--- Log opened Sun May 15 00:00:49 2011
--- Day changed Sun May 15 2011
00:00 < rael_wiki_> in the language specification there is an example of
select that I don't understand
http://golang.org/doc/go_spec.html#Select_statements
00:02 < rael_wiki_> I mean, if we put that select in a for loop, we
eventually get to know on the third case that c3 is closed but our select will
keep executing that case, right?
00:11 -!- Tuller [~tuller@c-69-143-48-210.hsd1.va.comcast.net] has joined #go-nuts
00:22 -!- jhawk28 [~jhawk28@user-387c58d.cable.mindspring.com] has joined #go-nuts
00:24 -!- gtaylor [~gtaylor@99-5-124-9.lightspeed.gnvlsc.sbcglobal.net] has joined
#go-nuts
00:29 -!- rlab_ [~Miranda@91.200.158.34] has quit [Read error: Connection reset by
peer]
00:41 < Ekspluati> rael_wiki_: Executing as in checking ig there's data
coming?
00:44 < rael_wiki_> Ekspluati: executing the body of the case, the checking
of data coming should result true since a read on a closed channel should return a
zero value
00:44 < rael_wiki_> Ekspluati: am I wrong?
00:45 < Ekspluati> rael_wiki_: Oh right, that's true.
00:45 -!- hallas [~hallas@x1-6-30-46-9a-b2-c5-1f.k891.webspeed.dk] has left
#go-nuts []
00:46 < rael_wiki_> Ekspluati: so I don't know how to manage a for loop with
many channels with some of them that might be eventually closed
00:46 < Ekspluati> That means the default should never execute.
00:46 < rael_wiki_> exactly
00:49 < rael_wiki_> Ekspluati: it would be nice to have a select with many
channels that keeps reading them until none of them is open and then executes the
default, exiting
00:51 -!- pearle [~pearle@blk-224-181-222.eastlink.ca] has quit [Ping timeout: 246
seconds]
00:53 -!- mwarning [~mwarning@ip-178-202-97-131.unitymediagroup.de] has joined
#go-nuts
00:53 < mwarning> Hi, may I ask what this program prints?
http://www.reddit.com/r/programming/comments/hb6m8/google_io_2011_writing_web_apps_in_go/c1u1xj6
00:54 < mwarning> I tried to compile the go compiler, but got some error I
couldn't fix right away.
00:56 < Ekspluati> "if(blah()) <newline> {" will become "if(blah());
{"
00:56 < Ekspluati> So the if statement is basically ignored.
00:56 < mwarning> ah, that's the problem.  thanks a lot
00:57 < mwarning> I wonder why if(..); isn't invalid?
00:59 < mwarning> nvm, apparently it doesn't compile
00:59 -!- jhawk28 [~jhawk28@user-387c58d.cable.mindspring.com] has quit [Quit:
Linkinus - http://linkinus.com]
01:01 -!- Tuller [~tuller@c-69-143-48-210.hsd1.va.comcast.net] has quit [Quit:
Computer has gone to sleep.]
01:01 -!- mwarning [~mwarning@ip-178-202-97-131.unitymediagroup.de] has quit
[Quit: Leaving]
01:12 -!- breadlorde [~dreadlord@c-24-11-39-160.hsd1.mi.comcast.net] has quit
[Ping timeout: 250 seconds]
01:17 < skelterjohn> Ekspluati, rael_wiki_: the default case also fires if
you have open channels that don't have something ready to send, immediately
01:17 < skelterjohn> it's not only if all the chans are closed, but that
will also trigger default
01:21 < rael_wiki_> skelterjohn: I agree on the fact that default case is
executed if open channels don't have data to send at the moment, but a closed
channel actually has data to send immediately (the zero-value for the specified
type) so it shouldn't execute the default
01:39 < fzzbt> how do i do length matching with go's regexp?  i can't seem
to get the {m,n} syntax working
01:39 < fzzbt> eg.  [a-z]{1,10} to match 10 character words
01:40 < fzzbt> it's not implemented?
01:42 < moraes> fzzbt, no
01:42 < moraes> fzzbt, but this one implements it:
http://code.google.com/p/sre2/
01:42 < moraes> http://code.google.com/p/re2/wiki/Syntax
01:43 < moraes> nitpick: that doesn't match 10, matches 1 to 10
01:43 < fzzbt> yeah :0.  i'll have to switch then
01:49 -!- Ekspluati [5b9a0bcf@gateway/web/freenode/ip.91.154.11.207] has quit
[Ping timeout: 252 seconds]
02:02 -!- fheller [~hellerf@c-24-9-120-132.hsd1.co.comcast.net] has quit [Remote
host closed the connection]
02:02 -!- rael_wiki_ [~chatzilla@87.18.214.175] has quit [Quit: ChatZilla 0.9.86.1
[Iceweasel 3.5.19/20110430164311]]
02:03 -!- pearle [~pearle@blk-224-181-222.eastlink.ca] has joined #go-nuts
02:26 < tsp> Why won't this relatively simple program compile under gccgo?
http://paste.pocoo.org/show/389151/
02:27 < tsp> readlines.go:9:23: error: reference to undefined field or
method ‘ReadLine’ -- it works just fine under 8g.
02:29 -!- danilo04 [~danilo04@129.21.100.41] has joined #go-nuts
02:30 -!- GutenLinux [~chatzilla@125.117.86.199] has joined #go-nuts
02:31 -!- GutenLinux [~chatzilla@125.117.86.199] has quit [Client Quit]
02:33 < edsrzf> tsp: Probably different versions of the Go core packages
02:33 < edsrzf> ReadLine is a relatively recent addition
02:34 < tsp> Can I update gccgo's core somehow?
02:35 < edsrzf> If you built it, it should be pretty easy to at least add in
that one method
02:35 < edsrzf> Other newer parts might not go in so easily
02:37 < tsp> Is there a changelog for the go trunk somewhere?
02:42 -!- quag [~quag@121-98-81-61.bitstream.orcon.net.nz] has joined #go-nuts
02:43 < quag> I've built a small go program and it currently has a horrible
build.sh script to build it instead of using Makefiles and packages.
02:43 < quag> It's up on https://github.com/quag/mcobj/tree/develop
02:44 < quag> is there a tutorial on how to build a program like this?
02:44 < quag> Ideally, I'd like multiple packages for different bits.
02:44 < quag> There is already one package there (nbt.go) and it is making
switching over to the normal cmd Makefile difficult.
02:53 -!- rcrowley [~rcrowley@c-71-202-44-233.hsd1.ca.comcast.net] has quit [Quit:
Computer has gone to sleep.]
02:53 -!- gaiusp [~gaiusparx@bb219-75-41-232.singnet.com.sg] has left #go-nuts []
02:55 -!- ExsysTech [~ExsysTech@50-46-213-60.evrt.wa.frontiernet.net] has quit
[Read error: Connection reset by peer]
02:56 -!- jemes [~jemeshsu@bb219-75-41-232.singnet.com.sg] has joined #go-nuts
02:56 -!- ExsysTech [~ExsysTech@50-46-213-60.evrt.wa.frontiernet.net] has joined
#go-nuts
02:57 -!- jemes [~jemeshsu@bb219-75-41-232.singnet.com.sg] has quit [Client Quit]
02:59 -!- jemeshsu [~jemeshsu@bb219-75-41-232.singnet.com.sg] has joined #go-nuts
03:03 -!- blackmagik [~blackmagi@unaffiliated/blackmagik] has joined #go-nuts
03:04 < steven> hi guys
03:04 < fzzbt> hey steve
03:04 < steven> do you have any idea how to notice when a file's contents
have changed in Go code?
03:04 < quag> hi steven
03:04 < steven> like, how to observe that?
03:05 < quag> steven: stat the file every so often and check if the modified
timestamp has changed?
03:06 < steven> i like that solution
03:06 < quag> normally, I store the current timestamp and then check to see
if the new timestamp is different.  (Not >)
03:06 < steven> but theres nothing platform-independent like inotify?
03:06 < quag> not to my knowledge.
03:10 < quag> if I recall correctly, windows has an api that fires an event
when anything in a directory changes, but not specific files.
03:10 < quag> But that may have changed
03:10 < quag> don't recall what osx has
03:10 < steven> i know mac has one, FSEvents
03:10 < steven> i wish there was some wrapper around it tho
03:10 < steven> since its C-based and cgo is tough
03:10 -!- boscop_ [~boscop@f055153169.adsl.alicedsl.de] has joined #go-nuts
03:11 -!- angasule_ [~angasule@190.2.33.49] has quit [Read error: Connection reset
by peer]
03:11 < quag> steven: could you call out to a commandline program to watch
the files?
03:13 -!- boscop [~boscop@f055049000.adsl.alicedsl.de] has quit [Ping timeout: 246
seconds]
03:14 < steven> hmm interesting idea
03:14 < steven> thanks
03:15 -!- breadlorde [~dreadlord@c-24-11-39-160.hsd1.mi.comcast.net] has joined
#go-nuts
03:23 -!- nteon [~nteon@c-98-210-195-105.hsd1.ca.comcast.net] has joined #go-nuts
03:25 -!- chomp [~chomp@c-67-186-35-69.hsd1.pa.comcast.net] has joined #go-nuts
03:26 -!- manveru [~manveru@b08s28ur.corenetworks.net] has joined #go-nuts
03:26 < manveru> heya
03:26 < quag> hi
03:26 < manveru> i'm trying to implement a line-based protocol server
03:27 -!- hq1 [~hq@ns1.insant.pl] has quit [Read error: Connection reset by peer]
03:27 < manveru> lines are separated by \r\n
03:27 < manveru> i have one command that acts like a header, giving the
number of bytes to follow until the next \r\n considered eol
03:28 < manveru> right now i make a []byte with the given size and simply
use bufio.Reader.Read with that buffer
03:28 < manveru> is that an ok approach?
03:29 < quag> depends on how large the data will be
03:29 < quag> if, say, it was 2GB, you may find yourself struggling for
memory
03:29 < manveru> heh
03:29 < manveru> no, shouldn't be that big :)
03:29 < manveru> i expect up to a few mb
03:29 < quag> could nasty people target your system?
03:29 < quag> :-)
03:30 < manveru> yeah
03:30 < manveru> well, gotta cap it for sure
03:30 < quag> what will you do with the bytes afterwards?
03:31 < chomp> heh sooo...  wrapping a C API whose use requires function
pointers for callbacks.  any good documentation on handling this?
03:31 < manveru> i store them in a job struct in a priority ordered
red-black tree until someone requests a job
03:31 < chomp> i've seen talk of //export but haven't found a clear example
or description of what it actually does
03:31 < manveru> i'll def.  also have to handle out-of-memory errors
03:31 < manveru> if the job queue is too big
03:31 < quag> so you're not going to spill to disk always?
03:32 < manveru> no
03:32 < quag> a simple way to handle this would be to use io.Copyn() to
write it to a file, then pull it back when the job needs it.
03:32 < quag> and rely on the OS caching files for speed :-)
03:33 < quag> if you're going to keep it all in memory anyway, then
allocating the []byte and reading into it seems like the way to go.
03:33 -!- hq1 [~hq@ns1.insant.pl] has joined #go-nuts
03:33 < manveru> that might be a nice feature later on...  but not something
i need for small queues right now
03:33 < manveru> ok...
03:33 < manveru> then the next issue i have
03:33 < quag> how large are these pieces of data?
03:34 < quag> heh
03:34 < manveru> how do i keep reading from the connection?
03:34 < manveru> without blocking everything else
03:34 < quag> do the read in a goroutine?
03:34 < manveru> i cannot find select or epoll...
03:34 < manveru> hm, yeah
03:35 < manveru> go server.accept(listener)
03:35 < manveru> basically that
03:36 < quag> yup
03:36 < manveru> but i don't know how to find out if the other side
disconnected or sent more data
03:37 < quag> handle each socket in its own goroutine
03:37 < quag> they'll block waiting for more data
03:38 < manveru> hm, ok
03:39 < quag> looking at http://golang.org/src/pkg/net/fd_linux.go
03:39 < manveru> maybe i need a continuous reader...
03:39 < quag> it looks like on linux it uses epoll for implementation anyway
03:39 < manveru> i use NewReaderSize right now
03:40 < danilo04> is there a binding to access usb devices?
03:41 < quag> danilo04: http://go-lang.cat-v.org/library-bindings lists
go-pkg-usb but that 404s
03:41 < danilo04> quag, yeah that is what I found
03:41 -!- blackmagik [~blackmagi@unaffiliated/blackmagik] has quit [Read error:
Connection reset by peer]
03:41 < danilo04> quag: thanks anyway
03:42 < quag> what are you wanting to do with usb?
03:42 < manveru> guess NewReader is good enough
03:43 < danilo04> maybe play with a kinect device
03:44 < quag> cool :-)
03:45 < danilo04> quag: can I import C and make a wrapper of libusb
03:45 < danilo04> quag: I am really new in go
03:45 < quag> danilo04: I haven't played with the cgo stuff.  Sorry.
03:45 < danilo04> ok
03:56 < nteon> danilo04: I'm new as well, but the go gtk bindings make heavy
use of cgo, as well as dealing with marshalling go callbacks to c
03:56 < nteon> chomp: ^^
03:56 < nteon> both of you I guess
03:57 < chomp> i found the life example in misc/cgo, it appears to be useful
after all :p
03:57 < danilo04> nteon: sounds good, let me check
03:57 < chomp> i'm wrapping PAM btw.  good times, good times :p
03:58 < nteon> hah
03:58 < nteon> nice
03:59 < manveru> :)
03:59 * manveru is working on a beanstalkd
03:59 < chomp> soon i'll have my httpsh server in go.  soon...
04:00 < chomp> going to serve a js/canvas vt100 emulator with a login shell,
over websockets
04:00 < skelterjohn> add a button for scp too :)
04:00 < chomp> hrh yeah i actually thought about that
04:01 < skelterjohn> would be a very useful little tool
04:01 < manveru> better than those java ones anyway :P
04:01 < chomp> it's next on my agenda
04:02 < chomp> i actually have a working shell server now, but it uses a ssh
client on the server side for authentication
04:02 < manveru> but isn't canvas really slow for text?
04:02 < chomp> vt100 emulator works pretty well though.  i can vim in my
browser
04:02 < chomp> yeah, im separating the rendering out and will be building a
plain old DOM terminal renderer
04:02 < manveru> cool
04:02 < chomp> canvas was quick n dirty
04:03 < manveru> i built a terminal emulator in tk before
04:03 < manveru> termcap was such a pita
04:03 < manveru> only vt100 would be nice :)
04:03 < chomp> one step at a time :)
04:03 < manveru> heh
04:04 < manveru> i gave up when i had to implement scrolling regions
04:04 < chomp> right now the extent of my tty control involves sending stty
commands over the wire upon successful login ><
04:04 < chomp> i got scrolling regions in!  :)
04:04 < manveru> w007
04:04 < chomp> vim really works beautifully
04:04 < manveru> yeah, that's where i encountered them
04:04 < chomp> and yes, that was a huge PITA.  documentation is so sparse
04:04 < manveru> couldn't for the life of me find docs for how they should
behave
04:05 < chomp> i basically just reverse engineered the CSI sequences coming
from about 10 different console applications
04:05 < manveru> even read half the urxvt source :P
04:05 < chomp> after getting all i could out of any "real" documentation
04:06 -!- alc [~arx@222.128.132.65] has joined #go-nuts
04:06 < manveru> well, at least it's a pretty efficient protocol
04:07 < chomp> for a computer, indeed :)
04:08 < manveru> please lemme know when you get a demo running :)
04:09 < nteon> chomp: that sounds sweet
04:09 < manveru>
https://github.com/manveru/yonde/blob/master/lib/yonde/terminfo.rb
04:09 < nteon> I'm working on an sshd implementation
04:10 < manveru> that's how i get termcap/terminfo
04:10 < chomp> nice
04:10 < manveru> sorry, not go
04:10 < nteon> manveru: beanstalk daemon or client (or both)
04:10 < manveru> daemon
04:10 < nteon> nice
04:10 < manveru> i figured goroutines are made for it :)
04:11 < nteon> oh, theres already a beanstalk go client :)
04:11 < manveru> there are clients for just about any language...
04:11 < manveru> it's a really easy protocol
04:11 < manveru> no XML crap like AMQP
04:13 < nteon> hah
04:13 < nteon> xml is a disease
04:14 < manveru> ok...  my prototype seems to work...
04:15 -!- rejb [~rejb@unaffiliated/rejb] has quit [Disconnected by services]
04:15 -!- rejb [~rejb@unaffiliated/rejb] has joined #go-nuts
04:17 < manveru> hm
04:17 < manveru> well, i do wish they'd use JSON instead of YAML in
beanstalk
04:21 < manveru> can't find a yaml lib...
04:22 < justinlilly> oh the fun of a new language ^_^
04:22 < manveru> oh well, the C beanstalkd builds the YAML by hand as well
04:27 -!- breadlorde [~dreadlord@c-24-11-39-160.hsd1.mi.comcast.net] has quit
[Ping timeout: 240 seconds]
04:28 -!- breadlorde [~dreadlord@c-24-11-39-160.hsd1.mi.comcast.net] has joined
#go-nuts
04:33 < nteon> heh, nice, TCP_NODELAY defaults to true in Go :)
04:33 -!- photron [~photron@port-92-201-244-191.dynamic.qsc.de] has joined
#go-nuts
04:33 < vsmatck> That's weird.
04:44 < nteon> vsmatck: idk, for most applications these days that makes
sense.  its the first thing i do when making a networked application in C, Python
or Java
04:48 -!- pearle [~pearle@blk-224-181-222.eastlink.ca] has quit [Quit: Leaving]
04:51 < nteon> does anyone know if there is a reason go doesn't support
python style slice notation: slice[:-1] in python is aSlice[:len(aSlice)-1]
04:52 -!- rcrowley [~rcrowley@c-71-202-44-233.hsd1.ca.comcast.net] has joined
#go-nuts
04:52 < edsrzf> Probably because it would penalize the performance of the
common case (when signed variables are involved, anyway)
04:53 < edsrzf> I remember somebody asking about it on the mailing list
once.  Or maybe it was an issue.  I can't remember.
04:53 < nteon> it seems like it would be a somewhat trivial change to detect
the unary op '-' in the high part of the slice expression and replace it with
len(aSlice) - ...  at compile tiem
04:53 < nteon> edsrzf: hmm, I briefly searched and didnt find it, I will try
looking harder
04:54 < edsrzf> Yes, but if you have an expression like slice[a:b], you have
to generate extra code.
04:54 -!- rcrowley [~rcrowley@c-71-202-44-233.hsd1.ca.comcast.net] has quit
[Client Quit]
04:54 < edsrzf> Since you don't know statically if a or b is negative.
04:55 < manveru> well, it def.  is not a search/replace issue :)
04:55 < nteon> edsrzf: i would be happy restricting it to compile-time
negative ints
04:56 < edsrzf> nteon: Here's Rob talking about it a little bit:
http://groups.google.com/group/golang-nuts/msg/0b0af95059d4dc11
04:57 < manveru> at least job insertion works now...
05:00 < nteon> edsrzf: thanks, that makes sense.  I suppose its not that big
a deal :)
05:01 -!- ios_ [~ios@180.191.131.160] has joined #go-nuts
05:04 < nteon> i still like the idea of allowing compile time negative
integer constants, but I see how that could add confusion
05:04 < edsrzf> There are a few other relevant posts in that thread
05:05 < manveru> hm
05:05 < manveru> how would you implement a priority queue?
05:06 < manveru> WP says a tree would work fine, but i fear that would give
me locking issues
05:07 < manveru> channels have FIFO order, so i cannot easily translate it
into that either
05:08 < edsrzf> manveru: There's the package container/heap, which you could
use for a priority queue, but it won't take care of synchronization for you.
05:10 < manveru> hm
05:11 < manveru> i guess i could sequencialize the access through a channel
05:11 < manveru> that way i don't have to lock manually
05:11 < |Craig|> heaps through channels really seems like the right approach
05:12 < manveru> ok...  out of battery, will tackle that tonight
05:12 < manveru> thanks guys
05:25 -!- danilo04 [~danilo04@129.21.100.41] has quit [Read error: Operation timed
out]
05:27 < chomp> so call me crazy, but bool is most certainly a Go type, yes?
05:28 < edsrzf> chomp: Yes
05:28 < chomp> rhetorical question reallay, of course it is
05:28 < chomp> but i'm trying to compile some code, say func foo()
(*C.char,bool) { // ...
05:28 < chomp> and i am getting an "unrecognized Go type *ast.Ident"
05:28 < chomp> which disappears if i remove the bool
05:28 -!- danilo04 [~danilo04@rm2348358874.student.rit.edu] has joined #go-nuts
05:29 < edsrzf> That kind of sounds like a cgo bug
05:29 < chomp> was thinking as much
05:30 < chomp> yep, looks like specifically a cgo bug with exporting bool
types
05:31 < chomp> or signatures that use bool types
05:31 < chomp> int it is, then!
05:35 -!- quag [~quag@121-98-81-61.bitstream.orcon.net.nz] has quit [Remote host
closed the connection]
05:35 -!- quag [~quag@121-98-81-61.bitstream.orcon.net.nz] has joined #go-nuts
05:35 -!- nteon [~nteon@c-98-210-195-105.hsd1.ca.comcast.net] has quit [Ping
timeout: 240 seconds]
05:42 -!- quag [~quag@121-98-81-61.bitstream.orcon.net.nz] has quit [Ping timeout:
240 seconds]
05:45 -!- hq1 [~hq@ns1.insant.pl] has quit [Ping timeout: 250 seconds]
05:46 -!- breadlorde [~dreadlord@c-24-11-39-160.hsd1.mi.comcast.net] has quit
[Ping timeout: 240 seconds]
05:48 -!- danilo04 [~danilo04@rm2348358874.student.rit.edu] has quit [Quit:
Saliendo]
05:50 -!- quag [~quag@121-98-81-61.bitstream.orcon.net.nz] has joined #go-nuts
05:57 -!- quag [~quag@121-98-81-61.bitstream.orcon.net.nz] has quit [Ping timeout:
240 seconds]
05:58 -!- zozoR [~Morten@2906ds2-arno.0.fullrate.dk] has joined #go-nuts
05:59 -!- quag [~quag@121-98-81-61.bitstream.orcon.net.nz] has joined #go-nuts
06:12 -!- jamesmiller5 [~jamesmill@184.17.105.74] has quit [Remote host closed the
connection]
06:17 -!- alc [~arx@222.128.132.65] has quit [Ping timeout: 248 seconds]
06:27 -!- ShadowIce [~pyoro@unaffiliated/shadowice-x841044] has joined #go-nuts
06:36 -!- ios_ [~ios@180.191.131.160] has quit [Quit: Leaving]
06:40 -!- monsieur_max [~maxime@ip-78.net-89-2-171.rev.numericable.fr] has joined
#go-nuts
06:41 < chomp> is there any way to cheat and do some pointer arithmetic?
i've got some data from C code and i essentially need to walk through an array of
*C.char
06:42 < chomp> ah of course i'd figure it out after asking
06:43 < chomp> foo := (uintptr)(unsafe.Pointer(C.terrible_C_function()));
foo += (uintptr)(unsafe.Sizeof(foo))
06:50 -!- jeremy_c_ [~jeremy@cowgar.com] has joined #go-nuts
06:53 -!- hq1 [~hq@ns1.insant.pl] has joined #go-nuts
06:57 -!- jeremy_c_ [~jeremy@cowgar.com] has quit [Quit: leaving]
07:04 -!- gtaylor [~gtaylor@99-5-124-9.lightspeed.gnvlsc.sbcglobal.net] has quit
[Quit: gtaylor]
07:17 -!- chomp [~chomp@c-67-186-35-69.hsd1.pa.comcast.net] has quit [Quit:
Leaving]
07:24 -!- heatxsink [u956@gateway/web/irccloud.com/x-ngylfklxhmstihlb] has joined
#go-nuts
07:24 < heatxsink> are there any examples of using a filewatcher like thing
in go?
07:25 < vsmatck> There is inotify on linux.  The Go API is very similar to
the C API.
07:26 < heatxsink> ah but there's no high level abstraction for other oses?
07:27 < vsmatck> File system events are a hard problem from the kernel
developer standpoint.  There is not a good common subset of functionality between
all major OS's such that high level abstraction could be created.
07:27 < vsmatck> Inotify is one of the better ones.
07:29 < heatxsink> vsmatck: dumb question does inotify work on osx?  last i
knew it was only for linux
07:30 < vsmatck> It's only for linux as far as I know.  A thing like that is
very tied to a specific kernel.
07:30 -!- Cobi [~Cobi@2002:1828:88fb:0:aede:48ff:febe:ef03] has quit [Ping
timeout: 264 seconds]
07:31 < heatxsink> vsmatck: k thx.
07:32 < vsmatck> I made a setup a while ago in C++ that did continual
scanning that checked for modification dates using c++/boost.  Sort-of yuck.
07:32 < vsmatck> But it was portable.  :-/
07:33 < str1ngs> heatxsink: there is a package for inotify in the stdlib if
you use linux
07:34 < heatxsink> vsmatck: nice ...  was it efficient?
07:34 < heatxsink> vsmatck: i'm trying to write something that is portable
across osx and linux that does that...
07:35 < heatxsink> vsmatck: i've already wrote it in C# using mono but i
kinda wanna ditch any dependancy on C# that's all..
07:36 < vsmatck> It did ok.  It was basically a while true, with a sleep in
it.  I'd disable the sleep after finding a modified file because the files I was
monitoring tended to be modified in groups.
07:37 < heatxsink> ah
07:37 < vsmatck> It was a p2p program share scanner.
07:37 -!- piranha [~piranha@5ED43A0B.cm-7-5a.dynamic.ziggo.nl] has joined #go-nuts
07:38 < manveru> heatxsink: osx has kqueue i think
07:39 < heatxsink> manveru: ya that's what mono is invoking under the hood
07:39 < manveru> you could use libevent, it works on both
07:40 < heatxsink> i was kinda blown away at the whole differences in
unix'es but I guess it makes sense that everyone for very specific system things
always go their own way
07:40 < manveru> but of course adds a dependency...
07:40 < manveru> https://secure.wikimedia.org/wikipedia/en/wiki/Libevent
07:44 < manveru> damn, i could swear it had stuff for file handles too
07:44 < manveru> guess not, then :(
07:44 < justinlilly> str1ngs: if I can talk you into outlining the work that
remains in your libgit2 bindings, I may help you.  I need the ability to traverse
something akin to git-log, which I don't think your version currently supports.
07:45 < justinlilly> ...  not that I know how to work with the cgo stuff or
have ever worked with libgit2...  but necessity and all that.
07:45 -!- Cobi [~Cobi@2002:1828:88fb:0:aede:48ff:febe:ef03] has joined #go-nuts
07:46 -!- |Craig| [~|Craig|@panda3d/entropy] has quit [Quit: |Craig|]
07:47 < str1ngs> justinlilly: hello
07:47 < str1ngs> currently yes it does support Rev walking
07:48 < str1ngs> the api is abit low level I'll put something together to
demostrate
07:48 < heatxsink> manveru: balls.
07:49 < justinlilly> str1ngs: yea.  I think I can kinda get the idea.  Must
have glanced over the revwalk bits.
07:49 -!- alc [~arx@222.128.132.65] has joined #go-nuts
07:49 < str1ngs> justinlilly: its modeled after the libgit api which is not
idea.  and I've sorta been selective about what I have included
07:50 -!- alc [~arx@222.128.132.65] has quit [Remote host closed the connection]
07:51 < str1ngs> justinlilly: for etc C object I bassically wrap it in a go
struct ie Oid = Oid.git_oid
07:51 -!- iant [~iant@216.239.45.130] has quit [Ping timeout: 276 seconds]
07:51 < str1ngs> but let me put together an example of a RevWalk
07:51 < justinlilly> see that.  Makes it a bit difficult to know what the
struct contains.
07:52 < str1ngs> not really Oid is a oid just the C parts are hidden
07:53 -!- hq1 [~hq@ns1.insant.pl] has left #go-nuts ["WeeChat 0.3.3"]
07:56 -!- ImpatientSpoon [~ashley@115.6.233.220.static.exetel.com.au] has joined
#go-nuts
07:56 < str1ngs> justinlilly: http://libgit2.github.com/api.html#revwalk
07:59 < justinlilly> I suppose the biggest issue I was having is that "push"
doesn't seem to jive with my thinking of traversing the data structure.
08:00 < str1ngs> justinlilly: make sure you pull I just merged my devel
branch to master
08:03 < str1ngs> justinlilly: push pretty much gives it a place to start
08:04 < str1ngs> for the Test I use head oid
08:06 -!- photron [~photron@port-92-201-244-191.dynamic.qsc.de] has quit [Quit:
Leaving]
08:08 < str1ngs> justinlilly: its not very go idium but my plan was sorta to
do a direct api mapping the raise the level of the api and make it more go
idiomatic
08:09 < str1ngs> justinlilly: let me put a example Rev walker together for
you.  if you could add a Readme that might be a start?
08:09 < str1ngs> basic install instructions I guess
08:09 < justinlilly> k.  just trying to map out what I think a revwalker
should look like.  I'll probably get to testing & installing tomorrow
08:10 -!- iant [~iant@adsl-71-133-8-30.dsl.pltn13.pacbell.net] has joined #go-nuts
08:10 -!- mode/#go-nuts [+v iant] by ChanServ
08:11 < str1ngs> justinlilly: awesome thanks
08:13 < str1ngs> justinlilly: I do live far from you btw
08:13 < str1ngs> dont*
08:13 < justinlilly> orly?  Neat.
08:13 < str1ngs> yes just north of the border
08:15 -!- dfr|mac [~dfr|work@ool-182e3fca.dyn.optonline.net] has quit [Remote host
closed the connection]
08:24 -!- wrtp [~rog@92.17.93.207] has joined #go-nuts
08:27 -!- araujo [~araujo@gentoo/developer/araujo] has quit [Quit: Leaving]
08:29 < str1ngs> justinlilly: https://gist.github.com/972976
08:29 < str1ngs> a basic walk
08:29 < str1ngs> I think revwalk.Next should return a Oid not use the
pointer..  off hand
08:36 -!- huin [~huin@91.85.171.238] has joined #go-nuts
08:37 -!- mehalelal [~androirc@76.103.175.11] has quit [Quit: AndroIRC]
08:38 -!- rlab [~Miranda@91.200.158.34] has joined #go-nuts
08:42 -!- jemeshsu [~jemeshsu@bb219-75-41-232.singnet.com.sg] has quit [Quit:
jemeshsu]
08:42 -!- ucasano [~ucasano@host185-90-static.5-79-b.business.telecomitalia.it]
has joined #go-nuts
08:44 -!- Project_2501 [~Marvin@dynamic-adsl-94-36-161-95.clienti.tiscali.it] has
joined #go-nuts
08:49 -!- mehalelal [~androirc@94.sub-174-253-244.myvzw.com] has joined #go-nuts
08:56 -!- jemeshsu [~jemeshsu@bb219-75-41-232.singnet.com.sg] has joined #go-nuts
09:06 -!- hallas [~hallas@x1-6-30-46-9a-b2-c5-1f.k891.webspeed.dk] has joined
#go-nuts
09:06 < hallas> hey all
09:07 -!- firwen [~firwen@2a01:e34:eea3:7e10:4a5b:39ff:fe51:e8ae] has joined
#go-nuts
09:07 < str1ngs> hello
09:22 -!- Mehak [~androirc@76.103.175.11] has joined #go-nuts
09:25 -!- mehalelal [~androirc@94.sub-174-253-244.myvzw.com] has quit [Ping
timeout: 248 seconds]
09:25 -!- TheMue [~TheMue@p5DDF4DC8.dip.t-dialin.net] has joined #go-nuts
09:26 -!- Loonacy [~loonacy@c-67-172-248-248.hsd1.ut.comcast.net] has quit [Ping
timeout: 248 seconds]
09:46 -!- ucasano [~ucasano@host185-90-static.5-79-b.business.telecomitalia.it]
has quit [Quit: ucasano]
09:54 -!- sacho [~sacho@83.228.17.177] has quit [Ping timeout: 246 seconds]
09:57 -!- Mehak [~androirc@76.103.175.11] has quit [Quit: AndroIRC]
09:57 -!- sebastianskejoe [~sebastian@188.114.142.217] has joined #go-nuts
09:59 -!- skelterjohn [~jasmuth@c-24-0-2-70.hsd1.nj.comcast.net] has quit [Ping
timeout: 246 seconds]
10:02 -!- Project_2501 [~Marvin@dynamic-adsl-94-36-161-95.clienti.tiscali.it] has
quit [Ping timeout: 240 seconds]
10:07 -!- sacho [~sacho@83.228.17.177] has joined #go-nuts
10:17 -!- Squeese [~squeese@cm-84.209.17.156.getinternet.no] has joined #go-nuts
10:25 -!- sacho [~sacho@83.228.17.177] has quit [Read error: Operation timed out]
10:26 -!- napsy [~luka@88.200.96.18] has quit [Ping timeout: 260 seconds]
10:27 -!- napsy [~luka@88.200.96.18] has joined #go-nuts
10:29 < jemeshsu> Is there any golang book under development?  Coming soon?
10:30 < TheMue> jemeshsu: There are already several
10:31 < TheMue> One in the US, two (mine and another one) in Germany and
afaik one in Japan
10:32 < jemeshsu> I can only find one by Baugh in Amazon
10:32 < TheMue> jemeshsu: http://www.amazon.de/gp/product/3898647129/
10:32 < TheMue> But it's in German
10:32 < jemeshsu> TheMue: where's yours?  url?
10:33 -!- m4dh4tt3r [~Adium@c-98-210-145-213.hsd1.ca.comcast.net] has quit [Ping
timeout: 252 seconds]
10:34 < jemeshsu> TheMue: I don't want to learn another language to learn a
new language :-) Thanks for the link
10:35 < fzzbt> jemeshsu: here is free go book http://miek.nl/files/go/
10:35 < TheMue> Hehe, sorry, we already asked US publishers for a
translation, but there is no interest
10:35 < TheMue> There are new English book in work
10:37 < jemeshsu> TheMue: i know about the Gieben book.  thanks
10:37 < jemeshsu> guess need more github projects, blog posts to make golang
more mainstream
10:46 -!- ExsysTech [~ExsysTech@50-46-213-60.evrt.wa.frontiernet.net] has quit
[Read error: Connection reset by peer]
10:47 -!- edsrzf [~edsrzf@122-61-221-144.jetstream.xtra.co.nz] has quit [Remote
host closed the connection]
10:48 -!- ExsysTech [~ExsysTech@50-46-213-60.evrt.wa.frontiernet.net] has joined
#go-nuts
11:00 -!- oal [~oal@5.79-160-122.customer.lyse.net] has joined #go-nuts
11:04 -!- Project_2501 [~Marvin@82.84.68.151] has joined #go-nuts
11:11 < TheMue> jemeshsu: You know the dashboard?
11:12 < TheMue> jemeshsu: And this week Go has been announced as third lang
on the GAE.  So for such a young lang it get's better each day.
11:13 < quag> fzzbt: what's the book like?
11:17 < fzzbt> quag: haven't read it.  seems promising, but it's still WIP.
11:27 -!- Squeese [~squeese@cm-84.209.17.156.getinternet.no] has quit [Remote host
closed the connection]
11:27 < heatxsink> TheMue: that's pretty crazy that it's the third lang for
app eng.
11:28 < TheMue> heatxsink: Yep, an important step for go.  So this week the
go related traffic on the web raised a lot.
11:28 < heatxsink> w00t
11:28 -!- keidaa [~keidaa@cm-84.210.56.138.getinternet.no] has joined #go-nuts
11:30 < quag> I wonder if it is because there is pressure to use Go
internally on google's internal 'app engine'
11:33 < str1ngs> I think it was more because there as alot of intreset for
it.  and go language seems quite logical for app engine
11:36 -!- Adys_ [~Adys@unaffiliated/adys] has joined #go-nuts
11:38 < TheMue> or just to lower the importance of (oracles) language (java)
11:39 < quag> TheMue: isn't Oracle doing a good job of that themselves?
11:39 -!- Adys_ [~Adys@unaffiliated/adys] has quit [Client Quit]
11:39 < TheMue> quag: hehe
11:40 < aiju> Oracle even sucks at sucking
11:40 < heatxsink> aiju: lolz
11:45 -!- Project-2501 [~Marvin@dynamic-adsl-94-36-170-9.clienti.tiscali.it] has
joined #go-nuts
11:48 < moraes> they want to place go as the gae option for highly intensive
tasks
11:48 -!- Project_2501 [~Marvin@82.84.68.151] has quit [Ping timeout: 264 seconds]
11:51 < moraes> heatxsink, and this is pretty funny:
http://twitter.com/#!/dougnaka/statuses/68039575997661184
11:52 < heatxsink> nobody has ever heard of?
11:52 < heatxsink> lolz
11:52 < heatxsink> geez
11:52 < heatxsink> if you haven't heard of it you have n00b status
11:55 < moraes> many php users only like languages that can run wordpress
11:56 < str1ngs> ya he uses php so we should not worry to much about his
opinion
11:56 < aiju> php users only know php
11:56 < aiju> and C
11:56 < aiju> because php is written in it
11:57 -!- Ekspluati [5b9a00ea@gateway/web/freenode/ip.91.154.0.234] has joined
#go-nuts
11:58 -!- Elspeth_ [ca2e8ec1@gateway/web/freenode/ip.202.46.142.193] has joined
#go-nuts
12:00 < str1ngs> struct fields should be created when creating a struct with
f := new(Foo) correct?
12:03 < quag> isn't it a relief that app-engine doesn't support php?
12:03 < str1ngs> ok I think I have declaring and new confused
12:05 < quag> str1ngs: all fields go to their zero values when using new()
12:05 < quag> if you want to populate a struct with values use Foo{a=...,
b=...} and so on
12:05 < quag> Is that what you're getting at?
12:05 < str1ngs> ok thats working for go types
12:06 < str1ngs> no I have a problem with new(Foo) and C type fields
12:06 < str1ngs> ie type Foo { cfield *C.foo }
12:07 < str1ngs> typo missed the struct keyword
12:07 < str1ngs> I'll paste some code will make it easier
12:07 < str1ngs> but my guess the the zero value for C fields is nil
12:08 < quag> sorry, I don't know about the C interation.
12:14 -!- skelterjohn [~jasmuth@c-24-0-2-70.hsd1.nj.comcast.net] has joined
#go-nuts
12:14 < str1ngs> doh pointers are nil
12:14 < str1ngs> sill me
12:14 < str1ngs> silly*
12:14 < str1ngs> ok that explains that
12:22 -!- aho [~nya@fuld-590c6141.pool.mediaWays.net] has joined #go-nuts
12:34 -!- ImpatientSpoon [~ashley@115.6.233.220.static.exetel.com.au] has quit
[Read error: Connection reset by peer]
12:35 -!- photron [~photron@port-92-201-244-191.dynamic.qsc.de] has joined
#go-nuts
12:39 -!- breadlorde [~dreadlord@c-24-11-39-160.hsd1.mi.comcast.net] has joined
#go-nuts
12:51 -!- XenoPhoenix [~Xeno@cpc13-aztw24-2-0-cust23.aztw.cable.virginmedia.com]
has quit [Quit: back in a bit]
12:53 -!- keidaa [~keidaa@cm-84.210.56.138.getinternet.no] has quit [Quit:
Leaving]
12:56 -!- l00t [~i-i3id3r_@189.105.66.232] has quit [Ping timeout: 246 seconds]
13:06 -!- ExsysHost [~ExsysTech@50-46-213-60.evrt.wa.frontiernet.net] has joined
#go-nuts
13:06 -!- snearch [~snearch@g225018121.adsl.alicedsl.de] has joined #go-nuts
13:08 -!- ExsysTech [~ExsysTech@50-46-213-60.evrt.wa.frontiernet.net] has quit
[Ping timeout: 252 seconds]
13:10 < uriel> 11:40 < aiju> Oracle even sucks at sucking
13:10 < uriel> nonsense, they are really good at it
13:10 < uriel> specially at sucking up all the money from corporate fools
13:10 < str1ngs> haha
13:12 -!- boscop [~boscop@f055153169.adsl.alicedsl.de] has quit [Ping timeout: 246
seconds]
13:12 -!- huin [~huin@91.85.171.238] has quit [Quit: luncheon]
13:25 -!- napsy [~luka@88.200.96.18] has quit [Quit: leaving]
13:25 -!- napsy [~luka@88.200.96.18] has joined #go-nuts
13:28 -!- Ekspluati [5b9a00ea@gateway/web/freenode/ip.91.154.0.234] has quit
[Quit: Page closed]
13:29 < manveru> container/heap really could do with some example...
13:32 -!- napsy [~luka@88.200.96.18] has quit [Quit: Lost terminal]
13:33 < str1ngs> manveru: see there tests
13:33 < str1ngs> manveru: also you can contribute to the docs for that
13:35 -!- XenoPhoenix [~Xeno@cpc13-aztw24-2-0-cust23.aztw.cable.virginmedia.com]
has joined #go-nuts
13:35 < manveru> heh
13:35 < manveru> if i knew how to write those docs...
13:36 -!- napsy [~luka@88.200.96.18] has joined #go-nuts
13:36 < manveru> the only useful test i can find is test/chan/sieve2.go -
which seems to say that any container/vector can be used as a container/heap
13:37 < manveru> now i just have to write my own vector because we still
don't have generics :(
13:38 -!- mattn_jp [~mattn@180-144-129-39f1.hyg1.eonet.ne.jp] has joined #go-nuts
13:39 < Soultaker> does Go have something like Java's Scanner or C++
iostream yet?  (Not go.scanner, which is for Go source only)
13:40 < manveru> 200 lines of useless boilerplate...
13:40 -!- rael_wiki [~chatzilla@unaffiliated/rael-wiki/x-8420294] has joined
#go-nuts
13:41 < exch> Soultaker: I guess the closest match are the fmt.Scan()
functions and friends
13:41 < exch> not sure if that is what you are after
13:42 -!- boscop [~boscop@f055145097.adsl.alicedsl.de] has joined #go-nuts
13:42 < TheMue> manveru: Why do you explicitely need vectors?
13:42 < manveru> TheMue: because that's what the test uses?
13:43 < manveru> needs push/pop, and the sort interface
13:43 < TheMue> manveru: do you need it or the test?
13:43 < manveru> so this seems like the best option
13:44 < Soultaker> exch: actually that looks pretty close to what I was
looking for, thanks!
13:44 < manveru> TheMue: i need container/heap
13:44 < manveru> anything else is boilerplate
13:45 < manveru> vector seems like the most mature option for that, and i
can translate it with two gofmts
13:47 -!- araujo [~araujo@gentoo/developer/araujo] has joined #go-nuts
13:50 -!- snearch [~snearch@g225018121.adsl.alicedsl.de] has quit [Quit:
Verlassend]
14:00 < Soultaker> is there a ReadLine function that allocates its own
buffer, so I don't have to bother with preallocating a slice of sufficient length?
14:01 -!- breadlorde [~dreadlord@c-24-11-39-160.hsd1.mi.comcast.net] has quit
[Ping timeout: 250 seconds]
14:03 < hallas> Soultaker, which package?
14:03 < Soultaker> I don't know, that's why I ask.  ;)
14:04 < Soultaker> (or if you meant ReadLine: I found that in bufio)
14:05 < hallas> I dont
14:05 < hallas> I mean
14:05 < aiju> ReadLine allocates its own buffer
14:05 < rael_wiki> if a channel is closed while there are some values still
left inside it a range operator on that channel will immediately stop or will it
keep getting the remaining values and then stop?
14:05 < hallas> If you talk about the ReadLine of bufio, then it doesnt make
any sence about pre allocating anything
14:05 < aiju> rael_wiki: the latter
14:05 -!- mattn_jp [~mattn@180-144-129-39f1.hyg1.eonet.ne.jp] has quit [Remote
host closed the connection]
14:05 < rael_wiki> aiju: thanks god ;)
14:07 < kamaji> java makes me sad
14:08 < kamaji> tomcat is so enterprisey it hurts
14:08 < heatxsink> kamaji: ya it is
14:09 < kamaji> it feels like a collection of libraries that are all trying
to do the same thing in ever-so-slightly different ways, and aren't really
managing it very well
14:10 < kamaji> I wish I could use twister for this
14:10 -!- angasule [~angasule@190.2.33.49] has joined #go-nuts
14:10 < aiju> kamaji: aren't most libraries like that?
14:10 < kamaji> ever the cynic
14:11 < kamaji> Maybe
14:11 < Soultaker> aiju/hallas: you are right, I misread that.
14:12 < kamaji> but it's just...  ugh
14:16 -!- monsieur_max [~maxime@ip-78.net-89-2-171.rev.numericable.fr] has quit
[Quit: Leaving.]
14:16 -!- PortatoreSanoDiI [~Marvin@82.84.74.1] has joined #go-nuts
14:20 -!- Project-2501 [~Marvin@dynamic-adsl-94-36-170-9.clienti.tiscali.it] has
quit [Ping timeout: 260 seconds]
14:28 -!- l00t [~i-i3id3r_@189.105.108.155] has joined #go-nuts
14:38 < str1ngs> I have a good idea for a project github markdow support for
doc.go
14:39 -!- huin [~huin@91.85.171.238] has joined #go-nuts
14:41 -!- Tuller [~tuller@c-69-143-48-210.hsd1.va.comcast.net] has joined #go-nuts
14:41 -!- firwen [~firwen@2a01:e34:eea3:7e10:4a5b:39ff:fe51:e8ae] has quit [Remote
host closed the connection]
14:42 -!- monsieur_max [~maxime@ip-78.net-89-2-171.rev.numericable.fr] has joined
#go-nuts
14:43 < hallas> doc.go ?
14:43 < hallas> godoc?
14:46 < str1ngs> right
14:49 -!- hallas [~hallas@x1-6-30-46-9a-b2-c5-1f.k891.webspeed.dk] has left
#go-nuts []
14:52 -!- ako [~nya@fuld-590c6993.pool.mediaWays.net] has joined #go-nuts
14:54 -!- bortzmeyer [~stephane@2a01:e35:8bd9:8bb0:c4f3:affa:111a:fa53] has joined
#go-nuts
14:54 -!- aho [~nya@fuld-590c6141.pool.mediaWays.net] has quit [Ping timeout: 246
seconds]
14:58 -!- monsieur_max [~maxime@ip-78.net-89-2-171.rev.numericable.fr] has quit
[Quit: Leaving.]
15:01 -!- merijn [~merijn@535134AB.cm-6-2a.dynamic.ziggo.nl] has joined #go-nuts
15:01 -!- monsieur_max [~maxime@ip-78.net-89-2-171.rev.numericable.fr] has joined
#go-nuts
15:02 < merijn> How is it decided what gets exported in a package?  Are
there explicit public/private declarations or...?
15:02 < aiju> merijn: first letter
15:02 < aiju> capital letter means exported
15:02 < merijn> ok, thanks :)
15:07 -!- nictuku [~nict@84-72-7-79.dclient.hispeed.ch] has joined #go-nuts
15:07 -!- nictuku [~nict@84-72-7-79.dclient.hispeed.ch] has quit [Changing host]
15:07 -!- nictuku [~nict@unaffiliated/nictuku] has joined #go-nuts
15:11 -!- hallas [~hallas@x1-6-30-46-9a-b2-c5-1f.k891.webspeed.dk] has joined
#go-nuts
15:11 < hallas> Small question, not entirely go related.  What kind of token
is { and } if you had to give them a name?  In most grammars they're simply
LEFT_PARAM RIGHT_PARAM.
15:13 < hallas> A seperator, or control character?  I'm not sure.
15:15 < ampleyfly> what about block delimiter
15:16 < skelterjohn> brackets
15:17 < skelterjohn> scope operator
15:17 < skelterjohn> dunno
15:17 -!- Tuller [~tuller@c-69-143-48-210.hsd1.va.comcast.net] has quit [Quit:
Computer has gone to sleep.]
15:17 < hallas> Meh alright
15:18 < aiju> SQUIRLYTHINGLEFT
15:18 < hallas> Writing a disposition for my thesis - and I just started to
wonder :P
15:18 < aiju> SQUIRLYTHINGRIGHT
15:19 < ampleyfly> seriously, I think block delimiter is an established term
15:22 < hallas> ampleyfly: alright thanks
15:28 -!- MX80 [~MX80@cust151.253.117.74.dsl.g3telecom.net] has joined #go-nuts
15:28 -!- FX80 [~MX80@cust151.253.117.74.dsl.g3telecom.net] has joined #go-nuts
15:35 -!- danilo04 [~danilo04@129.21.100.41] has joined #go-nuts
15:36 -!- dfr|mac [~dfr|work@ool-182e3fca.dyn.optonline.net] has joined #go-nuts
15:37 < skelterjohn> aren't delimiters things that go on the end?  like
semicolons w/ C?
15:37 < Soultaker> what's a good way to do for i := range(array) {} when I
don't use i in the loop?
15:37 < skelterjohn> it sets de-limit
15:37 < skelterjohn> _ = range array
15:37 < skelterjohn> don't need the parens - range is not a function
15:38 < Soultaker> ah, I was trying _ := range, but that's not allowed
apparently :)
15:38 < Soultaker> thanks.
15:38 < Soultaker> (and I'll get rid of the parentheses)
15:39 < Soultaker> (it's a bit confusing that builtins like make() and len()
require parentheses but other builtins like new and range apparently don't)
15:39 < aiju> new requires parentheses
15:39 < aiju> and range is not a builtin function
15:39 < aiju> range is a part of syntax
15:39 < aiju> like if or for
15:39 < Soultaker> yeah, but so is make.
15:40 < aiju> make is a builtin function
15:40 < aiju> valid as a value
15:40 < aiju> range is not valid as a value
15:40 < Soultaker> functions can't take types as arguments, can they?
15:40 < aiju> it's a magical function
15:40 < Soultaker> well, then the same could be said of range, couldn't it?
15:41 < aiju> you can't write i := range foo
15:41 < aiju> you need a for statement around that
15:44 -!- moraes_ [~moraes@189.103.176.38] has joined #go-nuts
15:47 -!- moraes [~moraes@189.103.176.38] has quit [Ping timeout: 248 seconds]
15:49 < rael_wiki> is it a good habit to close all open channels before the
end of programs or is it just to be done if useful?
15:50 -!- chomp [~chomp@c-67-186-35-69.hsd1.pa.comcast.net] has joined #go-nuts
16:00 < manveru> rael_wiki: only if you like :)
16:01 < str1ngs> godoc -html . > index.html and profit :P
16:03 -!- Fish [~Fish@9fans.fr] has joined #go-nuts
16:04 -!- Squeese [~squeese@cm-84.209.17.156.getinternet.no] has joined #go-nuts
16:10 -!- dju [dju@fsf/member/dju] has quit [Read error: Connection reset by peer]
16:10 -!- dju_ [dju@fsf/member/dju] has joined #go-nuts
16:14 -!- bockmabe [~bockmabe@c-24-143-97-84.customer.broadstripe.net] has joined
#go-nuts
16:17 -!- merijn [~merijn@535134AB.cm-6-2a.dynamic.ziggo.nl] has quit [Ping
timeout: 276 seconds]
16:18 < wrtp> Soultaker: make and new can be used in any expression context;
range can only be used in a for statement
16:22 < manveru> ok...  priority queue seems to work fine
16:23 -!- gtaylor [~gtaylor@99-5-124-9.lightspeed.gnvlsc.sbcglobal.net] has joined
#go-nuts
16:26 -!- Squeese [~squeese@cm-84.209.17.156.getinternet.no] has quit [Remote host
closed the connection]
16:27 -!- firwen [~firwen@2a01:e34:eea3:7e10:4a5b:39ff:fe51:e8ae] has joined
#go-nuts
16:28 < str1ngs> hmm godoc --http does no generate a table of contents for
each package
16:29 -!- oal [~oal@5.79-160-122.customer.lyse.net] has quit [Read error:
Connection reset by peer]
16:33 -!- oal [~oal@5.79-160-122.customer.lyse.net] has joined #go-nuts
16:38 -!- twopoint718 [~chris@fsf/member/twopoint718] has joined #go-nuts
16:38 -!- wrtp [~rog@92.17.93.207] has quit [Quit: wrtp]
16:44 -!- nteon [~nteon@c-98-210-195-105.hsd1.ca.comcast.net] has joined #go-nuts
16:48 -!- tvw [~tv@e176004005.adsl.alicedsl.de] has joined #go-nuts
16:49 -!- pearle [~pearle@blk-224-181-222.eastlink.ca] has joined #go-nuts
16:54 < justinlilly> how do you go about telling C where it can find
libraries?  Trying to compile a go module that uses C libraries.
16:54 < justinlilly> Its telling me that it can't find some header files.
16:54 < str1ngs> C or cgo?
16:55 < justinlilly> basically I'm trying to install go-git, and it can't
find git2.h
16:55 < justinlilly> not sure which environment variable tells it where to
find C packages.
16:55 < str1ngs> ah do you have libgit2 installed?
16:55 < justinlilly> I have a check out of it nearby
16:55 < justinlilly> just not sure how to say "look for libgit2 over THERE"
16:56 < str1ngs> what os are you using?
16:56 < justinlilly> linux
16:56 < str1ngs> distro?
16:56 < justinlilly> ubuntu 11.*
16:56 < str1ngs> apt-cache search libgit
16:56 < justinlilly> I have a checkout of libgit right nearby..  can I not
use that?
16:57 < str1ngs> you can but its more work
16:57 < justinlilly> It should seemingly be as easy as setting an
environment variable telling it where to search for libraries, no?
16:57 < str1ngs> you need to use -I cflag iirc
16:57 < str1ngs> let me mess with it here
16:57 < str1ngs> also make module should pull the libgit sub module
16:57 < nteon> justinlilly: tipically it searches /usr/include and friends,
so you can do make install in libgit2 and it should work
16:58 < str1ngs> justinlilly: or git submodule init;git submodule update
16:58 < nteon> but if it pulls ti in as a submodule do that :)
16:58 < justinlilly> str1ngs: I did that.  btw, git submodule update --init
is a shortcut for those two commands.
16:58 < str1ngs> no!  make module :P
16:58 < str1ngs> :)
16:59 < str1ngs> ok let me mess with the -L cflag
16:59 < nteon> a native git package would be really cool
16:59 < justinlilly> https://github.com/str1ngs/go-git
16:59 < str1ngs> not exactly native but a start
17:00 -!- FX80 [~MX80@cust151.253.117.74.dsl.g3telecom.net] has quit [Quit:
Leaving]
17:00 < str1ngs> justinlilly: hope you saw http://str1ngs.github.com/go-git/
17:00 < str1ngs> justinlilly: this way I only have to hack at one document
source
17:01 < justinlilly> I didn't.  I see the make rule for it though.
17:01 * justinlilly has just been reading the source.
17:01 < chomp> hrm so os provides geteuid but not seteuid?
17:02 -!- rcrowley [~rcrowley@c-71-202-44-233.hsd1.ca.comcast.net] has joined
#go-nuts
17:02 < nteon> str1ngs: very cool
17:09 < justinlilly> str1ngs: btw, only libgit stuff in apt is libgit-ruby
17:09 < str1ngs> justinlilly: the other option is to use /usr/local/ for now
17:09 < str1ngs> ./waf configure
17:09 < str1ngs> ./waf build
17:09 < str1ngs> ./waf install iirc
17:09 < justinlilly> right.
17:10 < justinlilly> Well I'll give it a go.
17:10 < str1ngs> but that crufts up /usr/local ..  which I dont mind much
but you may
17:10 * justinlilly would prefer it didn't happen, but wants to get this tool
working more.
17:10 < str1ngs> I'll work on the CFLAGS issue
17:11 < skelterjohn> justinlilly: // #cgo CFLAGS: -I somewhere
17:11 < skelterjohn> above import "C"
17:11 < skelterjohn> http://golang.org/cmd/cgo/
17:12 -!- hungrygruffalo
[~hungrygru@host86-186-120-61.range86-186.btcentralplus.com] has quit [Quit:
hungrygruffalo]
17:13 -!- hungrygruffalo
[~hungrygru@host86-186-120-61.range86-186.btcentralplus.com] has joined #go-nuts
17:15 < str1ngs> yes but if we change that then it breaks for goinstall
17:16 < str1ngs> which might not be the best option
17:18 < str1ngs> justinlilly: I think /usr/local is the best option
17:18 < justinlilly> that's what I'm doing for now.
17:18 < str1ngs> its easier to support people this way
17:18 < str1ngs> and goinstall will work also
17:19 < str1ngs> and for people with distro packages it will just work
provided is sane
17:19 < str1ngs> push comes to shove I'll go back to pkg-config but will
break goinstall
17:21 < str1ngs> also w/e you decide is fine for top level design is fine
with me.  I'm kinda weak in that regards.  you seem to be much better at it then I
17:22 < justinlilly> making up apis is fun.  I tend to just pseudocode then
try to make the code reflect it.
17:23 < str1ngs> thats perfect
17:24 -!- cinghialino [~Marvin@82.84.74.1] has quit [Read error: Connection reset
by peer]
17:25 < str1ngs> I'm also going to play with godef more.  its possible we
can do direct struct mappings like https://gist.github.com/9d118b7db3cfb9043aa9
17:26 < justinlilly> cool.
17:26 < str1ngs> which I think you were sorry after?
17:26 < str1ngs> sorta*
17:26 < justinlilly> kinda, yea.
17:27 < justinlilly> I need to walk git log and get a list of files touched
with each commit, then generate diffs for them.
17:27 < justinlilly> I'm making a git-backed static site generator, fwiw.
17:28 < justinlilly> https://gist.github.com/5ca661d29a354be0847c -- general
idea
17:30 -!- preflex [~preflex@unaffiliated/mauke/bot/preflex] has quit [Ping
timeout: 276 seconds]
17:30 < str1ngs> ah nice
17:31 < str1ngs> I'll forgive you for useing emacs :P
17:31 * justinlilly has used all other editors, including vim for 5 years, before
converting.
17:31 * justinlilly is happy with his choice.  ^_^
17:33 < str1ngs> I really tried to use emacs.  but sadly I could not convert
17:33 < justinlilly> Why's that?
17:33 < str1ngs> I started with vi
17:33 < str1ngs> and chaining commands I could not get use to
17:34 < str1ngs> I also set -o vi . to make matters worst so I never use gnu
style bindings
17:34 -!- preflex [~preflex@unaffiliated/mauke/bot/preflex] has joined #go-nuts
17:34 < justinlilly> I started with vim.  I greatly disliked emacs, for no
real reason.  I decided I wanted to write an objective article about why emacs was
awful, so I tried it for a month.  I've been with it since :)
17:34 < justinlilly> (though I still use vim almost daily)
17:34 < str1ngs> hehe
17:35 < hallas> VIM FTW!
17:35 < str1ngs> I can not say I dislike emacs.  I just like to bug emacs
users :P
17:35 < moraes> i'd like to try emacs but i have a pc
17:35 < str1ngs>
https://github.com/str1ngs/go-git/blob/master/pkg/git/git_test.go#L139 example
refwalk
17:35 * moraes is trolling too
17:35 < hallas> Install Ubuntu :-)?
17:36 < justinlilly> Can't say that I really want to convert anyone to
emacs, just feel obligated to help educate.
17:36 < moraes> eubuntus?
17:36 < hallas> Huh?
17:36 < moraes> my joke is lame.
17:36 < justinlilly> str1ngs: I have the refwalk working.  I'm down to
commit objects, but I need to access deeper than what Commit exposes.
17:36 < str1ngs> justinlilly: I think you want tree
17:36 < str1ngs> justinlilly: possibly use the oid to do a tree look up
17:36 < str1ngs> for each commit
17:38 < str1ngs> also make sure to defer repo.Free() and possibly
revwalk.Free() I need to check revwalk repo you should call
17:38 < justinlilly> what's the best way to develop this locally?  I have a
.a file under GOROOT?
17:38 < str1ngs> I use _test.go files
17:39 < str1ngs> use _test files for working on the go-git and then make
install and import git
17:42 < aiju> the funny thing is most arguments "why EMACS is bad" also
apply to vim
17:46 < moraes> i don't use any, my editor is awful
17:46 < str1ngs> justinlilly: ah I have limited tree support right now
17:46 < justinlilly> that's fine.  I'll add some in a bit.
17:50 -!- dju__ [dju@fsf/member/dju] has joined #go-nuts
17:51 -!- Loonacy [~loonacy@c-67-172-248-248.hsd1.ut.comcast.net] has joined
#go-nuts
17:52 -!- dju_ [dju@fsf/member/dju] has quit [Ping timeout: 240 seconds]
18:04 -!- awidegreen [~quassel@h-170-226.A212.priv.bahnhof.se] has quit [Remote
host closed the connection]
18:06 -!- awidegreen [~quassel@h-170-226.A212.priv.bahnhof.se] has joined #go-nuts
18:12 < chomp> mmm PAM wrapper's up- https://github.com/krockot/gopam
18:13 < aiju> yuck pam
18:13 -!- tvw [~tv@e176004005.adsl.alicedsl.de] has quit [Ping timeout: 246
seconds]
18:13 < chomp> i know right
18:13 < aiju> haha
18:13 < chomp> at least it's ever-so-slightly less painful than using the C
API directly
18:18 -!- angasule [~angasule@190.2.33.49] has quit [Ping timeout: 240 seconds]
18:18 -!- l00t [~i-i3id3r_@189.105.108.155] has quit [Read error: Connection reset
by peer]
18:20 -!- alexandere [~alexander@eijg.xs4all.nl] has joined #go-nuts
18:21 -!- l00t [~i-i3id3r_@20150141106.user.veloxzone.com.br] has joined #go-nuts
18:23 -!- angasule [~angasule@190.2.33.49] has joined #go-nuts
18:25 < nteon> is there a way to create something like pipes?  I need to
test a function that uses a bufio.Reader
18:25 < aiju> io.Copy?
18:26 -!- l00t [~i-i3id3r_@20150141106.user.veloxzone.com.br] has quit [Read
error: Connection reset by peer]
18:28 < str1ngs> nteon: io.Pipe()
18:28 < nteon> I think io.Pipe is what im looking for
18:28 < nteon> heh
18:28 < nteon> yes :)
18:28 < str1ngs> nteon: are you running something?
18:28 < str1ngs> with exec or StartProcess
18:30 -!- l00t [~i-i3id3r_@189.105.3.38] has joined #go-nuts
18:31 < nteon> str1ngs: no, basically I just want to test some code that
takes a reader as an argument.  maybe there is an easier way.  I guess I could
create a bufio.NewReader from some byte slice, that might be better
18:32 < chomp> that sounds perfectly reasonable
18:32 < str1ngs> nteon: bytes.Buffer
18:33 < nteon> str1ngs: thanks
18:33 < str1ngs> then use io.Copy
18:33 < str1ngs> you might not need bytes.Buffer with io.Copy
18:33 < str1ngs> just copy to os.Stderr for testing
18:35 < str1ngs> bytes.NewBuffer to create one from a bytes slice
18:35 -!- merijn [~merijn@inconsistent.nl] has joined #go-nuts
18:36 -!- twopoint718 [~chris@fsf/member/twopoint718] has quit [Ping timeout: 252
seconds]
18:42 < chomp> is there currently any way to get goinstall to play nicely
with cgo modules
18:42 < chomp> specifically to, say, become aware of an external linker
dependency
18:44 -!- nictuku [~nict@unaffiliated/nictuku] has quit [Quit: brb]
18:45 -!- eikenberry [~jae@ivanova.zhar.net] has quit [Quit: End of line.]
18:46 -!- rcrowley [~rcrowley@c-71-202-44-233.hsd1.ca.comcast.net] has quit [Quit:
Computer has gone to sleep.]
18:49 -!- l00t [~i-i3id3r_@189.105.3.38] has quit [Ping timeout: 264 seconds]
18:49 < chomp> ah, nevermind.  there it goes.
18:49 -!- alexandere [~alexander@eijg.xs4all.nl] has quit [Quit: alexandere]
18:51 < str1ngs> chomp: there is see http://golang.org/cmd/cgo/
18:51 < chomp> ya, just found the LDFLAGS stuff
18:51 < chomp> thanks
18:51 < str1ngs> np
18:58 -!- |Craig| [~|Craig|@panda3d/entropy] has joined #go-nuts
19:02 -!- hungrygruffalo
[~hungrygru@host86-186-120-61.range86-186.btcentralplus.com] has quit [Quit:
hungrygruffalo]
19:03 < justinlilly> str1ngs: how do you go about debugging issues with this
cgo stuff?  I'm not fairly familiar with C, if you couldn't tell, though I can
read it.
19:04 < justinlilly> Specifically, I'm doing something I think should work,
mostly following the C api docs, but I'm getting a rather opaque error: Cannot
read reference file 'packed-refs'
19:05 -!- aho [~nya@fuld-590c6172.pool.mediaWays.net] has joined #go-nuts
19:07 < str1ngs> ah from refwalk.Next() ?
19:07 < justinlilly> https://gist.github.com/8ab864d2a1f0cd19589a -- gist if
it makes it clearer.
19:08 < justinlilly> compare to: http://libgit2.github.com/api.html#trees
19:08 -!- m4dh4tt3r [~Adium@c-98-210-145-213.hsd1.ca.comcast.net] has joined
#go-nuts
19:08 < justinlilly> (btw, I think the ruby api is something worth basing
the go one off of.  It makes lots of sense)
19:08 < str1ngs> I assume you have a commit?
19:08 < justinlilly> I'm reusing the repo object from earlier in the tests.
It should have several commits.
19:08 -!- ako [~nya@fuld-590c6993.pool.mediaWays.net] has quit [Ping timeout: 246
seconds]
19:08 < justinlilly> ...  assuming I'm following this correctly?
19:09 < str1ngs> try with one created by git
19:09 < justinlilly> I tried with a real repository.  same effect.
19:10 < str1ngs> ok I'll look into it more . mean time you can get the error
number from the C function and look it up in error.h
19:11 < justinlilly> found the error earlier, just couldn't find where to
look it up.
19:11 < str1ngs> LastError is new and might be buggy or not completey on the
libgit2 side
19:12 < justinlilly> hmm.  GIT_EMISSINGOBJDATA
19:12 < str1ngs> but I think its an issue with GetHead
19:12 < str1ngs> thats rather hackish and not the right way
19:12 < str1ngs> also sounds perverted
19:13 < justinlilly> err.  I think it might be GIT_EINVALIDTYPE as I wasn't
reading the define properly.
19:15 < str1ngs> if you see GetHead I hardcode the ref path which is
probably wrong
19:17 < justinlilly> seems to return the correct oid in this case.
19:18 -!- breadlorde [~dreadlord@c-24-11-39-160.hsd1.mi.comcast.net] has joined
#go-nuts
19:19 < str1ngs> what is the error code?
19:19 < justinlilly> -9
19:20 -!- merijn [~merijn@inconsistent.nl] has left #go-nuts []
19:21 < str1ngs> GIT_EMISSINGOBJDATA
19:21 < justinlilly> no.  its -8
19:21 < justinlilly> because GIT_ERROR is -1 already
19:21 < justinlilly> -1 - 8 = -1 + -8 = -9
19:21 < str1ngs> sorry your right
19:21 < justinlilly> (I made the same mistake)
19:21 < str1ngs> I actually did the math but selected the wrong one
19:21 < str1ngs> for paste
19:22 < justinlilly> so wrong type for tree.
19:22 < str1ngs> http://libgit2.github.com/api.html#trees
19:23 < justinlilly> teehee.  I just linked that above ^_^
19:23 < str1ngs> sorry
19:23 < justinlilly> so it seems like I'm following the API they've defined
there.
19:23 < str1ngs> missed it
19:23 -!- bortzmeyer [~stephane@2a01:e35:8bd9:8bb0:c4f3:affa:111a:fa53] has quit
[Quit: Leaving.]
19:23 < str1ngs> ok my guess HEAD is not going to have a tree
19:23 < str1ngs> so try with a commit
19:24 < justinlilly> HEAD == commit
19:24 < justinlilly> specifically in my case, but I think generally as well.
19:25 < str1ngs> justinlilly: also I think this &tree.git_tree is a better
way to wrap
19:26 < str1ngs> offtopic but I'm thinking I not using pointer for fields
19:29 < justinlilly> apparently you can't use short shas with NewOidString.
19:29 < justinlilly> anyway, even pegging a commit sha to it in a repo with
a single file and single commit..  no luck.
19:29 < str1ngs> no there is another method for that
19:29 < str1ngs> can you push to my repo please.  and I'll mess with it
19:30 < justinlilly> sure.
19:30 < str1ngs> I you can make commits saves doing pull requests
19:33 < justinlilly> comitted.
19:33 < str1ngs> thanks
19:33 < str1ngs> also I there are some methods that can be converted to
static functions just not sure which yet
19:35 < str1ngs> oops sub :P
19:35 < str1ngs> I though of that but I forgot to change it
19:37 -!- Elspeth_ [ca2e8ec1@gateway/web/freenode/ip.202.46.142.193] has quit
[Ping timeout: 252 seconds]
19:38 < str1ngs> CommitCreate is hackish so becareful with it right now
19:40 -!- rcrowley [~rcrowley@c-71-202-44-233.hsd1.ca.comcast.net] has joined
#go-nuts
19:41 < kamaji> does go follow 'BODMAS'?
19:41 -!- dreadlorde [~dreadlord@c-24-11-39-160.hsd1.mi.comcast.net] has quit
[Ping timeout: 252 seconds]
19:42 -!- tobym [~tobym@cpe-72-229-2-6.nyc.res.rr.com] has joined #go-nuts
19:45 -!- twopoint718 [~chris@adsl-71-150-251-93.dsl.mdsnwi.sbcglobal.net] has
joined #go-nuts
19:45 -!- twopoint718 [~chris@adsl-71-150-251-93.dsl.mdsnwi.sbcglobal.net] has
quit [Changing host]
19:45 -!- twopoint718 [~chris@fsf/member/twopoint718] has joined #go-nuts
19:45 < justinlilly> kamaji: I would imagine so, yes.
19:50 -!- napsy [~luka@88.200.96.18] has quit [Read error: Operation timed out]
19:53 < nteon> heh, I had to look that up.  we called it PEDMAS :)
19:56 -!- boscop [~boscop@f055145097.adsl.alicedsl.de] has quit [Ping timeout: 246
seconds]
19:57 -!- boscop [~boscop@f055145097.adsl.alicedsl.de] has joined #go-nuts
19:59 < str1ngs> justinlilly: LastError is wrong in this regards so its
confusing the issue.
19:59 < justinlilly> *shakes a fist*
19:59 < justinlilly> how do you go about debugging this?
19:59 < str1ngs> thats from RevWalk test
20:00 < str1ngs> mainly I use the libgit headers and sometimes read the C
source.
20:00 < str1ngs> for the post parts though I use prints
20:05 < kamaji> nteon: I can't even think what the O is now...
20:05 < kamaji> I'm going to kick myself in a minute
20:06 < justinlilly> kamaji: orders
20:06 < justinlilly> powers/sqrrts
20:06 < kamaji> Ok i don't think i've ever heard it referred to as "orders"
20:06 < kamaji> I wonder why I learned that mnemonic :D
20:07 < justinlilly> dunno.  we always just called it "the order of
operations" afaicr
20:07 -!- TheMue [~TheMue@p5DDF4DC8.dip.t-dialin.net] has quit [Quit: TheMue]
20:10 -!- ExsysTech [~ExsysTech@50-46-213-60.evrt.wa.frontiernet.net] has joined
#go-nuts
20:11 -!- anticw [~anticw@c-98-210-108-13.hsd1.ca.comcast.net] has quit [Read
error: Connection reset by peer]
20:11 -!- monsieur_max1 [~maxime@ip-78.net-89-2-171.rev.numericable.fr] has joined
#go-nuts
20:11 -!- aho [~nya@fuld-590c6172.pool.mediaWays.net] has quit [Ping timeout: 246
seconds]
20:11 -!- monsieur_max [~maxime@ip-78.net-89-2-171.rev.numericable.fr] has quit
[Ping timeout: 246 seconds]
20:11 -!- chomp_ [~chomp@c-67-186-35-69.hsd1.pa.comcast.net] has joined #go-nuts
20:11 -!- ako [~nya@fuld-590c6172.pool.mediaWays.net] has joined #go-nuts
20:11 -!- ExsysHost [~ExsysTech@50-46-213-60.evrt.wa.frontiernet.net] has quit
[Ping timeout: 246 seconds]
20:11 -!- chomp [~chomp@c-67-186-35-69.hsd1.pa.comcast.net] has quit [Ping
timeout: 246 seconds]
20:11 -!- anticw [~anticw@c-98-210-108-13.hsd1.ca.comcast.net] has joined #go-nuts
20:15 -!- rlab [~Miranda@91.200.158.34] has quit [Ping timeout: 260 seconds]
20:16 < nteon> kamaji: Orders (exponents) according to the wikipeia
20:17 < nteon> oh, right, been answered
20:17 < nteon> heh
20:17 -!- rlab [~Miranda@91.200.158.34] has joined #go-nuts
20:18 -!- danilo04 [~danilo04@129.21.100.41] has quit [Ping timeout: 250 seconds]
20:21 < nteon> does anyone know anythign/have any advice for running 6cov?
I'd love to be able to run my unit tests through it, but its telling me "6cov:
stopped at 0x413203; not at breakpoint 1"
20:21 -!- m4dh4tt3r [~Adium@c-98-210-145-213.hsd1.ca.comcast.net] has quit [Quit:
Leaving.]
20:22 < nteon> tracked as issue 1319 it seems
20:25 -!- sebastianskejoe [~sebastian@188.114.142.217] has quit [Quit: Lost
terminal]
20:25 < str1ngs> common waf . waf install does that not imply build
20:25 -!- Urmel| [~11087Urme@82-136-196-44.ip.telfort.nl] has quit [Ping timeout:
248 seconds]
20:30 -!- Urmel| [~11087Urme@82-136-196-44.ip.telfort.nl] has joined #go-nuts
20:35 < str1ngs> justinlilly: hows your C?
20:35 < justinlilly> str1ngs: I can read it, but that's about the extent.
20:35 < justinlilly> Why?
20:36 < str1ngs> was going to see if you could replicate this with C
20:36 < str1ngs> I'll put a ctest together
20:36 < str1ngs> but it boilds down to the git_object_lookup
20:36 < str1ngs> boils*
20:38 < jeremy_c> no go-iup on appspot yet :-/
20:40 -!- zozoR [~Morten@2906ds2-arno.0.fullrate.dk] has quit [Ping timeout: 260
seconds]
20:43 -!- sacho [~sacho@83.228.17.177] has joined #go-nuts
20:44 -!- grncdr [~stephen@sdo.csc.UVic.CA] has quit [Ping timeout: 260 seconds]
20:45 -!- hungrygruffalo
[~hungrygru@host86-186-120-61.range86-186.btcentralplus.com] has joined #go-nuts
20:46 -!- dreadlorde [~dreadlord@c-24-11-39-160.hsd1.mi.comcast.net] has joined
#go-nuts
20:46 -!- hungrygruffalo
[~hungrygru@host86-186-120-61.range86-186.btcentralplus.com] has quit [Client
Quit]
20:50 -!- bockmabe [~bockmabe@c-24-143-97-84.customer.broadstripe.net] has quit
[Ping timeout: 258 seconds]
20:52 -!- Tuller [~tuller@c-69-143-48-210.hsd1.va.comcast.net] has joined #go-nuts
20:52 -!- firwen [~firwen@2a01:e34:eea3:7e10:4a5b:39ff:fe51:e8ae] has quit [Remote
host closed the connection]
20:53 -!- Dr_Who [~tgall@206.9.88.154] has quit [Ping timeout: 276 seconds]
20:54 -!- photron [~photron@port-92-201-244-191.dynamic.qsc.de] has quit [Read
error: Operation timed out]
21:02 -!- monsieur_max1 [~maxime@ip-78.net-89-2-171.rev.numericable.fr] has quit
[Quit: Leaving.]
21:04 -!- Fish [~Fish@9fans.fr] has quit [Quit: So Long, and Thanks for All the
Fish]
21:06 -!- Tuller [~tuller@c-69-143-48-210.hsd1.va.comcast.net] has quit [Quit:
muffins]
21:09 -!- nictuku [~yvesj@unaffiliated/nictuku] has joined #go-nuts
21:09 -!- awidegreen [~quassel@h-170-226.A212.priv.bahnhof.se] has quit [Remote
host closed the connection]
21:09 -!- vzx [~ryan@74-129-201-27.dhcp.insightbb.com] has quit [Quit: </d]
21:09 < nictuku> am I wrong to think that the documentation for package math
should say that Cos() Sin() and Sincos() take radians as input, confront to
degrees?
21:10 < nictuku> I realize most people used to math just assume it's
degrees, but not the rest of us ;-P
21:10 -!- danilo04 [~danilo04@rm2348358874.student.rit.edu] has joined #go-nuts
21:11 < chomp> nictuku, i am confused
21:11 < chomp> they operate in radians, yes?
21:11 < nictuku> yes.
21:11 -!- piranha [~piranha@5ED43A0B.cm-7-5a.dynamic.ziggo.nl] has quit [Quit:
Computer has gone to sleep.]
21:11 < chomp> that's the de facto standard across virtually all math
libraries
21:12 < chomp> though i suppose it's a fair point that it wouldn't hurt the
documentation to divulge such information :p
21:12 < nictuku> right, that doesn't mean it shouldn't be explicit.
21:12 < |Craig|> nictuku: what do you mean by "confront to degrees"?
21:12 < nictuku> "as opposed to"
21:13 < |Craig|> thats not what confront means, I assumed you had a typo of
convert
21:14 < |Craig|> confront = Meet (someone) face to face with hostile or
argumentative intent.
21:14 < nictuku> should have been confer, my bad.
21:14 < nictuku> I meant in the sense of "cf."
21:14 < nictuku> I incorrectly expanded that to "confront".  Sorry for the
confusion.
21:16 < nictuku> seems like that is an improper use of "cf." too
21:16 < chomp> i think "as opposed to" or "instead of" are the clearest ways
of expressing what you wanted :)
21:18 < nictuku> chomp, thank you, that's what I said before.
21:18 < chomp> but digression abounds and this is not #english.  i agree
with you that the documentation should be explicit about this
21:21 -!- AmourDeZombi [~jphillips@c-68-41-9-175.hsd1.mi.comcast.net] has joined
#go-nuts
21:26 -!- mehalelal [~androirc@76.103.175.11] has joined #go-nuts
21:27 -!- hallas [~hallas@x1-6-30-46-9a-b2-c5-1f.k891.webspeed.dk] has quit
[Remote host closed the connection]
21:34 -!- oal [~oal@5.79-160-122.customer.lyse.net] has quit [Remote host closed
the connection]
21:38 -!- dju_ [dju@fsf/member/dju] has joined #go-nuts
21:40 -!- gtaylor [~gtaylor@99-5-124-9.lightspeed.gnvlsc.sbcglobal.net] has quit
[Quit: gtaylor]
21:42 -!- dju__ [dju@fsf/member/dju] has quit [Ping timeout: 248 seconds]
21:43 -!- wrtp [~rog@92.17.93.207] has joined #go-nuts
21:43 -!- huin [~huin@91.85.171.238] has quit [Quit: bedtime]
21:45 -!- ExsysTech [~ExsysTech@50-46-213-60.evrt.wa.frontiernet.net] has quit
[Read error: Connection reset by peer]
21:46 -!- ExsysTech [~ExsysTech@50-46-213-60.evrt.wa.frontiernet.net] has joined
#go-nuts
21:48 < str1ngs> justinlilly: I'm going to post something to the libgit2 ML.
is this holding you up?
21:48 -!- mehalelal [~androirc@76.103.175.11] has quit [Quit: AndroIRC]
21:51 -!- rlab [~Miranda@91.200.158.34] has quit [Read error: Connection reset by
peer]
21:54 -!- mehalelal [~mehalelal@76.103.175.11] has joined #go-nuts
21:55 -!- espeed [~espeed@63.246.231.57] has joined #go-nuts
21:58 -!- Dr_Who [~tgall@206.9.88.154] has joined #go-nuts
22:03 < hpvincent> hi, why does 'package' takes a plain word, but 'import' a
string ?
22:04 < str1ngs> because import needs to be escaped in some way
22:04 < str1ngs> its quite common to have url's and number for import names
22:05 -!- wrtp [~rog@92.17.93.207] has quit [Quit: wrtp]
22:06 < hpvincent> so you do not necessarily import a package ?
22:06 < |Craig|> you may need a path along with the package name
22:07 < str1ngs> import github.com/pkg/foobar/blah is going to break the
compiler
22:07 < str1ngs> so it needs to be escaped
22:07 < |Craig|> or even just "image/png".  You can't have a / in an
identifier
22:08 < str1ngs> yep
22:08 < hpvincent> I see; so import actually refers to a path
22:09 < str1ngs> right
22:10 -!- ShadowIce [~pyoro@unaffiliated/shadowice-x841044] has quit [Quit:
Verlassend]
22:10 < hpvincent> I suppose the 'int' type implement the Stringer
interface, but do I need to import something to benefit from it ?
22:11 < skelterjohn> benefit from it in what way?  and i don't believe int
satisfies Stringer
22:11 < str1ngs> justinlilly: I have a good idea whats going on commit oid
!= tree oid
22:11 -!- pearle [~pearle@blk-224-181-222.eastlink.ca] has quit [Ping timeout: 246
seconds]
22:11 < hpvincent> skelterjohn, I mean, I have 'a' of type 'int', I want to
do a.String()
22:12 < skelterjohn> right.  i don't believe that happens.
22:12 < skelterjohn> i'll check
22:12 < skelterjohn> yeah, doesn't happen
22:12 < skelterjohn> none of the built-in types have methods
22:12 < skelterjohn> not even string implements Stringer :)
22:14 < str1ngs> hpvincent: you can do that still
22:14 < str1ngs> hpvincent: it may not impliment stringer but you can use
a.String() to return a string representation
22:14 < hpvincent> I though it was possible to add new methods to any type
:(
22:15 < str1ngs> you can
22:15 < skelterjohn> hpvincent: you can't add methods to a type you do not
define.  you have to create a new type
22:15 < skelterjohn> type MyInt int
22:15 < skelterjohn> func (i MyInt) String() string { ...  }
22:15 < hpvincent> ah
22:15 < skelterjohn> then you can do MyInt(aRegularInt).String()
22:15 < hpvincent> I see, you have to be very explicit
22:15 < skelterjohn> yes
22:15 < skelterjohn> "var a int; a.String()" <- does not compile
22:16 < hpvincent> so a given type has all its methods defined together in
only one module
22:17 < skelterjohn> yes
22:17 < skelterjohn> so you only have to import one package to know
everything about a type
22:17 < hpvincent> I see, that was not very clear from the doc I read so far
:)
22:18 < skelterjohn> otherwise two different pieces of code could think
something different about one particular type
22:18 < hpvincent> but it does make sense
22:18 < hpvincent> yes, that was actually bugging me
22:19 -!- ab3 [~abe@83.101.90.66] has joined #go-nuts
22:19 < ab3> is there something luke curl in go
22:19 < ab3> i mean a pkg
22:19 < skelterjohn> check http://godashboard.appspot.com/project
22:20 < taruti> ab3: http.Get() is not enough?
22:20 < mehalelal> I am trying to learn how to use the netchan package.
This code: http://pastie.org/1906711 . Is causing this error: multiple-value
netchan.Import() in single-value context.  What am I doing wrong?
22:21 < skelterjohn> mehalelal: netchan.Import() returns multiple values
22:21 < skelterjohn> you're only providing a landing zone for one
22:21 < skelterjohn> the other is probably os.Error, but i haven't looked at
the doc
22:22 < mehalelal> Yea, it is.  I changed it to err := imp.Import(service,
chan, Recv, 1) and same issue
22:22 < skelterjohn> imp, err := netchan.Import(...)
22:22 -!- dreadlorde [~dreadlord@c-24-11-39-160.hsd1.mi.comcast.net] has quit
[Ping timeout: 248 seconds]
22:22 < mehalelal> oh, that makes sense.  Thank you
22:22 < mehalelal> I feel real stupid now
22:22 < skelterjohn> if a function has signature "func Foo() (a, b int)"
22:22 < skelterjohn> you have to say "a, b = Foo()"
22:24 < mehalelal> I thought that the Import I was using was a method?
22:24 < ab3> taruti: why is there no Get func in the doc of http
http://golang.org/pkg/http/
22:24 -!- dreadlorde [~dreadlord@c-24-11-39-160.hsd1.mi.comcast.net] has joined
#go-nuts
22:25 < skelterjohn> ab3: Client.Get
22:25 < taruti> ab3: http://golang.org/pkg/http/#Response.Get
22:25 < mehalelal> Sorry, I was looking at the wrong line number
22:26 < mehalelal> Thank you for your help!  :-)
22:29 -!- allengeorge [~allengeor@c-24-7-17-50.hsd1.ca.comcast.net] has joined
#go-nuts
22:29 < ab3> taruti: why is it func Get(url string) ...  instead of func
(*Response) Get(url string) ...
22:30 < taruti> ab3: because the Get function _creates_ the response.
22:30 < str1ngs> Get is a helper function
22:30 < ab3> from the doc you are made to belief that Get is part of
Response
22:30 < ab3> k, so it is something like CreateBLaBla
22:31 < str1ngs> no
22:31 < taruti> meh, I want unsafeCoerce# in Go :(
22:32 < taruti> *(*foo)(unsafe.Pointer(&x)) is ugly
22:33 < str1ngs> ab3: you need to play with the http package abit after abit
it will make sense.  also dependant on how much you have used othere libraries
22:35 < hpvincent> say I defined "type MyInt int", and "type MyOtherInt
int", how can I cast from MyInt to MyOtherInt or vice versa ?
22:37 < ab3> str1ngs: guess so
22:37 -!- AmourDeZombi [~jphillips@c-68-41-9-175.hsd1.mi.comcast.net] has left
#go-nuts []
22:39 < |Craig|> hpvincent: I assume you can cast to an int, then to the
final type
22:39 < hpvincent> ok, I found how to cast to int, just type 'int
(expression)'
22:40 < hpvincent> one funny thing , if 'a' is of type 'MyInt', then
apparently 'a + 3' is still of this type, as well as '3 + a'
22:40 < jnwhiteh> why is that funny?
22:41 < jnwhiteh> its just straightforward type inference
22:41 < hpvincent> well, why MyInt rather than int ?
22:41 < jnwhiteh> because a is a MyInt
22:41 < hpvincent> but 3 is int ?
22:41 < jnwhiteh> yes, and they both have the same underlying type
22:42 < jnwhiteh> you could have it so the 'a' is basically coerced back to
that underlying type, or you can promote the 3 in order to keep the user defined
type
22:42 < jnwhiteh> having it the other way would make writing lots of code a
pain in the ass =/
22:42 -!- nteon [~nteon@c-98-210-195-105.hsd1.ca.comcast.net] has quit [Quit:
leaving]
22:43 -!- vpit3833 [~user@203.111.33.203] has joined #go-nuts
22:44 < hpvincent> ok; actually when writing 'a+3', 3 must have been
considered as MyInt
22:44 < jnwhiteh> yes
22:46 < hpvincent> but I canno write 'a + b' where b would be an int
22:46 < jnwhiteh> type MyInt int; func main() { var a MyInt = 4; var b int =
3; println(a + b) }
22:46 < jnwhiteh> that doe
22:46 < jnwhiteh> that doesn't work
22:46 < jnwhiteh> correct
22:46 < hpvincent> it's just int constant that are automatically coerced
22:46 < jnwhiteh> the 'type' of 3 is inferred
22:46 < jnwhiteh> its not even coerced
22:46 < hpvincent> I see
22:46 < jnwhiteh> the type is inferred
22:46 < hpvincent> ok
22:47 < hpvincent> it's getting clearer :)
22:47 < jnwhiteh> you can add two MyInts together, so a + b where a is a
MyInt needs a MyInt for b as well, and the compiler knows that the constant 3 can
be a MyInt =)
22:47 < jnwhiteh> I'm simplifying, but that's the general idea
22:49 -!- Squeese [~squeese@cm-84.209.17.156.getinternet.no] has joined #go-nuts
22:57 < hpvincent> I wonder why I cannot do that :
http://pastebin.com/LFu8NgLy
22:58 < hpvincent> basically, I have two types defined to be similar
structure, I want to convert one to the other
22:58 < hpvincent> to do that I try to convert to an anonymous structure
23:00 < skelterjohn> constants are untyped
23:00 < skelterjohn> 3 has no type
23:00 < skelterjohn> however, if you do x := 3, x has type int
23:00 < skelterjohn> cause you gotta choose something
23:00 < skelterjohn> but var x MyInt = 3 is fine too
23:01 < skelterjohn> as is x + 3
23:01 < skelterjohn> even though x + a regular int is a problem
23:01 < skelterjohn> and for the MyOtherInt, you can change
MyOtherInt(int(b)) to MyOtherInt(b)
23:02 < skelterjohn> i think
23:02 < skelterjohn> ?
23:02 < hpvincent> yes, but my problem is with the types t3 and t4 in my
pastebin now
23:02 < skelterjohn> yeah - those types have nothing to do with each other
23:02 < skelterjohn> that they have the same data layout is a coincidence,
as far as the compiler is concerned
23:02 < justinlilly> str1ngs: what I'm doing isn't time sensitive.  just a
side project.  :)
23:02 < hpvincent> but the way I can convert MyInt to int, why can't I
convert t3 to anonymous structure ?
23:03 < skelterjohn> you could probably convert t3 to a struct {a, b int}
23:03 < hpvincent> since it has same structure, it should be allowed, just
like I was allowed to convert MyInt to int
23:03 < str1ngs> justinlilly: ok if you look at the
23:03 < skelterjohn> i don't know what a "struct { int int
23:03 < skelterjohn> } is
23:03 < skelterjohn> i think it might be nonsense
23:03 < str1ngs> justinlilly: test dir in libgit2 they help
23:03 < hpvincent> the same but without the name of the fields ?
23:03 < skelterjohn> no - it isn't go synta
23:03 < skelterjohn> x
23:03 < skelterjohn> that's what i mean
23:04 < skelterjohn> "struct { int int }" doesn't mean anything at all
23:04 < str1ngs> justinlilly: maily the tree one you see that the sha1 for
the tree_oid does not match any commit sha1
23:04 < skelterjohn> but "struct {a, b int}" or maybe even "struct { int;
int }" would be ok
23:04 < str1ngs> justinlilly: so they are not one in the same
23:04 < skelterjohn> but "int int" is not go
23:04 < hpvincent> skelterjohn, ok
23:05 < hpvincent> indeed, I guess it meant a field of type int called int
:)
23:05 < justinlilly> So you have to get the treeoid for the given commit
then.
23:05 < skelterjohn> hpvincent: ah yes, it is go
23:05 < skelterjohn> just not what you meant
23:05 < skelterjohn> right
23:05 < str1ngs> yes I'm thinking you have to some how get the tree oid from
the commit
23:06 < str1ngs> justinlilly: I just got a new keyboard so dont mind all the
typos
23:06 < hpvincent> skelterjohn, after fixing the syntax, it refuses to do
the conversion anyway
23:06 < hpvincent> apparently , it cares not only about the type of the
fields, but also their names
23:07 < str1ngs> justinlilly: this is partyly why I do not like the examples
on the libgit2 site they are not complete
23:07 < skelterjohn> http://pastebin.com/LLYaNKF0
23:07 < skelterjohn> this compiles, hpvincent
23:07 < skelterjohn> oh interesting
23:07 < skelterjohn> right you are
23:08 < hpvincent> this works http://pastebin.com/DnzMrp1C , but I need the
fields to have the same name in both types
23:09 < skelterjohn> btw the use of anonymous types like that is very rare
23:10 < hpvincent> I guess so :) I was just curious if it was possible to do
similar to the MyInt to MyOtherInt conversion with more complicated types
23:11 < hpvincent> actually, I could have just wrote t4(e)
23:12 < skelterjohn> rule of thumb for a new language - if what you're doing
seems unnecessarily complicated, there is probably a better way :)
23:12 -!- twopoint718 [~chris@fsf/member/twopoint718] has quit [Ping timeout: 240
seconds]
23:13 < str1ngs> justinlilly: git_commit_tree
23:13 < justinlilly> str1ngs: aha!  they have api docs.
http://libgit2.github.com/libgit2/group__git__commit.html
23:13 < str1ngs> lol
23:13 < justinlilly> ^_^ found it at the same time
23:14 < str1ngs> just use the header files
23:14 < str1ngs> those docs are meh.  or am I to hardcore?
23:14 < justinlilly> those docs are effectively prettier header files :)
23:15 -!- pearle [~pearle@blk-224-181-222.eastlink.ca] has joined #go-nuts
23:15 < str1ngs> I cant grep them :P
23:15 -!- Adys [~Adys@unaffiliated/adys] has quit [Ping timeout: 240 seconds]
23:17 < str1ngs> also I'm not a fan of how they hardcode so many sha1's
23:17 < str1ngs> atleast not in the example's there not very useful when
they do that
23:18 -!- bombuzal [~bombuzal@cpc5-newc14-2-0-cust836.gate.cable.virginmedia.com]
has joined #go-nuts
23:20 < justinlilly> nice!  Have it parsing commits to trees now.
23:21 < str1ngs> good stuff.  ok I'm not going to touch this code then.
commit away
23:21 < justinlilly> will flesh out Tree with the bits I need, then commit.
23:21 < str1ngs> no worries
23:22 -!- dfc [~dfc@eth59-167-133-99.static.internode.on.net] has joined #go-nuts
23:23 < str1ngs> I'll surf around for a libgit2 ppa
23:29 -!- Adys [~Adys@unaffiliated/adys] has joined #go-nuts
23:40 -!- gtaylor [~gtaylor@99-5-124-9.lightspeed.gnvlsc.sbcglobal.net] has joined
#go-nuts
23:56 -!- gtaylor [~gtaylor@99-5-124-9.lightspeed.gnvlsc.sbcglobal.net] has quit
[Quit: gtaylor]
23:56 -!- gtaylor [~gtaylor@99-5-124-9.lightspeed.gnvlsc.sbcglobal.net] has joined
#go-nuts
--- Log closed Mon May 16 00:00:50 2011