Frank wrote:
Why care? [about how efficient pipe is] There's not much you can do about it, if it's too slow, except writing a better [pipe] in C/Lua/etc.
Well that's exactly why I need to know: to decide whether I need to write a "better" [pipe] in Lua or not!!
An approach carmen did in the past is to use list2symbol and symbol2list (the latter an external)
Yes that's an interesting solution! I usually try to go as far as possible in pd-vanilla without any external.... but maybe I've reached the "as far as possible". You may argue that I am reluctant to use an external while I would use pdlua, but one thing is to install an external that solves a specific problem, another thing is to install one single extension mechanism (pdlua) that allows you to develop any "externals" (especially when we're talking of externals for manipulating data, not for using hardware etc).
I think it's a pity that PD Vanilla doesn't include [symbol2list] or any objects capable of "breaking" symbols a little bit (with which one could build a symbol2list abstraction and maybe more), since it does give the possibility to join symbols.
Note that with [makefilename] and using the standard syntax of sprintf-like funcions (namely %.<n>s where <n> is a number) you can take the first n characters of a symbol: if only there were a way to take the remaining characters (or a way to know a symbol length + a way of taking the last m characters), one could build a symbol2list abstraction, or split a symbol into its characters etc.
.....well now I'm thinking the length of a symbol _can_ be calculated, in a loop which at iteration k takes the first k characters and compares the result with the whole symbol..... so the missing piece is the possibility to take the last n chars... or is it not?
Thank you very much for the help.
Ciao m.
On Thu, 2008-03-13 at 10:49 +0100, matteo sisti sette wrote:
I think it's a pity that PD Vanilla doesn't include [symbol2list] or any objects capable of "breaking" symbols a little bit (with which one could build a symbol2list abstraction and maybe more), since it does give the possibility to join symbols.
you're speaking from my heard, dude. [symbol2list] or any other mechanism to split symbols is on the top of my 'most wanted internals' list. i tried to bring that up at pd-conv07, but i wasn't good at all in explaining and unfortunately rather caused confusion than making my point clear. but yeah, i think you are very right: since it is possible to join symbols, not being able to split them is an obvious lack.
roman
___________________________________________________________ Der frühe Vogel fängt den Wurm. Hier gelangen Sie zum neuen Yahoo! Mail: http://mail.yahoo.de
Hallo, matteo sisti sette hat gesagt: // matteo sisti sette wrote:
You may argue that I am reluctant to use an external while I would use pdlua, but one thing is to install an external that solves a specific problem, another thing is to install one single extension mechanism (pdlua) that allows you to develop any "externals" (especially when we're talking of externals for manipulating data, not for using hardware etc).
I think it's a pity that PD Vanilla doesn't include [symbol2list] or any objects capable of "breaking" symbols a little bit (with which one could build a symbol2list abstraction and maybe more), since it does give the possibility to join symbols.
I very much agree that pdlua is a wonderful extension ("best thing since [list]") that should become one of the standard externals to learn for every Pd user, because it can be used to implement all this missing (message) functionality of vanilla Pd while still requiring only a single tiny binary to install.
Some people may miss [s2l] in vanilla Pd, others miss a [pipe] for variable length lists or a multi-period [metro], a flexible markov chain object and whatnot. With pdlua most of this is almost trivial to write and can be installed without the need to compile anything once pdlua is installed.
Frank Barknecht _ ______footils.org__
On Thu, 2008-03-13 at 14:19 +0100, Frank Barknecht wrote:
I very much agree that pdlua is a wonderful extension ("best thing since [list]") that should become one of the standard externals to learn for every Pd user, because it can be used to implement all this missing (message) functionality of vanilla Pd while still requiring only a single tiny binary to install.
yo... i totally agree...
Some people may miss [s2l] in vanilla Pd, others miss a [pipe] for variable length lists or a multi-period [metro],
those can all be done with built-in objects already, as far as i can see (with more or less effort, though).
a flexible markov chain object and whatnot.
to "whatnot": i still think, that some mechanism to split symbols isn't just some arbitrary wish, because yet there is _no way at all_ to cut a symbol into pieces. no matter how much effort you put into it, it's not possible right now. the ability to create arbitrary symbols out of lists (since pd 0.40) is only half valuable without the opposite function. actually, i wouldn't care much about how it'd be implemented: something like [symbol2list] or something like [any2string], but at least the ability to apply _some_ processing on symbols would complete what was started with [$1$2$3( and became possible with the new methods shown in [pd changing-messages] for the message box.
btw: [symbol2list] and [netclient] are currently the only externals used by netpd (i would even have removed [symbol2list], if there'd be a way).
With pdlua most of this is almost trivial to write and can be installed without the need to compile anything once pdlua is installed.
not that i am any familiar with lua yet, but it seems to me that pdlua would magically solve all of those issues.
roman
Telefonate ohne weitere Kosten vom PC zum PC: http://messenger.yahoo.de