Hi, I'm working with some form of abstraction which requires that I create a pd-file-specific table ($0-table) so that they are kept separate per instance of pd-script.

I've got a table I'm trying to dump like this:

bang
|
tabdump tablename
|
print

I get a nice little list of the table contents, some 24 numbers followed by a bunch of 0's.
Then I try this

bang
|
tabdump tablename
|
;$0-table $1 $2 $3 $4 $5 $6 $7 $8 $9 $10 $11 $12 $13 $14 $15 $16 $17 $18 $19 $20 $21 $22 $23 $24 $25 $26 $27 $28(

and I get
"0-table: no such object".

What exactly should I do so that I can dump the contents to a table to another table?

After writing all this, I thought to do it like this:

bang
|
tabdump tablename
|
tabwrite $0-table
|
tabdump $0-table
|
print

but the print-out from $0-table comes out empty.
What am I misunderstanding/not comprehending about how tables are written to and read from?

Ok, tabdump isn't a workable solution (pd extended only).

The $0 variable doesn't exist inside message boxes so that'd be why you're having trouble with part 2, you'd need to store $0 somewhere as you instantiate the abstraction and pass it later.
I'm not entirely sure what you're trying to do in the third part though?

@jm547ster said:

The $0 variable doesn't exist inside message boxes so that'd be why you're having trouble with part 2, you'd need to store $0 somewhere as you instantiate the abstraction and pass it later.
I'm not entirely sure what you're trying to do in the third part though?

Hi, I'm basically doing this:

write to a table, let's say it's called orig-table.
Then I'm using a select to select between 7 different tables (they'll all have different names). When that selection is made, one of the 7 tables is written to a $0-table (or similar type thing, specific to that pd-script-within-a-script). Then one of the numbers on the $0-table is banged out by sending a number to it.

Let's say if the $0-table contains 0 1 2 3 4 5 6 7, it might receive 3, so number 2, and that will later be calculated for something else.

I'm trying to keep the pd-script-within-script-tables separate from eachother, so i could have say, 32 instances of the same script, each set to a specific one of the 7 tables, and not have them overwrite eachother..

it looks to me like you have the blinkers on a bit. You might be trying to force things to work the way you want, and making them far more complicated than they really should be.

It is quite rare for something not to have a simple and elegant solution in pd.

if you post your patches here (.zip em up if there are a few) and outline what your end goal is, i think we can help you get it going in a way that gels better with the way pd is structured.

Never let things get too complicated unless you're quite sure that a complicated route is the only way possible.

Okay, I'll isolate it into a separate .pd.

While I'm doing it, I'll try and break it down in text form too.

loadbang
|
7 messageboxes which each create a single table. (ionian,dorian,phrygian,lydian,mixolydian,aeolian,locrian)

space for 7 tables

list 2 5 60(
|
unpack f f f
|
pd i-viiosc
|
output~

i-viiosc "is supposed to be" designed to receive three numbers - one to select one of the entries in a table, one to select the table, and one to provide the "basis", so to speak. So:

list 2 5 60(

would result in 2nd note selected from 5th table, and transposed by +60.

pd i-viiosc is the main issue. I have three inlets, one to receive the number-to-read-from-table, one to receive select-which-table, and one to receive the basis for the calculation.
i-viiosc also contains a simple

resultant-number-read-from-table + basisnote
|
mtof
|
osc~
|
outlet~

and that's it.

At first I thought a simple

select 1 2 3 4 5 6 7

would be enough, this would select which table to read. But then deciding how to "mute" the other tables while banging all of them with a specific table-content-number, became the issue. so now, instead of connecting the order-number-to-retrieve-from-table, i'm trying to go "well, I selected the 5th table, so bang that table to $0-table, and select the 3rd or 8th number from that table, and output that to (number-from-table + basis) and then mtof it.

I must be doing this wrong. I've tried to think about it for a while, but $0-table was the only solution I could come up with, that would allow me to have, say, 32 instances of pd i-viiosc in such a way that none of them touch eachother's tables (the 7 tables are not written to after loadbang).also, i realize that $0-table just won't work if it's not a separate .pd file, I was just trying to shove the whole thing into a single .pd.

outline:
basically if I can fetch a number from any of the seven tables, I'll then create another set of tables that will contain "which numbers to fetch". Let's say there's 4 numbers to fetch from one table, these 4 numbers will be split amongst 4 of those pd i-viioscs in such a way that can press a messagebox of numbers and out come 4 different notes, a 4 note chord, that is.

Then I'll start crafting a way to use a metronome counter to bang out specific notes from a master-table, which then toggle another scale-table's notes to come out to it's specific oscillators.

Basically the objective is to say:
1, 5, 7, 3

and those numbers will read from a table which 3 notes to come out, per number, and then talk to the scale-tables and output those notes. this would result in a chord-progression that could be controlled.

@jm547ster said:

The $0 variable doesn't exist inside message boxes so that'd be why you're having trouble with part 2, you'd need to store $0 somewhere as you instantiate the abstraction and pass it later.

Right. Do this instead:

f $0
|
$1-table ...(

edit: I was only thinking of the "$0 inside a message", not the "send a bang to a table" part; edited to be less wrong and embarrassing.

you can't "bang" tables, like you were trying to do.

i think what you're looking for is the [set tablename( message, for tabread.

does this adjustment help?

http://www.pdpatchrepo.info/hurleur/tables_within_tables-edit.pd

@mod said:

you can't "bang" tables, like you were trying to do.

i think what you're looking for is the [set tablename( message, for tabread.

does this adjustment help?

That explains what I'm supposed to do completely! Thanks! I guess from now on I'll learn to use set and just think way, way simpler than how I was trying to go about doing it.

really appreciate it. Should I switch the "tabread ionian" to "tabread table-to-be-fed" or something, so it doesn't seem like everything is read as ionian?
I mean, set just over-writes the name, so maybe tabread could even be an empty name?

Okay, by modifying the original messages to have 4 digits and doing unpack f f f f, I was able to connect the first float number to control the vradio that selects which scale to go with.

This solves everything, now I can move on to doing the tables of numbers that are called to create major/minor chords. I don't think I would've been able to come up with the "set" thing at all on my own!

@esaruoho said:

Should I switch the "tabread ionian" to "tabread table-to-be-fed" or something, so it doesn't seem like everything is read as ionian?
I mean, set just over-writes the name, so maybe tabread could even be an empty name?

yeah, as long as you use the 'set' message before reading the table, you can just leave [tabread] with any argument.
i just put 'ionian' there as a default, so even if no set message is received, it will still read one of the tables.

good luck with the rest of your patch