Thanks a lot! I think I understand now.
Hmm so it doesn't look like I get much info from the headers:
2013/02/20 22:49:00 [lang:3] harbor connect
2013/02/20 22:49:00 [lang:3] header: Authorization: Basic c291cmNlOmhhY2ttZQ==
2013/02/20 22:49:00 [lang:3] header: User-Agent: libshout/2.0.0
2013/02/20 22:49:00 [lang:3] header: Content-Type: application/ogg
2013/02/20 22:49:00 [lang:3] header: ice-name: no name
2013/02/20 22:49:00 [lang:3] header: ice-public: 0
Is there no way to pass any arbitrary data to the on_connect function?
(Aside from modifying my source client...)
I'd like to set certain metadata when I recieve a connection from the
harbor. (Like setting the title to the name of the dj that
connects...) If I could at the very least get the mount name?
If not, I'm sure I can change my architecture to solve this problem in
a different way. I could register a different on_connect
function for each harbor source I guess.
Maybe I'm trying to accomplish this in a silly way.
Thank you for being so helpful! Liquidsoap seems very powerful, but I
am having difficulties getting the hang of it.
Hopefully your reply will help me debug more type issues when I run
into them, after some studying.
Thanks,
-Tony
On Sun, Feb 17, 2013 at 2:54 AM, Samuel Mimram <smim...@gmail.com> wrote:
> Hi,
>
> Here your harbor_connect is a function which takes headers as
> argument, and since you log headers, headers must be a string.
> Therefore harbor_connect is of type
>
> (string)->unit
>
> which means that it is a function, taking a string as argument and
> returning nothing. Now, if you look at the documentation of
> input.harbor (liquidsoap -h input.harbor) you can see that the
> expected type for the on_connect argument is
>
> ([(string*string)])->unit
>
> which means that it should not take a string as argument, but a list
> of pairs of strings (pairs header name / header value). This is what
> the error
>
> At line 16, char 78:
> this value has type
> (string,...)->_ (infered at line 7 char 1 - line 10 char 3)
> but it should be a subtype of
> ([_],...)->_
>
> means. The morale is that you cannot directly use log to print the
> headers. So, if you insist on printing them you have to do something a
> bit more involved such as
>
> def harbor_connect(headers)
> list.iter (fun(h) -> log("header: #{fst(h)}: #{snd(h)}"), headers)
> log("harbor connect")
> end
>
> which prints a list of pairs of strings.
>
> I think that this changed between 0.9 and 1.0 which is why the example
> is not valid anymore (and actually I don't think it's present anymore
> in 1.0 documentation).
>
> Hope this helps!
>
> Cheers,
>
> Samuel.
>
> On Sun, Feb 17, 2013 at 8:44 AM, Tony Miller <mcfiredr...@gmail.com> wrote:
>> Having trouble getting on_connect to work with this small example. The
>> error I get is:
>>
>> At line 16, char 78:
>> this value has type
>> (string,...)->_ (infered at line 7 char 1 - line 10 char 3)
>> but it should be a subtype of
>> ([_],...)->_
>>
>> I can't really tell what my script is doing differently than say, this
>> one: http://savonet.sourceforge.net/doc-0.9.2/radiopi.html
>> I guess my harbor_connect function is the wrong type, but why?
>> Here is my script:
>>
>> set("log.file",false)
>> set("log.stdout",true)
>> set("log.level",3)
>>
>> set("harbor.bind_addr","0.0.0.0")
>>
>> def harbor_connect(headers)
>> log(headers)
>> log("harbor connect")
>> end
>>
>> backup_playlist = playlist(reload=600,"~/grimeyblimey.m3u",conservative=true)
>> output.dummy(fallible=true,backup_playlist)
>> on_fail = single("/home/tony/samples/vox/datafruitsdotfm_processed.wav")
>>
>> dj = input.harbor("dj",port=9000,password="hackme",
>> on_connect=harbor_connect)
>>
>> source = fallback(track_sensitive=false,[dj,backup_playlist,on_fail])
>>
>> output.icecast(%vorbis, host = "localhost", port = 8000, password =
>> "hackme", mount = "liq.ogg", input)
>>
>> Here's the script on gist, so you can see the line numbers if you need
>> to: https://gist.github.com/mcfiredrill/4970578
>>
>> --
>> -Tony Miller
>> github.com/mcfiredrill
>> @freedrull
>> freedrool.us
>>
>> ------------------------------------------------------------------------------
>> The Go Parallel Website, sponsored by Intel - in partnership with Geeknet,
>> is your hub for all things parallel software development, from weekly thought
>> leadership blogs to news, videos, case studies, tutorials, tech docs,
>> whitepapers, evaluation guides, and opinion stories. Check out the most
>> recent posts - join the conversation now. http://goparallel.sourceforge.net/
>> _______________________________________________
>> Savonet-users mailing list
>> Savonet-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/savonet-users
>
> ------------------------------------------------------------------------------
> The Go Parallel Website, sponsored by Intel - in partnership with Geeknet,
> is your hub for all things parallel software development, from weekly thought
> leadership blogs to news, videos, case studies, tutorials, tech docs,
> whitepapers, evaluation guides, and opinion stories. Check out the most
> recent posts - join the conversation now. http://goparallel.sourceforge.net/
> _______________________________________________
> Savonet-users mailing list
> Savonet-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/savonet-users
--
-Tony Miller
github.com/mcfiredrill
@freedrull
freedrool.us
------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_feb
_______________________________________________
Savonet-users mailing list
Savonet-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/savonet-users
