XSLT question; DTD?

I’m trying to pull information out of some html using the XSLT patch. When I use this XSLT, I get an error:
XSLT input
(Note: I changed leading angle brackets to leading ‘[’, just so it wouldn’t look like html. Is there a better way to do that?)
[xsl:output method=“xml”

        [xsl:template  match="/">
              <xsl:value-of  select="/html"/>


On Line 4:8 The element ‘html’ is used but not declared in the DTD /Schema

What DTD?

best shot.txt (399 Bytes)

Very Interesting!!!

I was using the ‘tidy’ patch to make sure my input html was good. When I chose “auto” as the Doctype Mode (the default), it added a “Doctype html” line to the input. Since my first “select” included the word ‘html’, I went off on a blind chase for the DTD.

The solution was to set Doctype Mode = None. Now it works MUCH better. Problem solved.

if you try to validate the html agains" wich the xslt node tries to do when pin “validate on parse” is enablen, and you disallow the node to load external ressources… it will stop the parser. because the validation failed.

Tidy’s auto detect Document Type tries to find out what kind of document it is and set the right DOCTYPE declaration.


Thanks. I have it working now, and I wonder if you could help me with the next part of what I’m learning to do.

If I have a spread of strings which changes over time, how can I recognize new strings, and store them in a “seen this” spread, so I have a spread of just unique strings?

Also, is there a hash function for strings? I can see that it could be useful.

I’m really impressed with the way vvvv works. Neat stuff!


i am not shure if i understand what you want to do, but maybe the Clean (String Spreads) module from kalle can help you: kalle-Modules-String (you have to sort your spread before using it)


Thanks for the help, both now and earlier.

Clean (String Spreads) is perfect. Order log n rather than the n^2 approach my mind kept giving me.

??? I’m still not completely comfortable with how to process the urls I’m getting, one at a time. My question has to do with how to interact with the file system. I don’t think I can just throw a bunch of spreads at it, so I need some way of sequencing through a set of web reads and file writes. ???

I have midi working quite happily now. My son and I will be building an FTIR multitouch screen soon, and I’d like to figure out the best way to interface it with vvvv. I’ve been remapping my thought processes from languages like c++, python, pl1, and apl to vvvv. It’s great fun!

Where is the appropriate place to chat. I’d like to share my background and interest in vvvv. I asked in ‘shout’ if it’s possible to communicate directly with fans, but no one answered.

I’ll leave this note here since we’re communicating here, but I know it’s not about XSLT and DTDs any longer.


rphoenix, do you have skype installed? id like to exchange ideas with you about midi-note music recognition. im currently work on a live visual performance with an improvised solo piano. for that im patching a bunch of modules that regognice certain aspects of the music and that can be mapped on the fly to any visual parameter.
maybe i can also help you with some general vvvv questions. im also interessted in building a touchtable, because im working on some multiinput gui elements for live performances.

my skypename is elektromeier

hi rphoenix,
you asked:

Where is the appropriate place to chat. I’d like to share my background and interest in vvvv.

its the channel #vvvv on IRC.

unfornately our companys firewall is blocking that ports.
i have to drink a coffee with the admin…

I don’t have skype or IRC, but I’ll let you know when I do. I have email (ray.phoenix at gmail) and a cell phone 619-961-7390. Anytime is fine to call.

I’m currently listening to midi on 2 keyboards (3 channels each), 1 computer (4 channels), to music I composed algorithmically with vvvv. I like it best at 1 beat per second, very relaxing. With 3 sound sources, there’s some intersting interplay between the voices. Next I want to capture the phrases I like best and study them algorithmically. I also want to display what’s coming out so I have a visual record to study as well as an auditory one.

I’m interested in the multitouch displays as interactive surfaces for cooperative music making. A couple of videos showing some large MT displays:

A bar in a cocktail lounge – 10 meters+ long…

A wall sized display

Exciting times, yes?


it is really nice to have you here.
perhaps you don’t know that the iBar is realized with vvvv.
btw: the second link you pasted the same one…

for sure you will make your own userpage soon
just click that blue ? behind
and start editing…
it is really easy…

i hope that i am able to visit #vvvv on IRC soon again and meet you there!



I’m back online. I’ll be installing skype tomorrow, also IRC. In the meantime…

  1. We’re progressing on the touchtable. We’ve done experiments with visible leds and glass (compared to ir and lucite), and believe it will help us with debug and materials cost. We also have ir leds for later. Coupling appears to be quite adequate for the visible-glass arrangement.

  2. What is involved in writing a patch in C++? I’d like to write a hash function for strings, for example, not something I’d want to write a v4p patch for, I think.

  3. Another question… What happens if patch processing isn’t complete by the time the cycle is over?

  4. And finally, is it overloading of the video processor that causes strange vectors and areas to appear in the display, or just bad parameter values?

That’s it for now.


helo riverp.

ad 2)
“writing a patch” in c++ is not (yet) possible. we’ll let everybody know as soon as we have something in that respect…

ad 3)
the cycle is only over when patch processing is complete. you see? every single operation in a patch is blocking execution.

ad 4)
“strange vectors and areas to appear in the display” where is that? can you screenshot that?

  1. I tried IRC, no one on #vvvv. Do I have to be on some particular network, or was it just a bad time?
  2. What vvvv patch(es) exist(s) for ftir screens?
  3. I assume there is no way to write in any language? It doesn’t have to be c++ for me. I look forward to being able to write my own patch.
  4. Got it! Makes perfect sense.
  5. I’ll try to recreate it (now that I want to, it will probably elude me… :-)
  1. #vvvv is on ircnet…

ad 2) the only thing you can already do is writing FreeFrame video plugins for use with vvvv. that is nodes that per design have one videoinput and one videooutput.

  1. I’m trying to use chatzilla (a firefox irc client) and it says [dalnet, efnet, freenode, hybridnet, moznet, quakenet, serenia, slashnet, undernet, webbnet](dalnet, efnet, freenode, hybridnet, moznet, quakenet, serenia, slashnet, undernet, webbnet) are the only nets available. ??

  2. (and 2) is the vvvv FTIR interface an example of a video plugin? How is FTIR multitouch handled in vvvv?

Thanks for the answers.


ad 0) here is a list of all ircnet servers. you should have the option somewhere in chatzilla to specify one of these.

ad 1) what do you mean by FTIR?

  1. FTIR = the majority blogged frustrated total internal reflection technique. you would need the colortracker freeframe plugin in vvvv to get the coordinates of the white blotches in the video image.