Streaming to SHOUTcast v1 using C#

[waynel1059]

Hi Everyone,

What I am trying to do is write a program to feed mp3s to a hosted shoutcast stream.

I am doing this in c#. I have been successful logging into my stream, passing my password and station data and I receive the "OK2" message. According to a number of articles, posts, and webpages, the next thing I have to do is send the mp3 data.

I've tried sending it as binary, it didn't work. I've tried sending it as ASCII, that didn't work.

The mp3 files are saved to my hard drive, read in, in binary, and sent to my shoutcast host. I know this data is getting there because I use Wireshark to track the the tcp connection and see the data getting there. When I check my shoutcast stats page it says that the stream is up and running, but nothing plays on my player.

Could someone PLEASE point me in the right direction to attain the information on sending the mp3 to my shoucast host? Algoritmns, block size, example code, etc........ I've been kicking this around so long now that it's driving me nuts!

Thank you in advance for ANY help!!

[DrO]

it'd be a bit easier to do it against the supported 2.x DNAS than the unsupported 1.x DNAS as you can enable the source debugging options on 2.x to get a better idea of what the DNAS is actually seeing from your connection (in addition to what you're checking on sending). as 2.x fully supports 1.x sources (however much people seem to think that is not the case).

but it sounds like you've got most things right, just ensure that you're not sending any ID3 tags and other related tag data - it just needs to be the raw audio (preferably ensuring that you're only sending whole audio frames - not doing that will break a number of HTML5 / Flash players and is something we've had to add to the recent 2.x release to improve stream playback under such cases).

[waynel1059]

Thanks for the response, the reason I have not moved to Shoutcast v2.0 is the php code that runs my station website. It would all have to be recoded. Do you have any insight on how to EXACTLY send the mp3 data?

I'm not quite sure where the header breaks and the raw mp3 data begins. Are there resources out there that you know of that I could study?

[DrO]

Originally Posted by waynel1059

Thanks for the response, the reason I have not moved to Shoutcast v2.0 is the php code that runs my station website. It would all have to be recoded.

er no. you don't have to re-code anything to be able to work with the 2.x DNAS if you're running a single-stream setup (as all api actions are treated like a 1.x DNAS with the 2.x DNAS assuming stream #1 as that's all a 1.x DNAS could do). it's only if you need to support more than 1 stream that you then might need to ensure you're specifying the 'sid' parameter on the the requests made to the DNAS.

can I ask what (person, article, etc) has made you think you'd need to re-code everything?

Originally Posted by waynel1059

Do you have any insight on how to EXACTLY send the mp3 data?

I'm not quite sure where the header breaks and the raw mp3 data begins. Are there resources out there that you know of that I could study?

• send the password followed by a CR-LF line-break (\r\n)
• wait for OK2\r\n or treat any other response as a failure to connect (the response should give a hint as to why)
• send the headers as per the HTTP spec i.e. with another CR-LF line-break (\r\n) to represent the end of the headers
• then you send the MP3 data as long as you've not received anything back from the DNAS

there's nothing special about the MP3 data, it's just as it's read from the file sans tags and all that and sent down the connection you've made.

[waynel1059]

It was a bad assumption on my part regarding Shoutcast v2 I guess.

Do you know what block size to use when sending the mp3 to Shoutcast?

[DrO]

you just keep trying to send what you've got until either you decide to stop the connection or the DNAS closes the connection (which you'll need to monitor for). there is no concept of block size (especially with 1.x) and the DNAS will just keep taking what it's given.

as long as what you're sending is valid MP3 data (and if possible you can ensure you're only sending complete MP3 frames) then it should help improve the likelihood that when the DNAS then provides that data to the listeners, then it will work (subject to my notes in my first reply re: 1.x and HTML5/some Flash players).

[waynel1059]

What you have told me is basically everything that I had learned from reading different threads and articles in the past. I do appreciate you educating me on the fact that there isn't really a need to reprogram my site based on the difference between Shoutcast v1 & v2.

What I am asking for is specific code snippets used to send the mp3s to Shoutcast. I've tried all that I've said above and am at a dead end.

Is there anyone out there that has specifics. Frame size when sending? Delay times when sending frames? I'm looking for the Exact procedural practices used.

I'm not sure what the big mystery is behind the concept of what should be a simple task of sending mp3s to a Shoutcast Steam. There are CLEAR and definitive instructions on how to do millions of things in other areas of programming on the net, but when you research writing shoutcast streaming it all gets criptic.

[DrO]

I have made it as clear as possible when I cannot provide you example C# code (which would constitute me having to write a C# source to be able to make it a viable example, which is not something I can do).


you just send the password + headers + MP3 data + dealing with any responses back from the DNAS as needed. that's it, there's nothing magic or special needing to be done. the DNAS doesn't care about framing, delays, etc as I've already said - you just have to ensure you're sending things as soon as you can and when the DNAS wants to accept things over the socket connection.

the DNAS then just shoves it into a buffer and it's down to the listener to regulate how it gets the data and how it processes what is provided. the DNAS (subject to very recent builds) doesn't do anything on the audio (and none will do any re-processing of the audio data). hence why there is basically nothing out there as there's nothing special to do. if anything, it's like a HTTP PUT with a custom header setup i.e. about as basic as you can get.


you can always try looking at one of the open source broadcasting tools / libraries to see how they're doing it and try to port that to C#. that's about as much as I can do / suggest than what I've already said. but other than seeing what the DNAS is reporting &/or dumping the output and checking that it's correct, without anything to actually test against (as it could be the stream is correct and it's something with the player you're trying to use), there's little else I can suggest.

[jaromanda]

Originally Posted by waynel1059

but when you research writing shoutcast streaming it all gets criptic.

Are you able to read and understand C/C++ code? The source code for AltaCast (derivative of EdCast) is available at https://github.com/DustyDrifter/AltaCast

[umitemree]

Hi!

I was thinking same thing past couple months. I haven't tried it but I'm 100% sure that's how it'll be done. Download Wireshark for sniffing packets your current DSP (SAM? Maybe?) sends to DNAS v2 or v1 and DNAS sends to your client. If you are using transcoder then do it the same thing for it. Then apply everything to your C# Code.

[jaromanda]

Originally Posted by umitemree

Download Wireshark for sniffing packets your current DSP (SAM? Maybe?) sends to DNAS v2 or v1 and DNAS sends to your client.

What a brilliant idea, pay through the nose for SAM broadcaster to sniff packets so you can write your own source client

The only flaws in that are

a) SAM broadcaster
b) if you're going to go for v2 (ultravox 2.1 protocol) it's fully documented (I wrote a nodejs source client based on the documentation)
c) SAM broadcaster
d) sniffing packets gives very little info compared to the documentation mentioned above, or even the source code in another language (if you're able to understand the other language); and finally
e) SAM broadcaster

[waynel1059]

I have Wireshark and SAM Broadcaster. I have sniffed the packets coming and going. I am able to connect to my Shoutcast Stream in the exact manner that SAM Broadcaster does. I receive the ok to send, BUT, when I send the mp3 data I am either sending too much of the header, or sending too fast or too slow to the server. What I really need is to see some example code of sending the mp3 to shoutcast. The handshake between client and server is made just fine. I can see the mp3 data streaming to my shoutcast stream, but hear nothing playing on a player connected to the stream. I know the player works because I can hear the stream when I use SAM to play the mp3 file.

This is extremely frustrating. Why wouldn't there be more information out there to perform what seems like a simple task?

I'll check that AltaCast out thanks!

[DJ-Garybaldy]

Originally Posted by jaromanda

The only flaws in that are

a) SAM broadcaster
b) if you're going to go for v2 (ultravox 2.1 protocol) it's fully documented (I wrote a nodejs source client based on the documentation)
c) SAM broadcaster
d) sniffing packets gives very little info compared to the documentation mentioned above, or even the source code in another language (if you're able to understand the other language); and finally
e) SAM broadcaster

Hahahahaha! Nothing says "FLAW" quite like SAM Broadcaster.

[waynel1059]

Here is a snippet of the View log file after connecting to Shout Cast and trying to have a player connect to retrieve the stream and play the music.

Can anyone translate/tell me what is going wrong?

<09/01/15@21:02:58> [source] connected from 24.188.98.3
<09/01/15@21:02:58> [source] icy-nameJ Rider @ All Request Radio 4U ; icy-genre:Various
<09/01/15@21:02:58> [source] icy-pub:1 ; icy-br:128 ; icy-url:http://www.allrequestradio4u.com
<09/01/15@21:02:58> [source] icy-irc:#audiorealm ; icy-icq:NA ; icy-aim:NA
<09/01/15@21:02:59> [dest: 24.188.98.3] starting stream (UID: 1677)[L: 1]{A: NSPlayer/12.00.10011.16384 WMFSDK/12.00.10011.16384}(P: 1)
<09/01/15@21:02:59> [dest: 24.188.98.3] connection closed (0 seconds) (UID: 1677)[L: 0]{Bytes: 18723}(P: 1)
<09/01/15@21:02:59> [dest: 24.188.98.3] starting stream (UID: 1678)[L: 1]{A: NSPlayer/12.00.10011.16384 WMFSDK/12.00.10011.16384}(P: 1)
<09/01/15@21:02:59> [dest: 24.188.98.3] starting stream (UID: 1679)[L: 2]{A: NSPlayer/12.00.10011.16384 WMFSDK/12.00.10011.16384}(P: 2)
<09/01/15@21:02:59> [dest: 24.188.98.3] connection closed (0 seconds) (UID: 1679)[L: 1]{Bytes: 15819}(P: 2)
<09/01/15@21:02:59> [dest: 24.188.98.3] starting stream (UID: 1680)[L: 2]{A: NSPlayer/12.00.10011.16384 WMFSDK/12.00.10011.16384}(P: 2)
<09/01/15@21:02:59> [dest: 37.59.25.124] starting stream (UID: 1681)[L: 3]{A: SHOUTcast Directory Tester}(P: 3)
<09/01/15@21:02:59> [dest: 24.188.98.3] connection closed (0 seconds) (UID: 1680)[L: 2]{Bytes: 15800}(P: 2)
<09/01/15@21:02:59> [dest: 24.188.98.3] connection closed (0 seconds) (UID: 1678)[L: 1]{Bytes: 161500}(P: 1)
<09/01/15@21:03:00> [dest: 24.188.98.3] starting stream (UID: 1682)[L: 2]{A: NSPlayer/12.0.10011.16384 WMFSDK/12.0}(P: 1)
<09/01/15@21:03:00> [dest: 37.59.25.124] connection closed (0 seconds) (UID: 1681)[L: 1]{Bytes: 35947}(P: 3)
<09/01/15@21:03:00> [dest: 24.188.98.3] connection closed (0 seconds) (UID: 1682)[L: 0]{Bytes: 49780}(P: 1)
<09/01/15@21:03:00> [yp_add] yp.shoutcast.com added me successfully
<09/01/15@21:03:00> [dest: 24.188.98.3] starting stream (UID: 1683)[L: 1]{A: NSPlayer/12.0.10011.16384 WMFSDK/12.0}(P: 1)
<09/01/15@21:03:00> [dest: 24.188.98.3] connection closed (0 seconds) (UID: 1683)[L: 0]{Bytes: 57040}(P: 1)
<09/01/15@21:03:00> [dest: 24.188.98.3] starting stream (UID: 1684)[L: 1]{A: NSPlayer/12.00.10011.16384 WMFSDK/12.00.10011.16384}(P: 1)
<09/01/15@21:03:01> [dest: 24.188.98.3] connection closed (0 seconds) (UID: 1684)[L: 0]{Bytes: 15819}(P: 1)
<09/01/15@21:03:01> [dest: 24.188.98.3] starting stream (UID: 1685)[L: 1]{A: NSPlayer/12.00.10011.16384 WMFSDK/12.00.10011.16384}(P: 1)
<09/01/15@21:03:30> [dest: 24.188.98.3] connection closed (28 seconds) (UID: 1685)[L: 0]{Bytes: 4421166}(P: 1)
<09/01/15@21:03:30> [dest: 24.188.98.3] starting stream (UID: 1686)[L: 1]{A: NSPlayer/12.00.10011.16384 WMFSDK/12.00.10011.16384}(P: 1)
<09/01/15@21:03:30> [dest: 24.188.98.3] connection closed (0 seconds) (UID: 1686)[L: 0]{Bytes: 51151}(P: 1)
<09/01/15@21:03:53> [source] source dropped connection. disconnecting.
<09/01/15@21:03:55> [main] connecting to yp.shoutcast.com to remove myself [no source]
<09/01/15@21:03:55> [yp_rem] yp.shoutcast.com removed me!

[jaromanda]

Streaming to SHOUTcast v1 using C#

trying to have a player connect to retrieve the stream and play the music.

playing a stream wont help you learn how to source a stream

[waynel1059]

Shoutcast supposed experts quoting? That is very helpful!

I come here for help I get sarcastic remarks.

Thanks!

[DrO]

the log output doesn't give much to work with, hence why I've said in my first reply to use the 2.x DNAS with debugging enabled as that gives more information on what the DNAS is trying to process and may even show what the issue could be (when there's no url to a DNAS to check the output off).

[waynel1059]

Does anyone have an algorithm that calculates packet size and frequency of sending the mp3 data to Shoutcast? Also, is it sent in binary format?

[jaromanda]

packet size

I tend to either use:

1) max packet size negotiated in the initial connection phase; or
2) A single complete chunk of mp3 (or HE-AAC for that matter) audio, if I'm doing the compressing and not simply passing on already compressed audio (subject to the limit in point 1)

frequency of sending the mp3 data

for a 64k stream, send 8kBYTES per second,
for a 96k stream, send 12kBYTES per second,
for a 128k stream, send 16kBYTES per second,

There's a formula for figuring it out, I just can't seem to find it

[waynel1059]

Do you have examples of your code Jaromanda? That you would be willing to share?

Your response was helpful. I am able to send the mp3 data, but my stream shuts down as soon as I do. I think I'm sending it in the wrong format.

[jaromanda]

Not any more, I gave up on edcast years ago and two computer upgrades later I've lost all the source code. You may be able to find it on sourceforge or maybe google code

[waynel1059]

Through more trial and error I have found that my connection is closing just before sending the Mp3 data to the shoutcast server. I am getting an error 10053. It's a cryptic error even in Microsoft forums. Any info on circumventing this error would be appreciated.

[jaromanda]

WSAECONNABORTED
10053



Software caused connection abort.

An established connection was aborted by the software in your host computer, possibly due to a data transmission time-out or protocol error.

[waynel1059]

Yes, Thanks Jaromanda, that is exactly what I found when I googled the error also. There is no advice anywhere on how to circumvent the error. That is what I am looking for.

Expect that if I posed the question, I've looked on the internet for an answer. I am coming to the 'professionals' that are supposed to be here for help.

Re-quoting or reposting the obvious may get you more points on here as a Major Dude, but does little to help.

[DrO]

as was stated in the first reply to this thread, you need to be using a current DNAS and not 1.x so you can enable the debugging options to see why things are failing and so those of us who might be able to give you a reason have that information.

the 1.x DNAS is not going to give that information and if that's all you're going to work against then there is no 'official' help going to be coming as the 1.x DNAS is not supported (and as it does not provide detailed debug output, it's hindering you finding a cause of your issues).

so if you want official help, then you need to do as was initially asked to do.

[jaromanda]

Originally Posted by waynel1059

Expect that if I posed the question, I've looked on the internet for an answer.

and yet, here you are because you can't find someone elses code to steal use in your little toy program

Originally Posted by waynel1059

Re-quoting or reposting the obvious may get you more points on here as a Major Dude, but does little to help.

wayne - the professionals here have no idea what monkey code you write, so stick your attitude where the sun don't shine, and learn to at least read a real programming language so you can duplicate oddsocks (bad) code in the toy language you write in

[waynel1059]

Very professional reply Jaromanda. I don't want to 'steal' code, I am attempting to learn a process.

Thank you Dr. O I will move to v2.

[jaromanda]

Why do I need to make a "professional" reply to your "I can't be bothered doing any research, someone do it for me" attitude? I have no affiliation with winamp/shoucast, so I can be as rude to you as you are to me.

Good luck on Stack Overflow

http://stackoverflow.com/questions/3...ast-mp3-server

I've posed this same question in the winamp forums only to be pacified with general information that the responders received from the radiotoobox and smackfu sites or forums.

Really?

They seem to enjoy responding to questions with expressions like "when I did this it worked", etc... They never give a straight answer on procedure. They just like to hear themselves talk and repeat VERY general information.

Aww, nobody will do the coding for you, suck it up, princess

[waynel1059]

Lmao, Forum King? More like a Forum Queen! You should respond less and get on with
your transformation like Caitlyn Jenner!

Just stay out of the thread asshole if you have no insight to provide.

[jaromanda]

First thing I did in this thread was give you a link to altacast (edcast derivative) source.

I showed you how to "calculate" the rate of sending data

I made sarcastic comments in response to some dimwit who thought it would be a good idea to BUY SAM Broadcaster just to wireshark the interaction (guaranteed NOT to be helpful, as SAM Broadcaster is a piece of SHIT)

I mistakenly pointed out what that error was, because I thought you couldn't find ANY info about it, my mistake, since you couldn't find ANY code to copy, I assumed you wouldn't be able to find out ANY description for that error

I started out trying to help, now I make it my mission to be a pain in your arse, Wayne, you little bitch

[DJ-Garybaldy]

(guaranteed NOT to be helpful, as SAM Broadcaster is a piece of SHIT)

About the only thing i agree on.....

Not any more, I gave up on edcast years ago and two computer upgrades later I've lost all the source code. You may be able to find it on sourceforge or maybe google code

So I'm guessing RioCast won't ever see the light of day now? That was your project right?

[jaromanda]

Originally Posted by garybaldy72uk

So I'm guessing RioCast won't ever see the light of day now? That was your project right?

yeah, riocast was my project

I actually did finish riocast - the core is written in javascript, encoding done by ffmpeg (or any other suitable program), a couple of other external programs to source the actual audio from a sound card or winamp

I was trying to write a "source" plugin for radiodj, but could not find any documentation on doing so, so I gave up the whole lot

[DJ-Garybaldy]

I was trying to write a "source" plugin for radiodj, but could not find any documentation on doing so, so I gave up the whole lot

You should have asked Marius on the RadioDJ forums.... he's the man with the source code for plugins.

[jaromanda]

Originally Posted by garybaldy72uk

You should have asked Marius on the RadioDJ forums.... he's the man with the source code for plugins.

I may yet do that, when I get back into it - the wife has been hankering to get back into it

[DJ-Garybaldy]

What RDJ really needs is someone to write a streaming plugin that doesn't cause the program to crawl when there's a connection issue..... It's the only downside to the Altacast plugin.

The developer of Altacast has a few things going on in his life right now and can't give any more updates until that's sorted.

If i knew where to start i'd write something .... but I don't.

[jaromanda]

Originally Posted by garybaldy72uk

The developer of Altacast has a few things going on in his life right now and can't give any more updates until that's sorted.

Oddsock's curse!!!

p.s. I'm "not allowed" to post on this thread any more ... pffft

[jaromanda]

Originally Posted by garybaldy72uk

What RDJ really needs is someone to write a streaming plugin that doesn't cause the program to crawl when there's a connection issue..... It's the only downside to the Altacast plugin.

The way riocast works is, the "DSP" just pumps out audio, the streaming program is completely separate, so if there are connection issues you'll get lost audio, but that's all, no negative impact on the source

but many people don't like the idea of separate processes for some reason. having to start two or more programs at a time seems to be too difficult

[DJ-Garybaldy]

having to start two or more programs at a time seems to be too difficult

Oh I've had all the "Why do i need a separate encoding program" brigade to deal with over the past 3/4 years.

It's no more difficult than having them built into the program IMHO.

Currently using what was meant to be a temporary streamer plugin written by Marius which is actually working OK (Mp3 128K 22khz) but it could do with a few tweaks.

When it comes to standalone streaming software the options are somewhat limited....It's a right pain in the BUTT!! (See what i did there)

[jaromanda]

Originally Posted by garybaldy72uk

It's a right pain in the BUTT!! (See what i did there)

Saw it

[jaromanda]

radiodj DSP "system" is BASS_WADSP? No "native" interface (i.e. non-wadsp?)