joins: res99, scoeri_, mpl, cco3-hampster, matti__, impl,
Tonnerre, kkress, suiside, falconindy (+3 more)
07:58 -!- rlab [~Miranda@91.200.158.34] has joined #go-nuts
08:01 -!- napsy [~luka@88.200.96.18] has joined #go-nuts
08:11 -!- tvw [~tv@212.79.9.150] has joined #go-nuts
08:13 -!- jmettraux [~jmettraux@211.19.55.177] has quit [Remote host closed the
connection]
08:13 -!- photron_ [~photron@port-92-201-165-189.dynamic.qsc.de] has joined
#go-nuts
08:17 -!- sahid [~sahid@LNeuilly-152-21-22-10.w193-253.abo.wanadoo.fr] has joined
#go-nuts
08:18 -!- dju [dju@fsf/member/dju] has joined #go-nuts
08:22 -!- dju [dju@fsf/member/dju] has quit [Max SendQ exceeded]
08:23 -!- dju [dju@fsf/member/dju] has joined #go-nuts
08:25 -!- ronnyy [~quassel@2001:6f8:12c6:1c86:224:1dff:fed7:9541] has quit [Remote
host closed the connection]
08:38 -!- femtoo [~femto@95-89-196-80-dynip.superkabel.de] has joined #go-nuts
08:41 -!- skejoe [~skejoe@188.114.142.231] has joined #go-nuts
08:59 -!- wrtp [~rog@92.17.69.32] has quit [Ping timeout: 245 seconds]
09:03 -!- wrtp [~rog@92.17.69.32] has joined #go-nuts
09:05 -!- napsy [~luka@88.200.96.18] has quit [Ping timeout: 276 seconds]
09:08 -!- rlab [~Miranda@91.200.158.34] has quit [Read error: Connection reset by
peer]
09:11 -!- virtualsue [~chatzilla@nat/cisco/x-uxcizjdpkclbmhkx] has joined #go-nuts
09:11 -!- napsy [~luka@88.200.96.18] has joined #go-nuts
09:17 -!- rlab [~Miranda@91.200.158.34] has joined #go-nuts
09:22 -!- mat_ [~mat@mx3.absolight.net] has quit [Read error: Connection reset by
peer]
09:22 -!- mat_ [~mat@mx3.absolight.net] has joined #go-nuts
09:25 < jnwhiteh> I would like to use encoding/binary with a struct that I
won't know the dimensions of until runtime.  Is there any way to make it work?
09:26 -!- wjlroe [~will@212.169.34.114] has joined #go-nuts
09:28 -!- mat__ [~mat@mx3.absolight.net] has joined #go-nuts
09:28 -!- mat_ [~mat@mx3.absolight.net] has quit [Read error: Connection reset by
peer]
09:30 -!- noktoborus_ [debian-tor@gateway/tor-sasl/noktoborus] has quit [Ping
timeout: 245 seconds]
09:30 -!- Zoopee [alsbergt@zoopee.org] has joined #go-nuts
09:43 -!- noktoborus_ [debian-tor@gateway/tor-sasl/noktoborus] has joined #go-nuts
09:45 < wrtp> jnwhiteh: you mean the dimensions of a slice within the
struct?
09:45 < jnwhiteh> yes
09:45 < jnwhiteh> I need to read data in blocks from the disk, but the block
size of the disk may vary.
09:46 < jnwhiteh> so the slice in the struct may need to be 16 length, 32,
etc.  depending on block size.
09:46 < wrtp> i haven't tried it, but can't you just do:
binary.Read(&myStruct{make([]int32, numWordsInBlock)})
09:46 < wrtp> or something like that?
09:46 < jnwhiteh> no, I don't believe you can
09:46 < jnwhiteh> but I'll try that
09:46 < jnwhiteh> I believe it looks at the TYPE of the struct, sees that it
doesn't have fixed length dimensions, and fails.
09:46 < jnwhiteh> that's what it did when I tried last time, but its
possible I messed something up.
09:47 -!- ucasano [~ucasano@95.231.79.71] has joined #go-nuts
09:48 < wrtp> you might well be right
09:49 < wrtp> hmm, looking at the source it seems as if a slice is only
allowed at the outermost level
09:51 -!- iomox [~ios@180.191.86.11] has joined #go-nuts
09:51 < wrtp> it looks as if it would be trivial to fix, but maybe there's a
deeper reason for the limitation
09:52 < jnwhiteh> possibly
09:55 < wrtp> the possible reason that i'm thinking of is that with the
current API, it would be possible for encoding/binary to "pre-compile" the size
and encoding instructions for a given type.  if slices are allowed anywhere, then
this optimisation is no longer possible for such types.
09:57 -!- mduft [~mduft@gentoo/developer/mduft] has left #go-nuts ["Leaving"]
10:04 -!- virtualsue [~chatzilla@nat/cisco/x-uxcizjdpkclbmhkx] has quit [Ping
timeout: 276 seconds]
10:13 < wrtp> jnwhiteh: you're probably best off dividing the block type
into two - the header and the data, and calling binary.Read twice.
10:14 -!- napsy [~luka@88.200.96.18] has quit [Ping timeout: 255 seconds]
10:14 < jnwhiteh> wrtp: there is no header
10:15 < wrtp> in which case you've got no problem.  just don't put your data
in a struct.
10:15 < jnwhiteh> oh, good point =)
10:15 < wrtp> or call binary.Read directly on the slice inside the struct
10:15 < jnwhiteh> I have a slice of structs, which I should still be able to
do
10:16 < wrtp> yup, that should work fine
10:16 < jnwhiteh> I'll work on that then, thanks for talking it though with
me =0
10:16 < wrtp> as long as the structs are fixed-size
10:16 < wrtp> np
10:16 < jnwhiteh> they always will be
10:18 -!- niemeyer [~niemeyer@201-40-152-123.pltce701.dsl.brasiltelecom.net.br]
has joined #go-nuts
10:19 < soul9> hi
10:19 -!- ucasano [~ucasano@95.231.79.71] has quit [Read error: Connection reset
by peer]
10:20 < soul9> anyone know how i can customize the binaries' path if i
include make.cmd
10:20 < soul9> ?
10:20 -!- ucasano [~ucasano@95.231.79.71] has joined #go-nuts
10:23 -!- zozoR [~zozoR@5634798d.rev.stofanet.dk] has quit [Quit: Morten.  Desu~]
10:25 < wrtp> soul9: write your own target?
10:29 -!- artefon [~thiago@189.59.182.83.dynamic.adsl.gvt.net.br] has joined
#go-nuts
10:30 -!- XenoPhoenix [~Xeno@cpc5-aztw24-2-0-cust39.aztw.cable.virginmedia.com]
has quit [Quit: Lost terminal]
10:31 -!- XenoPhoenix [~Xeno@cpc5-aztw24-2-0-cust39.aztw.cable.virginmedia.com]
has joined #go-nuts
10:32 -!- XenoPhoenix [~Xeno@cpc5-aztw24-2-0-cust39.aztw.cable.virginmedia.com]
has quit [Client Quit]
10:32 < soul9> wrtp: wish I knew what that meant.  i started using Makefiles
when I started coding in go, so up to now it's been simple..
10:33 < soul9> now i want to make some go commands that install into a bin
dir different from $GOROOT/bin
10:33 < soul9> do you know any place i could find some info/help for this?
10:35 -!- napsy [~luka@88.200.96.18] has joined #go-nuts
10:37 < wrtp> if you are prepared to use a different command than "make
install" e.g.  "make mybininstall" then you can add a couple of lines to your
makefile: mybininstall: $(MYBIN)/$(TARG)
10:37 -!- wjlroe [~will@212.169.34.114] has quit [Ping timeout: 240 seconds]
10:37 < wrtp> and $(MYBIN)/$(TARG): $(TARG)\n\tcp $(TARG) $(MYBIN)/$(TARG)
10:38 < wrtp> where \n\t mean a literal newline and tab
10:38 < wrtp> soul9: your best bet is to find a good gnu make tutorial
10:38 < soul9> yeah..
10:39 < soul9> haven't been able to yet.
10:39 -!- wjlroe [~will@212.169.34.114] has joined #go-nuts
10:44 -!- wrtp [~rog@92.17.69.32] has quit [Ping timeout: 252 seconds]
10:49 -!- wrtp [~rog@92.17.69.32] has joined #go-nuts
10:55 -!- ucasano [~ucasano@95.231.79.71] has quit [Quit: ucasano]
10:58 < jnwhiteh> wrtp: works great, thank you!
10:58 < wrtp> cool
10:59 < soul9> wrtp: i just include make.cmd in the beginning of the
makefile and overwrite GOBIN ;)
11:00 < wrtp> soul9: that seems fair enough
11:10 -!- HollyRain [~HollyRain@87.223.104.30] has joined #go-nuts
11:15 -!- falconin1y [~noclaf@unaffiliated/falconindy] has joined #go-nuts
11:15 -!- falconindy [~noclaf@unaffiliated/falconindy] has quit [Remote host
closed the connection]
11:15 -!- mpl [~mpl@smgl.fr.eu.org] has quit [Ping timeout: 272 seconds]
11:15 -!- mpl [~mpl@213.251.175.204] has joined #go-nuts
11:20 -!- felipe [~felipe@my.nada.kth.se] has joined #go-nuts
11:23 -!- cenuij [~cenuij@109.8.8.202] has joined #go-nuts
11:23 -!- cenuij [~cenuij@109.8.8.202] has quit [Changing host]
11:23 -!- cenuij [~cenuij@base/student/cenuij] has joined #go-nuts
11:28 -!- nighty^ [~nighty@x122091.ppp.asahi-net.or.jp] has joined #go-nuts
11:30 < serverhorror> is there an "easy" way to test a http handler func.  I
scanned the packages but couldn't quite find something obiouse that let's create a
http.Responsewriter and http.Request that I could passt to my handler function?
11:35 -!- bortzmeyer [~bortzmeye@batilda.nic.fr] has joined #go-nuts
11:36 -!- Fish- [~Fish@86.65.182.207] has quit [Ping timeout: 240 seconds]
11:55 < wrtp> serverhorror: i think you need to listen on a localhost port
and use ClientConn
11:57 < wrtp> although all the fields in http.Request are public and
ResponseWriter is an interface that you could easily implement yourself, so that's
another possible option too
12:15 -!- kanru [~kanru@61-30-10-70.static.tfn.net.tw] has quit [Ping timeout: 252
seconds]
12:15 -!- creack [~charme_g@163.5.84.215] has joined #go-nuts
12:21 -!- Project_2501 [~Marvin@82.84.80.32] has quit [Quit: E se abbasso questa
leva che succ...]
12:23 -!- tensorpudding [~user@99.148.202.191] has joined #go-nuts
12:24 -!- virtualsue [~chatzilla@nat/cisco/x-mshvjiwskfsmwsgu] has joined #go-nuts
12:26 -!- HollyRain [~HollyRain@87.223.104.30] has left #go-nuts []
12:28 < plexdev> http://is.gd/gb8Eu by [David Symonds] in go/src/pkg/net/ --
net: fix comment on Dial to mention unix/unixgram.
12:30 -!- andrebq [~andre@bhe201062224134.res-com.wayinternet.com.br] has joined
#go-nuts
12:37 -!- gabriel9 [~gabriel9@93.157.192.28] has quit [Ping timeout: 265 seconds]
12:37 < andrebq> when using Go tcp sockets, when the data is received and
after that a response i send, the protocol closes the tcp connection on the
server?
12:39 < andrebq> after sending the response back to the client, the server
send a TCP FIN packet and when the client tries to send more data an error is
throw
12:40 -!- iomox [~ios@180.191.86.11] has quit [Ping timeout: 245 seconds]
12:42 -!- ct529 [~quassel@77-44-78-159.xdsl.murphx.net] has joined #go-nuts
12:42 -!- gabriel9 [~gabriel9@93.157.192.28] has joined #go-nuts
12:50 < wrtp> andrebq: have you got some small example code that
demonstrates the problem?
12:54 < andrebq> wrtp, http://pastebin.com/k286A89k
12:56 < andrebq> the function HandleConnect is responsible to process the
received package and sent the response
12:57 < andrebq> after that, the client should be able to send DISCONNECT,
but it can't
12:58 -!- RobertLJ [~RobertLJ@c-68-44-163-191.hsd1.nj.comcast.net] has joined
#go-nuts
13:01 -!- devrim [~Adium@ip-95-223-189-66.unitymediagroup.de] has joined #go-nuts
13:02 < wjlroe> Anyone know why I get "intermediate certificate does not
have CA bit set" for an ssl connection when curl can get the same url without
issue?
13:03 <@adg_> wjlroe: maybe curl doesn't report the issue?
13:03 < wrtp> andrebq: that's not complete code.  for instance, i'd like to
see the definition of NewFrameFromCommand and CONNECT_CMD etc
13:03 -!- Soultaker [~Soultaker@hell.student.utwente.nl] has joined #go-nuts
13:04 -!- kanru [~kanru@118-160-163-40.dynamic.hinet.net] has joined #go-nuts
13:04 < wrtp> like, if NewFrameFromReader returns err!=nil, does it return a
valid frame.Command?
13:04 -!- Fish [~Fish@86.65.182.207] has joined #go-nuts
13:04 < wjlroe> adg: I'm trying to get verbose output from curl but seems to
connect fine
13:05 -!- Pablosan [9bbcb713@gateway/web/freenode/ip.155.188.183.19] has joined
#go-nuts
13:05 < wjlroe> adg: maybe it's the ca-bundle being different - but as I see
it, the error comes from validating the cert sent by the server
13:06 -!- ct529 [~quassel@77-44-78-159.xdsl.murphx.net] has quit [Remote host
closed the connection]
13:07 <@adg> hmm.  sorry, i don't know much about SSL
13:08 <@adg> you could try posting your question to golang-nuts if you
haven't already
13:08 < wjlroe> okey dokey
13:08 < andrebq> CONNECT_CMD is just a const that stands for "CONNECT"
13:10 < andrebq> NewFrameFromCommand receives a io.Reader and don't call
Close on it, That functions returns an empty Frame when there is a error
13:11 -!- SRabbelier [~SRabbelie@ip138-114-211-87.adsl2.static.versatel.nl] has
joined #go-nuts
13:13 < andrebq> wrtp, thanks for the help, but Dave ant golang-nuts group
found that I was closing the connection in the for.  Maybe I removed some *if*
that checked if the connection should be closed
13:17 < wrtp> oh yes, how did i miss that?!
13:19 -!- Rennex [rennex@giraf.fi] has joined #go-nuts
13:23 -!- andrebq [~andre@bhe201062224134.res-com.wayinternet.com.br] has quit
[Quit: Ex-Chat]
13:24 -!- plainhao [~plainhao@mail.xbiotica.com] has joined #go-nuts
13:30 -!- virtualsue [~chatzilla@nat/cisco/x-mshvjiwskfsmwsgu] has quit [Ping
timeout: 252 seconds]
13:39 < uriel> and this is another reason why we need a "The State of the
Gopher" talk:
http://www.reddit.com/r/programming/comments/du9zu/the_language_i_wish_go_was/c12yjyx
13:39 < uriel> (it was at +20 a moment ago)
13:41 -!- napsy [~luka@88.200.96.18] has quit [Ping timeout: 265 seconds]
13:41 -!- ExtraSpice [~XtraSpice@88.118.33.48] has quit [Quit: Leaving]
13:46 -!- Pablosan [9bbcb713@gateway/web/freenode/ip.155.188.183.19] has quit
[Quit: Page closed]
13:47 -!- artefon [~thiago@189.59.182.83.dynamic.adsl.gvt.net.br] has quit [Quit:
bye]
13:50 -!- devrim [~Adium@ip-95-223-189-66.unitymediagroup.de] has quit [Quit:
Leaving.]
13:54 -!- dj2 [~dj2@216.16.242.254] has joined #go-nuts
13:56 -!- Pablosan [9bbcf712@gateway/web/freenode/ip.155.188.247.18] has joined
#go-nuts
14:03 -!- virtualsue [~chatzilla@nat/cisco/x-dkfbdxbrepxqtaos] has joined #go-nuts
14:07 -!- Boney_ [~paul@124-168-80-178.dyn.iinet.net.au] has quit [Ping timeout:
252 seconds]
14:17 -!- femtoo [~femto@95-89-196-80-dynip.superkabel.de] has quit [Quit:
Leaving]
14:19 -!- RobertLJ [~RobertLJ@c-68-44-163-191.hsd1.nj.comcast.net] has quit [Ping
timeout: 264 seconds]
14:21 -!- RobertLJ [~RobertLJ@c-68-44-163-191.hsd1.nj.comcast.net] has joined
#go-nuts
14:22 -!- nsf [~nsf@jiss.convex.ru] has joined #go-nuts
14:22 -!- ExtraSpice [~XtraSpice@88.118.33.48] has joined #go-nuts
14:32 -!- eikenberry [~jae@ivanova.zhar.net] has quit [Quit: End of line.]
14:40 -!- kanru [~kanru@118-160-163-40.dynamic.hinet.net] has quit [Ping timeout:
240 seconds]
14:41 -!- qutron_xyxy [~xxx@mm-139-211-84-93.dynamic.pppoe.mgts.by] has joined
#go-nuts
14:41 -!- napsy [~luka@88.200.96.18] has joined #go-nuts
14:43 -!- kanru [~kanru@118-160-163-40.dynamic.hinet.net] has joined #go-nuts
14:45 -!- Project_2501 [~Marvin@82.84.80.32] has joined #go-nuts
14:48 -!- Soultaker [~Soultaker@hell.student.utwente.nl] has quit [Read error:
Connection reset by peer]
14:48 -!- Soultake1 [~Soultaker@hell.student.utwente.nl] has joined #go-nuts
14:54 -!- Soultake1 [~Soultaker@hell.student.utwente.nl] has quit [Read error:
Connection reset by peer]
14:55 -!- Soultaker [~Soultaker@hell.student.utwente.nl] has joined #go-nuts
14:58 -!- eikenberry [~jae@ivanova.zhar.net] has joined #go-nuts
14:59 -!- artefon [~thiagon@150.164.2.20] has joined #go-nuts
15:02 -!- Soultaker [~Soultaker@hell.student.utwente.nl] has quit [Read error:
Connection reset by peer]
15:03 -!- MX80 [~MX80@cust222.253.117.74.dsl.g3telecom.net] has joined #go-nuts
15:03 -!- Soultaker [~Soultaker@hell.student.utwente.nl] has joined #go-nuts
15:05 -!- Fish [~Fish@86.65.182.207] has quit [Remote host closed the connection]
15:06 -!- soul9 [~none@unaffiliated/johnnybuoy] has quit [Ping timeout: 264
seconds]
15:06 -!- qutron_xyxy [~xxx@mm-139-211-84-93.dynamic.pppoe.mgts.by] has quit [Ping
timeout: 265 seconds]
15:08 -!- uggedal [~uggedal@hydrogen.uggedal.com] has joined #go-nuts
15:13 -!- rejb [~rejb@unaffiliated/rejb] has joined #go-nuts
15:16 -!- Soultaker [~Soultaker@hell.student.utwente.nl] has quit [Ping timeout:
250 seconds]
15:18 -!- Soultaker [~Soultaker@hell.student.utwente.nl] has joined #go-nuts
15:21 -!- DerHorst [~Horst@e176098211.adsl.alicedsl.de] has joined #go-nuts
15:22 -!- bortzmeyer [~bortzmeye@batilda.nic.fr] has quit [Quit: Leaving.]
15:23 -!- Venom_X [~pjacobs@74.61.90.217] has joined #go-nuts
15:30 < plexdev> http://is.gd/gbnkz by [Russ Cox] in
go/src/pkg/encoding/binary/ -- encoding/binary: give LittleEndian, BigEndian
specific types
15:33 -!- Tv [~tv@cpe-76-168-227-45.socal.res.rr.com] has quit [Read error:
Operation timed out]
15:34 -!- Soultaker [~Soultaker@hell.student.utwente.nl] has quit [Read error:
Connection reset by peer]
15:34 -!- Soultaker [~Soultaker@hell.student.utwente.nl] has joined #go-nuts
15:42 -!- skelterjohn [~jasmuth@c-76-124-135-199.hsd1.nj.comcast.net] has quit
[Quit: skelterjohn]
15:44 -!- awidegreen [~quassel@62.176.237.78] has joined #go-nuts
15:46 -!- ronnyy [~quassel@2001:6f8:12c6:1c86:224:1dff:fed7:9541] has joined
#go-nuts
15:47 < plexdev> http://is.gd/gboNj by [Russ Cox] in 4 subdirs of
go/src/cmd/ -- ld: abandon symbol-driven archive loading
15:47 < plexdev> http://is.gd/gboNm by [Robert Griesemer] in
go/src/pkg/go/parser/ -- go/parser: consume auto-inserted semi when calling
ParseExpr()
15:58 -!- Soultake1 [~Soultaker@hell.student.utwente.nl] has joined #go-nuts
15:59 -!- Soultaker [~Soultaker@hell.student.utwente.nl] has quit [Ping timeout:
272 seconds]
15:59 -!- DerHorst [~Horst@e176098211.adsl.alicedsl.de] has quit [Remote host
closed the connection]
15:59 -!- saschpe [~quassel@77-23-177-40-dynip.superkabel.de] has joined #go-nuts
16:00 -!- iant [~iant@adsl-71-133-8-30.dsl.pltn13.pacbell.net] has quit [Ping
timeout: 276 seconds]
16:01 -!- nsf [~nsf@jiss.convex.ru] has quit [Quit: WeeChat 0.3.3]
16:01 -!- iant [~iant@216.239.45.130] has joined #go-nuts
16:01 -!- mode/#go-nuts [+v iant] by ChanServ
16:03 -!- mattn_jp [~mattn_jp@112-68-82-125f1.hyg1.eonet.ne.jp] has joined
#go-nuts
16:03 -!- Fish [~Fish@9fans.fr] has joined #go-nuts
16:04 -!- ronnyy [~quassel@2001:6f8:12c6:1c86:224:1dff:fed7:9541] has quit [Remote
host closed the connection]
16:06 -!- RobertLJ [~RobertLJ@c-68-44-163-191.hsd1.nj.comcast.net] has left
#go-nuts []
16:06 -!- napsy_ [~luka@88.200.96.18] has joined #go-nuts
16:06 -!- littlebobby [~bob@unaffiliated/littlebobby] has joined #go-nuts
16:06 -!- napsy [~luka@88.200.96.18] has quit [Read error: Connection reset by
peer]
16:07 -!- Tv [~tv@gige.bur.digisynd.com] has joined #go-nuts
16:19 -!- mattn_jp [~mattn_jp@112-68-82-125f1.hyg1.eonet.ne.jp] has left #go-nuts
[]
16:23 -!- cco3 [~conley@c-69-181-138-209.hsd1.ca.comcast.net] has quit [Ping
timeout: 265 seconds]
16:27 -!- vrasidakos [~lala@77.49.144.38.dsl.dyn.forthnet.gr] has quit [Ping
timeout: 250 seconds]
16:34 -!- skelterjohn [~jasmuth@lawn-gw.rutgers.edu] has joined #go-nuts
16:34 -!- virtualsue [~chatzilla@nat/cisco/x-dkfbdxbrepxqtaos] has quit [Ping
timeout: 272 seconds]
16:35 -!- krakensden [~krakensde@c-67-166-147-17.hsd1.ca.comcast.net] has joined
#go-nuts
16:37 -!- ShadowIce [~pyoro@unaffiliated/shadowice-x841044] has joined #go-nuts
16:40 -!- devrim [~Adium@ip-95-223-189-66.unitymediagroup.de] has joined #go-nuts
16:41 -!- Soultake1 [~Soultaker@hell.student.utwente.nl] has quit [Ping timeout:
272 seconds]
16:47 -!- Soultaker [~Soultaker@hell.student.utwente.nl] has joined #go-nuts
16:48 -!- saschpe [~quassel@77-23-177-40-dynip.superkabel.de] has quit [Remote
host closed the connection]
16:49 < plexdev> http://is.gd/gbtsx by [Russ Cox] in go/src/pkg/debug/gosym/
-- fix build
16:57 -!- lmoura [~lauromour@200.17.208.118] has joined #go-nuts
17:07 < plexdev> http://is.gd/gbuFp by [Russ Cox] in go/src/pkg/runtime/arm/
-- fix arm build
17:07 < plexdev> http://is.gd/gbuFv by [Russ Cox] in
go/src/pkg/runtime/windows/386/ -- fix windows build
17:08 -!- napsy [~luka@88.200.96.18] has quit [Ping timeout: 245 seconds]
17:09 -!- twittard [~acts_as@208.236.105.27] has left #go-nuts []
17:10 -!- virtualsue [~chatzilla@nat/cisco/x-tibfslvwlsirafjs] has joined #go-nuts
17:10 -!- Soultaker [~Soultaker@hell.student.utwente.nl] has quit [Ping timeout:
276 seconds]
17:11 -!- napsy [~luka@88.200.96.18] has joined #go-nuts
17:14 -!- virtualsue [~chatzilla@nat/cisco/x-tibfslvwlsirafjs] has quit [Ping
timeout: 240 seconds]
17:22 -!- dahankzter [~henrik@92-244-3-192.customers.ownit.se] has joined #go-nuts
17:24 -!- itrekkie [~itrekkie@ip72-200-114-232.tc.ph.cox.net] has quit [Quit:
itrekkie]
17:25 -!- dahankzter [~henrik@92-244-3-192.customers.ownit.se] has quit [Client
Quit]
17:29 -!- flix [~flix@APuteaux-152-1-49-24.w82-120.abo.wanadoo.fr] has joined
#go-nuts
17:35 -!- lmoura [~lauromour@200.17.208.118] has quit [Ping timeout: 252 seconds]
17:39 -!- lmoura [~lauromour@200.17.208.118] has joined #go-nuts
17:42 -!- zozoR [~zozoR@5634798d.rev.stofanet.dk] has joined #go-nuts
17:43 -!- skejoe [~skejoe@188.114.142.231] has quit [Quit: leaving]
17:44 -!- lmoura [~lauromour@200.17.208.118] has quit [Quit: Leaving]
17:47 -!- flix [~flix@APuteaux-152-1-49-24.w82-120.abo.wanadoo.fr] has quit [Ping
timeout: 276 seconds]
17:48 -!- flix [~flix@APuteaux-152-1-65-109.w82-120.abo.wanadoo.fr] has joined
#go-nuts
17:52 -!- dj2_ [~dj2@216.16.242.254] has joined #go-nuts
17:54 -!- araujo [~araujo@gentoo/developer/araujo] has joined #go-nuts
17:54 -!- dj2 [~dj2@216.16.242.254] has quit [Ping timeout: 240 seconds]
18:09 < plexdev> http://is.gd/gbA1r by [Rob Pike] in go/src/ -- Make.inc:
delete unnecessary -no-inline flag to quietgcc
18:13 -!- res99 [~anonymous@201.237.130.70] has quit [Quit: res99]
18:15 -!- napsy [~luka@88.200.96.18] has quit [Ping timeout: 240 seconds]
18:24 -!- res99 [~anonymous@201.237.130.70] has joined #go-nuts
18:34 -!- enherit [~enherit@cpe-98-149-170-48.socal.res.rr.com] has joined
#go-nuts
18:34 -!- tvw [~tv@212.79.9.150] has quit [Read error: Connection reset by peer]
18:35 -!- scm [scm@d038144.adsl.hansenet.de] has quit [Ping timeout: 252 seconds]
18:37 -!- scm [scm@d165241.adsl.hansenet.de] has joined #go-nuts
18:39 -!- femtoo [~femto@95-89-196-80-dynip.superkabel.de] has joined #go-nuts
18:41 -!- IRWolfie- [irwolfie@ircnoob.com] has quit [Ping timeout: 265 seconds]
18:53 -!- zozoR [~zozoR@5634798d.rev.stofanet.dk] has quit [Quit: Morten.  Desu~]
18:59 -!- bnjmn [~bnjmn@siegel.dreamhost.com] has joined #go-nuts
19:00 -!- femtooo [~femto@95-89-196-80-dynip.superkabel.de] has joined #go-nuts
19:00 < bnjmn> what's the point of wrigint a bunch of concurrent Go code
when the GC sucks with multiple processors
19:01 < artefon> bnjmn, why to u say that?
19:02 -!- Project_2501 [~Marvin@82.84.80.32] has quit [Quit: E se abbasso questa
leva che succ...]
19:03 -!- femtoo [~femto@95-89-196-80-dynip.superkabel.de] has quit [Ping timeout:
255 seconds]
19:04 < uriel> because trolling is fun?
19:05 < kimelto> ;)
19:08 < skelterjohn> because the language has to come before the
implementation
19:09 < kimelto> I remember that a GC rewrite was planned (at the time of
the first release).  what the status?  is it in the tree?
19:09 < skelterjohn> there are a bunch of people working on go full time,
and the gc is one of their highest priorities
19:13 < wrtp> bnjmn: concurrent code is useful even when you don't need
speedup through parallelism.
19:13 -!- napsy [~luka@88.200.96.18] has joined #go-nuts
19:18 -!- XenoPhoenix [~Xeno@cpc5-aztw24-2-0-cust39.aztw.cable.virginmedia.com]
has joined #go-nuts
19:25 -!- dj2 [~dj2@216.16.242.254] has joined #go-nuts
19:25 -!- dj2_ [~dj2@216.16.242.254] has quit [Ping timeout: 240 seconds]
19:28 -!- gabriel9 [~gabriel9@93.157.192.28] has quit [Read error: Connection
reset by peer]
19:32 -!- nsf [~nsf@jiss.convex.ru] has joined #go-nuts
19:36 -!- goosebumper [3bb38272@gateway/web/freenode/ip.59.179.130.114] has joined
#go-nuts
19:37 < goosebumper> Anybody know of any stemming implementation in Go?
19:38 -!- virtualsue [~chatzilla@93-97-62-8.zone5.bethere.co.uk] has joined
#go-nuts
19:40 < skelterjohn> what is a stemming implementation?
19:41 < goosebumper> any stemming library such as a porter stemmer
19:42 < skelterjohn> still no clue what you're talking about :) CS is such a
wide field...
19:44 < skelterjohn> ahhh a stemmer (after a brief google search)
19:44 < goosebumper> I'd agree.  Stemmer is used in NLP..stemmer reduces
words to roots..such as Catty and Catter to Cat.
19:44 < goosebumper> yup, a stemmer :)
19:44 < skelterjohn> i actually do know what those are.  in fact i wrote one
in java maybe 10 years ago
19:45 < skelterjohn> i haven't done one in go, though
19:45 < skelterjohn> but it was very simple - and had lots of false
positives
19:45 < skelterjohn> just a find/replace that only looked at the end of a
word
19:46 < skelterjohn> i used the table that came with wordnet
19:46 < goosebumper> Yeah, I've been using a porter stemmer in python for
long.  So i think my best bet would be to write one in Go myself and post here
19:46 < goosebumper> All my work is NLP, so a lot of scratch work in Go :)
19:46 < goosebumper> wordnet used to be THE thing at one time
19:47 < skelterjohn> i worked for wordnet a couple summers when i was an
undergrad
19:47 < skelterjohn> i'm in a different field, now.  explaining why i didn't
remember what a stemmer was
19:48 < goosebumper> yup, i guessed that.  what you work on?
19:48 < goosebumper> *in
19:48 < skelterjohn> for wordnet i did internal tools to help lexicographers
make additions to the database
19:48 < skelterjohn> specifically linking verbs and nouns that were derived
from one another back together
19:49 < skelterjohn> now i do machine learning, spec.  reinforcement
learning
19:50 < goosebumper> sweet.  i am surprised to see an ML professional
interested in Go.
19:50 < skelterjohn> strictly amateur
19:50 < skelterjohn> i'm a phd student
19:51 < skelterjohn> i like go because i like good programming languages
19:51 < skelterjohn> and inference is hard - having a language that makes
parallelization easier is a good thing for ML
19:52 < goosebumper> well i am an undergrad but i am working on a natural
language search startup
19:53 < skelterjohn> difficult problem
19:53 < goosebumper> i know.  thats the whole point of working on it :)
19:54 < skelterjohn> what clever new idea is this startup bringing to the
market?
19:57 -!- gmilleramilar [~gmiller@184-106-207-119.static.cloud-ips.com] has quit
[Remote host closed the connection]
19:58 < goosebumper> greedily map natural language to semantics.  i figured
out a way to attach semantic metadata to every webpage
19:58 < goosebumper> interface wise, you speak a question in your mobile, it
speaks back the exact answer
19:59 < skelterjohn> how exactly do you intend to map natural language to
semantics?
19:59 < skelterjohn> just sense disambiguation alone is hard
19:59 < skelterjohn> and that's not even the whole problem you're discussing
20:00 -!- Soultaker [~Soultaker@hell.student.utwente.nl] has joined #go-nuts
20:03 < goosebumper> thats a hack.  i intended to solve two different
problems with one shot.  i use a really flawed language translation model and an
occurrence based clustering on top of it.  then statistically perform sanity
checks.
20:03 < goosebumper> we end up with rich correctness data of language
translation as well as decently extracted semantics, good enough to answer
questions aske in natural language
20:05 -!- kanru1 [~kanru@118-168-234-250.dynamic.hinet.net] has joined #go-nuts
20:06 < goosebumper> from what i see with TREC questions, its oing things
just fine...i dont know how good a reflection is trec performance
20:06 < skelterjohn> i don't know what TREC is
20:07 < skelterjohn> the thing that worries me when you say "greedily map X
to Y" is the number of different Xs and Ys when you're talking about the internet
20:07 < skelterjohn> things that might work in a proof-of-concept operation
might not scale well
20:07 -!- plainhao [~plainhao@mail.xbiotica.com] has quit [Quit: plainhao]
20:07 -!- kanru [~kanru@118-160-163-40.dynamic.hinet.net] has quit [Ping timeout:
252 seconds]
20:08 -!- rboyd [~rboyd@72-161-203-70.dyn.centurytel.net] has quit [Quit: rboyd]
20:08 -!- soul9 [~none@em.r0x0r.me] has joined #go-nuts
20:08 -!- soul9 [~none@em.r0x0r.me] has quit [Changing host]
20:08 -!- soul9 [~none@unaffiliated/johnnybuoy] has joined #go-nuts
20:10 < goosebumper> the mapping happens for each query.  for a natural
language question asked by a user, the 'X' is usally very small (< 4).  I know
this is hard to scale, but addressing this issue while crawling is a bigger issue
for a startup..too many resources needed.
20:12 < goosebumper> TREC is a conference i think.  they have large
collection of Q/A data for AI bots testing..
20:14 -!- rboyd [~rboyd@72-161-203-70.dyn.centurytel.net] has joined #go-nuts
20:19 -!- Eridius [~kevin@unaffiliated/eridius] has joined #go-nuts
20:21 -!- Boney [~paul@203-214-44-173.dyn.iinet.net.au] has joined #go-nuts
20:34 < soul9> hi
20:34 < soul9> i have a question about channels
20:34 < goosebumper> yup, go ahead
20:34 < soul9> if i want to return the stuff of my function using channels
20:34 < soul9> basically it has a signature in which it states that it takes
a sending channel as arguments
20:35 < soul9> and then passes the return values through that channel
20:35 < soul9> the problem is, that i can't seem to do it using
uni-directional channels
20:35 < soul9> i.e.: paste comming..
20:35 -!- wjlroe [~will@212.169.34.114] has quit [Ping timeout: 240 seconds]
20:35 < skelterjohn> don't paste into irc
20:35 < skelterjohn> use pastebin
20:36 < soul9> http://pastie.org/1239087
20:36 < skelterjohn> phew
20:36 < soul9> yeah.
20:36 < soul9> ;)
20:36 < soul9> this is basically the unix "wc"
20:36 < skelterjohn> so what's going wrong with that function?
20:36 < skelterjohn> nothing jumps out as stupid when i skim it :)
20:36 < wrtp> soul9: your channels are going in the wrong direction
20:37 < soul9> the thing is i don't know how to create a channel suitable
for it in the calling function
20:37 < soul9> gosh, ok, gimme a sec
20:37 < wrtp> in Count(...), ch should be ch chan<- int64 etc
20:37 < skelterjohn> there we go
20:37 < skelterjohn> wrtp: have you done any more of that rog-go stuff with
drawing?
20:37 < soul9> wrtp: then in the calling function i make(chan <-int64)?
20:39 < soul9> this doesn't work either
20:40 < skelterjohn> what is the error you're seeing
20:40 < soul9> http://pastie.org/1239102
20:41 < skelterjohn> in your original function, make a bidirectional channel
20:41 < skelterjohn> it can be treated as a unidirectional channel in the
called function
20:41 < soul9> yeah..
20:41 < skelterjohn> but the caller needs to be able to 1) receive on it and
2) pass it off as a send-only
20:42 < skelterjohn> so it needs to be make(chan int64)
20:42 < soul9> :(
20:42 < soul9> yep
20:42 < skelterjohn> and the param to Count can be chan<-int64
20:42 < skelterjohn> why the frownie?  this should work fine
20:42 < soul9> but why do i need to make a bidirectional channel if all i
want is receive through it?
20:43 -!- flix [~flix@APuteaux-152-1-65-109.w82-120.abo.wanadoo.fr] has quit
[Remote host closed the connection]
20:43 < soul9> not sure how the logic goes
20:43 < skelterjohn> but that isn't all you want to do
20:43 < bnjmn> what's the point of wrigint a bunch of concurrent Go code
when the GC sucks with multiple processors
20:43 < skelterjohn> you want to send to it from somewhere else
20:43 < skelterjohn> bnjmn: do you just say that every hour or so?  we
answered you last time
20:43 -!- femtoo [~femto@95-89-196-80-dynip.superkabel.de] has joined #go-nuts
20:43 < bnjmn> nobody gave a good answer
20:43 < soul9> yeah, i guess
20:43 < bnjmn> i just don't get it
20:44 < skelterjohn> bnjmn: you didn't give a good question.  that's why you
respond to the answers with a more precise question, etc
20:44 < bnjmn> it's like rah rah rah concurrency oh but....  yea, sucks on
multiple cpus?  wha?
20:44 < bnjmn> the GC sucks what's up with that
20:44 <+iant> bnjmn: 1) Concurrency is not *only* about performance
20:44 < skelterjohn> you clearly didn't read my answer
20:44 <+iant> bnjmn: 2) The GC is not fixed in its current state; it will
improve
20:44 < bnjmn> but i've heard for years that GCs will improve
20:44 < skelterjohn> years?
20:44 <+iant> Go has been public for less than 1 year
20:44 < bnjmn> i feel like a retard for believing that for like a decade
20:45 < skelterjohn> this language has been public for less than a year
20:45 < cbeck> GC has improved...
20:45 < bnjmn> yeah so what kind of magic GC fairy dust is Go going to have
20:45 <+iant> Java garbage collectors have improved enormously in the last
few years
20:45 < nsf> mono 2.8 has a new GC as well
20:45 < bnjmn> yea they went from horrendous to sucky
20:45 < skelterjohn> bnjmn: so your problem isn't with go's GC, it's with
GCs + concurrency in general?
20:45 <+iant> bnjmn: or just GC in general?
20:45 < skelterjohn> and when you say something is "bad" there has to be
something "better" for that to be a valid observation
20:45 < skelterjohn> "bad" is a relative term
20:45 < bnjmn> my problem is that GCs seem to suck, in general, i'd be happy
to use one if there was a good one
20:46 <+iant> what do you do in a concurrent program without a GC? And how
do you keep it reliable?
20:46 < bnjmn> i love GC, i love how easier it is to write code with a GC,
but if it's slow then what's the point if i need it to be fast
20:46 < bnjmn> how do i keep it reliable??  i manage my own memory...
20:46 < bnjmn> it sucks but i do it
20:46 < bnjmn> that's all
20:46 <+iant> if you need to run at the fastest possible machine
performance, then clearly GC is not the right approach
20:46 < bnjmn> i can't say it's a fun experience
20:46 -!- femtooo [~femto@95-89-196-80-dynip.superkabel.de] has quit [Ping
timeout: 265 seconds]
20:46 -!- IRWolfie- [irwolfie@ircnoob.com] has joined #go-nuts
20:46 < goosebumper> GC's seem to suck -> i love GC. You make NLP a hard
problem.
20:47 < skelterjohn> people used to say this during the transition from
assembly to higher level languages
20:47 <+iant> managing your own memory in a concurrent program is much
harder than in a single-threaded program
20:47 < bnjmn> i love gc if speed doesn't matter
20:47 <+iant> but, yes, it can be done
20:47 < bnjmn> sorry if that is too nuanced for you
20:47 <+iant> you are looking for a different language
20:47 < bnjmn> i manage memory just fine, i lock stuff if i have to
20:47 <+iant> in fact, you are looking for assembly code
20:47 < bnjmn> i guess i'm looking for C huh
20:47 < cbeck> ...
20:47 <+iant> that locking is a performance penalty in itself, you know
20:47 < bnjmn> why would i use assembly
20:47 < bnjmn> yea i know
20:47 <+iant> that will give you the fastest possible code
20:47 < bnjmn> it's better than the GC locking all over the place
20:48 <+iant> if runtime performance is the only important thing you need
20:48 < bnjmn> i don't know assembly well enough
20:48 < bnjmn> to do the fastest possible
20:48 < bnjmn> no it's a tradeoff
20:48 <+iant> but even the current the GC doesn't lock all over the place
20:48 <+iant> right, it's a tradeoff
20:48 <+iant> exactly
20:48 < bnjmn> but the GCs i see, for the code speeds i need, they are not
good enough
20:48 <+iant> OK, you need a different language
20:48 < bnjmn> anyway the locking is pretty minimal
20:48 < bnjmn> ok what do you recommend sir
20:48 <+iant> your tradeoffs are different
20:48 <+iant> I recommend assembly
20:48 < bnjmn> ok deal
20:48 < bnjmn> what
20:48 <+iant> if you rule that out, I recommend C
20:48 < bnjmn> no i am not smart enough for that
20:49 < bnjmn> ok
20:49 < soul9> heh
20:49 < bnjmn> that's better
20:49 <+iant> it's not a matter of intelligence
20:49 <+iant> it's just another language
20:49 < bnjmn> it's a matter of intelligence, time, memory
20:49 < skelterjohn> no, i believe him about that
20:49 < bnjmn> no it's true
20:49 < bnjmn> i don't have the memory or the intelligence to know all the
optimizaitons
20:49 < bnjmn> and i don't have the time to write it out like that so
20:49 < bnjmn> im not good enough for that
20:49 < skelterjohn> why do you think you do for memory management?
20:49 -!- rlab [~Miranda@91.200.158.34] has quit [Quit: Miranda IM! Smaller,
Faster, Easier.  http://miranda-im.org]
20:49 < wrtp> bnjmn: what sort of stuff are you doing if performance is so
crucial?  (out of interest)
20:49 < bnjmn> because if i don't the GC fucks up the performance?
20:50 < bnjmn> network processing at wire speed
20:50 < skelterjohn> no, why do you think you know all the optimizations for
memory management
20:50 < bnjmn> i don't
20:50 < bnjmn> i'm saying the C compiler can do a better job than me
20:50 < bnjmn> for opts like that
20:50 < skelterjohn> the C compiler doesn't do memory management for you
20:50 < goosebumper> "Premature optimization is the root of all evil."
20:50 < bnjmn> i'm not smarter than the C compiler sometimes
20:51 < bnjmn> skelterjohn i think you lost the plot, iant was recommending
asm to me
20:51 < bnjmn> please pay attention
20:51 < skelterjohn> 1) you're being fairly rude and 2) i prefaced what i
said with the context i meant it in
20:51 < skelterjohn> so please pay attention to that.
20:52 -!- Pablosan [9bbcf712@gateway/web/freenode/ip.155.188.247.18] has left
#go-nuts []
20:52 < bnjmn> well you said "< skelterjohn> no, i believe him about
that" when i was talking about my intelligence
20:52 < skelterjohn> i was also being rude.
20:53 < bnjmn> so maybe you should do some introspection and take a break
from irc if it is stressing you out
20:53 < goosebumper> i think he meant that for you saying that assembly is
hard
20:53 < skelterjohn> no, he understood me
20:53 < goosebumper> ok :)
20:53 -!- rbraley [~rbraley@ip72-222-128-245.ph.ph.cox.net] has joined #go-nuts
20:54 < skelterjohn> i just find it irritating when someone comes and says
"your way sucks because i think it doesn't work for the task i want to do"
20:54 < bnjmn> listen what irritates me is how Go seems to be marketed
20:55 < bnjmn> so i hear about it and it sounds great and then oh
20:55 -!- araujo [~araujo@gentoo/developer/araujo] has quit [Quit: Leaving]
20:55 < bnjmn> oh...  i thought...  it was going to be suitable for me...
and...  it's not
20:55 < bnjmn> super
20:55 -!- teralaser [~teralaser@unaffiliated/teralaser] has joined #go-nuts
20:55 -!- teralaser [~teralaser@unaffiliated/teralaser] has left #go-nuts []
20:55 <+iant> did you read the bit about how Go generates the fastest
possible running code?
20:55 < bnjmn> um no
20:55 <+iant> or did you read the bit about how Go is some 10% to 20% behind
C?
20:55 <+iant> because we only wrote the latter
20:55 < goosebumper> Go is marketed as a sensible systems language.  Not
natural language assembly.
20:55 <+iant> and it sounds like you want the former
20:56 < bnjmn> no i read the part that says "Its concurrency mechanisms make
it easy to write programs that get the most out of multicore and networked
machines"
20:56 < bnjmn> how it gets the most out of multicore is frankly beyond my
comprehension
20:56 < bnjmn> i guess i just don't get it, i'm an idiot or something
20:56 <+iant> you're right, you have to read "most" in conjunction with
"easy"
20:57 < bnjmn> funny how that works
20:57 <+iant> I apologize on the Go developer's behalf for deceiving you
20:57 <+iant> we will return your money
20:57 < bnjmn> that's very gracious of you
20:57 <+iant> it's nothing
20:57 <+iant> really
20:57 < bnjmn> would you like my mailing address
20:57 < nsf> all 0.00$ you've paid them, hehe
20:58 < goosebumper> nsf: nice sn.  you scared me for a sec.  :)
20:58 < skelterjohn> anyway, "garbage collected" is one of the things go
advertises
20:59 -!- jcao219 [~jcao219@pool-96-226-238-248.dllstx.fios.verizon.net] has
joined #go-nuts
20:59 < skelterjohn> so if your problem is that you think that good code for
multi-core can't be done with garbage collection, then you should have known that
go was not your language from the start
21:00 < bnjmn> i don't know if it can or not
21:00 < bnjmn> so far i haven't seen it
21:00 < bnjmn> doesn't mean it can't be done
21:00 < skelterjohn> it means your problem isn't with the language go
21:00 < bnjmn> it's with the implementation of go
21:00 < skelterjohn> you should gripe in #garbageCollectionForMultiCore
21:00 < nsf> I don't understand why people are complaining about free and
open source project..  like they saw an advertisement, then they've paid their
money and now here we go..  like a consumer habit or something
21:00 < bnjmn> which, last time i checked, is what actually matters
21:01 <+iant> we all agree that the garbage collector needs improvement
21:01 <+iant> and we all agree that Go is not yet a complete product
21:01 -!- twittard [~acts_as@208.236.105.27] has joined #go-nuts
21:01 < skelterjohn> then we can get back to my response an hour ago - the
gc and the rest of the language is a work in progress
21:05 < bnjmn> well make sure to let me know when it's ready
21:07 -!- dj2 [~dj2@216.16.242.254] has quit [Remote host closed the connection]
21:08 -!- artefon [~thiagon@150.164.2.20] has quit [Quit: Leaving]
21:08 < goosebumper> btw if there are python coders here.  how do you feel
about Go? Not performance, I am asking purely on language aesthetics.
21:09 -!- ShadowIce` [pyoro@unaffiliated/shadowice-x841044] has joined #go-nuts
21:10 -!- ShadowIce [~pyoro@unaffiliated/shadowice-x841044] has quit [Read error:
Connection reset by peer]
21:10 < skelterjohn> aesthetically there are a lot of similarities
21:10 < skelterjohn> i used to write a lot of python, but not so much lately
21:11 < skelterjohn> i find type safety to be a big bonus to my productivity
21:11 < skelterjohn> and python's lack of it drove me away
21:12 < bnjmn> i've done python and i agree about the type stuff and also
the GIL kills concurrent threads
21:13 < skelterjohn> what is GIL?
21:13 < goosebumper> ok, i agree to both points.  i miss lack of curly
braces, type safety is actually refreshing..brings back the high school java
memories.  i dont appreciate type safety that much, however.  still refreshing.
21:13 < bnjmn> the global interpreter lock
21:13 < bnjmn> it's an implementation detail
21:13 -!- tav [~tav@92.7.82.61] has quit [Ping timeout: 245 seconds]
21:13 < skelterjohn> oh that might be the other thing that drove me off
python - i could never get it do use more than one processor at a time
21:13 < goosebumper> python and threads dont go all that well
21:14 < bnjmn> exactly..
21:14 < Tonnerre> s/ and threads//
21:15 < KirkMcDonald> As someone who writes Python for a living, I can tell
you it is possible to utilize more than one core.
21:15 < KirkMcDonald> You just need to use multiple processes.
21:15 < Tonnerre> I know it is
21:15 -!- wjlroe [~will@78-86-14-131.zone2.bethere.co.uk] has joined #go-nuts
21:16 < goosebumper> it is possible.  its not that good.  python was never
my choice tool for anything multi-core.  even our web server doesn't use thread.
its like NASCAR in F1 race.  not good
21:16 < bnjmn> right.  multiple processes.  and what if i want to share
state?
21:17 < KirkMcDonald> Pfft, state.
21:17 < bnjmn> oh...thats not how it's done anymore.  right.
21:17 < Tonnerre> bnjmn, man shm
21:17 < bnjmn> just import multiprocessing huh
21:17 < goosebumper> i dont want a share state
21:17 < bnjmn> a web server is mostly IO. python code, ...could be anything
21:18 < bnjmn> so yes i would love if python would actually be able to do
threads properly
21:20 -!- tav [~tav@92.7.91.84] has joined #go-nuts
21:20 < goosebumper> web server for me wasn't just IO, it was a long-polling
engine.  so i had to implement a lot of async handling.  but i like python in its
process oriented state.
21:20 < bnjmn> so the async handling...what...is it doing...other than I/O?
21:21 < bnjmn> like how involved does the computation get
21:21 -!- ikaros [~ikaros@dslb-084-059-072-178.pools.arcor-ip.net] has joined
#go-nuts
21:21 < bnjmn> i bet it is pretty straightforward otherwise one computation
that wasn't interruptable would kill everything else
21:22 < bnjmn> since you just have one thread
21:22 < Tonnerre> Multiple processes can share a port
21:22 < goosebumper> edge cases.  closed broken connections.  shutting tasks
taking too long.  monitoring task state.  calculating if task is taking or wud
take longer (eg: stuck on some REST API), then shut it and re-initiate.
21:23 < bnjmn> ok
21:23 -!- ShadowIce` [pyoro@unaffiliated/shadowice-x841044] has quit [Quit:
Verlassend]
21:24 < goosebumper> its funny, a lot of code i wrote...in Go, its a damn
goroutine :\
21:24 < goosebumper> 1 line in Go :)
21:25 < skelterjohn> you couldn't write one function,
launchThisInAThread(foo, params) in python?
21:28 < goosebumper> i didnt want threads.  i still dont want them until its
completely hidden (like in Go).  i want proccesses an task queues.  i do bad
things with threads.
21:28 < goosebumper> *and task queues
21:28 < KirkMcDonald> skelterjohn: threading.Thread(target=foo, args=params)
21:28 -!- Fish [~Fish@9fans.fr] has quit [Remote host closed the connection]
21:28 < skelterjohn> there you go
21:29 < KirkMcDonald> (Then call .start() on the Thread object.)
21:31 -!- adu [~ajr@pool-173-66-10-221.washdc.fios.verizon.net] has joined
#go-nuts
21:32 < goosebumper> in a long-polled server environment, you want to know
exactly how many open connections can a box support.  also a thread per open
connection is insanely resource hogging.  what i do is queue up task and then let
background proccesses do them in sequence.  an open connection is like 2KB in
memory..
21:34 < Namegduf> I would write longpolling separately to most other things,
and have it terminate the "connection's" goroutine quickly after putting the
longpolling connection on a queue, yeah.
21:34 < Namegduf> Then I could use a goroutine-per-connection model in
general, but efficiently handle that.
21:37 -!- Boney [~paul@203-214-44-173.dyn.iinet.net.au] has quit [Ping timeout:
255 seconds]
21:39 -!- Boney [~paul@203-217-71-205.dyn.iinet.net.au] has joined #go-nuts
21:41 < scyth> KirkMcDonald, you can't use threads in python
21:42 < scyth> where by "threads" i mean nptl/posix threads
21:42 < KirkMcDonald> scyth: Sure you can.
21:42 < KirkMcDonald> They sure are.
21:42 < KirkMcDonald> Well, pthreads.
21:42 < KirkMcDonald> And on Windows, it uses native Windows threads.
21:43 < KirkMcDonald> scyth: The issue is that it uses this big ol' lock
around everything.
21:43 < scyth> Threading lib doesn't use pthreads, or at least it didn't
when I tried it
21:43 < KirkMcDonald> scyth: It has used pthreads since time immemorial.
21:44 < scyth> KirkMcDonald, I'm pretty sure I saw different pids for
threads on process list
21:44 -!- femtooo [~femto@95-89-196-80-dynip.superkabel.de] has joined #go-nuts
21:46 < KirkMcDonald> scyth: I am not sure you saw what you thought you saw.
21:47 < scyth> KirkMcDonald, could be
21:48 -!- femtoo [~femto@95-89-196-80-dynip.superkabel.de] has quit [Ping timeout:
250 seconds]
21:48 < scyth> maybe I misread ppid/pid columns
21:49 -!- rboyd [~rboyd@72-161-203-70.dyn.centurytel.net] has quit [Quit: rboyd]
21:50 -!- Soultaker [~Soultaker@hell.student.utwente.nl] has quit [Ping timeout:
240 seconds]
21:52 -!- rboyd [~rboyd@72-161-203-70.dyn.centurytel.net] has joined #go-nuts
21:52 -!- rboyd [~rboyd@72-161-203-70.dyn.centurytel.net] has quit [Client Quit]
21:54 -!- araujo [~araujo@gentoo/developer/araujo] has joined #go-nuts
21:55 -!- femtooo [~femto@95-89-196-80-dynip.superkabel.de] has quit [Quit:
Leaving]
21:56 -!- ikaros [~ikaros@dslb-084-059-072-178.pools.arcor-ip.net] has quit [Quit:
Leave the magic to Houdini]
21:57 -!- napsy [~luka@88.200.96.18] has quit [Quit: Lost terminal]
22:00 -!- zerd [~quassel@tor.zerd.net] has quit [Ping timeout: 264 seconds]
22:01 -!- wrtp [~rog@92.17.69.32] has quit [Quit: wrtp]
22:01 -!- skelterjohn [~jasmuth@lawn-gw.rutgers.edu] has quit [Quit: skelterjohn]
22:03 -!- zerd [~quassel@tor.zerd.net] has joined #go-nuts
22:06 -!- virtualsue [~chatzilla@93-97-62-8.zone5.bethere.co.uk] has quit [Ping
timeout: 265 seconds]
22:10 -!- rinzai [~rinzai@host86-158-180-200.range86-158.btcentralplus.com] has
joined #go-nuts
22:10 -!- lmoura [~lauromour@189.26.29.175.dynamic.adsl.gvt.net.br] has joined
#go-nuts
22:13 -!- DerHorst [~Horst@e176098211.adsl.alicedsl.de] has joined #go-nuts
22:34 < scyth> KirkMcDonald, just checked and remembered what was happening
:) for true paralelism I was refered to multiprocessing module, which indeed forks
processes, because threading GIL makes threads pointless for paralelism
22:36 -!- gzmask [~ray@corwin.cat.uregina.ca] has quit [Quit: gzmask]
22:39 -!- awidegreen [~quassel@62.176.237.78] has quit [Remote host closed the
connection]
22:41 < KirkMcDonald> scyth: Ah, yes.
22:43 -!- DerHorst [~Horst@e176098211.adsl.alicedsl.de] has quit [Remote host
closed the connection]
22:47 -!- slashus2 [~slashus2@74-141-107-118.dhcp.insightbb.com] has joined
#go-nuts
22:53 < soul9> yea the gil in python is retarded, yes threads in python can
not be used for parallelization, at most asynchronous operations (ie launch
blocking routines in new threads), and yes, with multiprocessing comes the problem
that processes on unix are expensive
22:54 < soul9> so you have to juggle with multiprocessing and threading to
be really effective
22:54 < soul9> in go this is done automatically :)
22:57 < KirkMcDonald> Multiprocessing is not the only way to use multiple
processes.
22:58 < soul9> i guess you can communicate over sockets
22:58 < soul9> or, even better, /dev/shm
22:58 < soul9> or are you talking about alternative threading
implementations?
22:58 < KirkMcDonald> I'm thinking of making a good ol' fashioned preforking
server.
22:59 < KirkMcDonald> s/thinking of/talking about/
22:59 < soul9> isn't that...a process?
22:59 < KirkMcDonald> Yes?  I was talking about multiple processes.
23:00 -!- adu [~ajr@pool-173-66-10-221.washdc.fios.verizon.net] has quit [Quit:
adu]
23:00 < soul9> i don't get it
23:00 < soul9> are you talking about the multiprocessing module in python?
23:01 < KirkMcDonald> I was when I used the word "multiprocessing," yes.
23:01 < soul9> ok
23:01 < soul9> still, processes aren't cheap on unix
23:01 < KirkMcDonald> In what sense?
23:02 < soul9> in the sense that it doesn't make sense to spawn thousands of
them very quickly ;)
23:02 < KirkMcDonald> Nah.  You spawn a fixed number of workers.
23:02 < goosebumper> a single process can have 1000s of open connections.
23:02 < soul9> yes, exactly
23:02 < soul9> that is what i meant
23:03 < soul9> then you start jiggling with processes, and threads in them
23:03 < soul9> which go automatically does for you
23:04 < goosebumper> if you need to.  or you can just have a stack of tasks
which another army of processes complete asynchronously
23:04 < soul9> ok, true.
23:05 < goosebumper> its not the best way, its way to avoid threads which i
religiously do :D
23:05 < soul9> ok, i see where KirkMcDonald was getting to, i guess
23:05 < soul9> yeah, me too heheh
23:13 -!- felzix [~rdavidson@65.203.131.194] has joined #go-nuts
23:13 < felzix> Is Go a good language for using agents such as those found
in Clojure?
23:15 < felzix> er, by 'good' I mean, is it fairly easy and efficient?
Goroutines using closures sounds like it'd do it but I don't know how realistic
that is for a large number of goroutines
23:16 -!- devrim [~Adium@ip-95-223-189-66.unitymediagroup.de] has quit [Quit:
Leaving.]
23:17 < cbeck> goroutines are very cheap.  I don't think I've ever worked
with a parralellization contruct that consumed fewer resources.
23:17 < cbeck> parallelleilllazation
23:18 < cbeck> Someday I'll spell that right
23:19 < felzix> ok.  thanks!
23:21 -!- jmettraux [~jmettraux@211.19.55.177] has joined #go-nuts
23:22 -!- skejoe [~skejoe@188.114.142.231] has joined #go-nuts
23:25 -!- photron_ [~photron@port-92-201-165-189.dynamic.qsc.de] has quit [Quit:
Leaving]
23:34 -!- skelterjohn [~jasmuth@c-76-124-135-199.hsd1.nj.comcast.net] has joined
#go-nuts
23:41 -!- dj2 [~dj2@CPE001f5b35feb4-CM0014048e0344.cpe.net.cable.rogers.com] has
joined #go-nuts
23:41 -!- pvarga [~pvarga@pool-72-88-229-188.nwrknj.east.verizon.net] has joined
#go-nuts
23:43 -!- rinzai [~rinzai@host86-158-180-200.range86-158.btcentralplus.com] has
quit [Quit: Leaving]
23:51 -!- ExtraSpice [~XtraSpice@88.118.33.48] has quit [Ping timeout: 240
seconds]
23:58 -!- wjlroe [~will@78-86-14-131.zone2.bethere.co.uk] has quit [Quit: Computer
has gone to sleep]
23:58 < plexdev> http://is.gd/gbYzb by [Andrew Gerrand] in go/doc/ --
playground: s/it/this/
--- Log closed Fri Oct 22 00:00:12 2010