Can't dynamically name a fluid.dataset~

[rodrigo.constanzo]

Not sure if this is an intentional restriction, but the name attribute to fluid.dataset~ doesn’t allow you to change it. This happens either with the attrui or with name $1.

----------begin_max5_patcher----------
396.3ocyT1taCBBFE925Uggre5ZJTmS2sxRyBUYcznfAvNWZ5t1G7h1tO5rM
sMYKlPjiuxgG3.aBCPKjcLMJ5gnGiBB1DFD.RNgf99AnZZWQEUCkgpYZMcIC
E6+lg0Y.cZSS0N0Fpo3EtX4SJVgwO5jzISiilQf1LWKw9dz79+fWBihbwpay
GFEQaMWTwLfwj8hxVyfJtW0KYdqg4cCghl69x1vPWS7EBmfVyhtAeT7Ru20d
2cigW1+O7VPE1mQnKOY+lWR9Xzk92Qmf8pcF7C3dtpkWNojZnZl48iBIdJPI
lLJl3ChI9fXN62vztruDE6lgRpw8x4wM0XTs7iBlOaRvIiwEYTtbNs67.5R1
SAuQUbw2u7Al9N8uBrV1pJFFp9SQQ6InjoMbA0vkhOUC1WyAWROUexOAextB
9jdB9PtB9PNq0M+1k8J90LktuXvBaRbkDxDYwPWtv2ENwiTr07g58JTkMcZr
QyVELsPcoIH+uJKYJgMH6RQgN3rVBobWZS2P8b.GFB2F9A.zF6Hb
-----------end_max5_patcher-----------

The use case here is instantiating a random/unique fluid.dataset~ inside an abstraction, (as detailed in this thread).

I can give it a name on instantiation (e.g. fluid.dataset~ #0), but I want to give it a nicer random name, or may want to programmatically name several ones.

Alternately I can just fluid.dataset~ generate its own random name (as it does if you instantiate one without a declared name), there’s then no way of querying what that random name is…

[weefuzzy]

@rodrigo.constanzo

Not sure if this is an intentional restriction, but the name attribute to fluid.dataset~ doesn’t allow you to change it. This happens either with the attrui or with name $1.

Yes, it’s intentional. You can change which dataset a box points to with refer <name>

there’s then no way of querying what that random name is…

getname returns the name…

----------begin_max5_patcher----------
365.3ocoRtsaCBCCF9Z3oHJWypBPoG1qxT0TJ3wREjfxgNlp5d1GwAl15Asp
1aB3e6D6OaeHNhtU0CFJ4YxKjnnCwQQnjWHZzNh1x6Ka3FLLZKXL7ZflD7Yg
dKpK4s.wwVlyXKXY4S9aDRnT4jXPYihRWqP1.VyIhJmcRMcTUTgutZ6tmJld
yPX1O6fPgSojMit531x2Ex5W0PoM3cNaFKgjxV6+rJ0elWLiQ13uww3X+QxC
BeMX87SeH9xuO9xPjxV5OKJv+ytS7jvGCExYz8ViSTMqha4Fv90EgL8hPleN
joWExsbYMMwmME25+4FndQQX1xtH138wMvS1vw2yq+2dgQ4zkSkyXwRx9oJp
.iUH4VgR9qXF1JIoWoaeq4YXxSX+SdRCwbFd7tt8f1LFLlhgg5Nk1atJAMEx
fItJR0vdwT7LTgqGZu1gdqSikEsewbZ3ppJPKcBbFG6gaHk3Biee2zwCbf6U
wGi+F7MVCvF
-----------end_max5_patcher-----------

[rodrigo.constanzo]

@weefuzzy

Yes, it’s intentional. You can change which dataset a box points to with refer <name>

That doesn’t work if that other dataset doesn’t already exist. But I can’t give the other dataset a name for the same reason. So if I want to create a new dataset but don’t have a name for it, its the same problem.

@weefuzzy

getname returns the name…

Handy!
That doesn’t show up in the “right-click input” thing, the helpfile, or the reference page though.

[weefuzzy]

@rodrigo.constanzo

That doesn’t work if that other dataset doesn’t already exist. But I can’t give the other dataset a name for the same reason. So if I want to create a new dataset but don’t have a name for it, its the same problem.

So you want to dynamically create datasets?

@rodrigo.constanzo

That doesn’t show up in the “right-click input” thing, the helpfile, or the reference page though.

It’s magic that Max adds for attributes if you have a dump outlet

[rodrigo.constanzo]

@weefuzzy

So you want to dynamically create datasets?

Well, I will manually create the dataset by creating an object called fluid.dataset~. I want to dynamically name it. So like, a patch may have a few of these “unnamed” fluid.dataset~s which I’ll (programmatically) name unique identifiers + strings.

@weefuzzy

It’s magic that Max adds for attributes if you have a dump outlet

Nice. Is that like common knowledge? Never in a million years would I have started typing arbitrary requests to an object that doesn’t say what it’s list of messages are.

[weefuzzy]

@rodrigo.constanzo

Well, I will manually create the dataset by creating an object called fluid.dataset~. I want to dynamically name it. So like, a patch may have a few of these “unnamed” fluid.dataset~s which I’ll (programmatically) name unique identifiers + strings.

Ok. Not possible, I’m afraid, but what’s wrong with the unique names that come with putting no name in?

@rodrigo.constanzo

Nice. Is that like common knowledge? Never in a million years would I have started typing arbitrary requests to an object that doesn’t say what it’s list of messages are.

It’s certainly true for all the jitter objects, not sure what of the newer Max object use dump outlets. Will make sure we add to docs somewhere, because evidently not common enough knowledge. It applies to any attribute on one of our objects that has a dump outlet at the right

----------begin_max5_patcher----------
455.3ocsTssTCBCD8Y3qHSdF6j.EZwm7+vwwIkFqoCjvjDTbbzucS1.dE0Zw
9BgrYyd1ytmMOFGg2n54FL5bzknnnGiih.SdCQC6ivMr9pZlAbC2vMF1NNNI
blk2aGra0hJT43AxtFgrlagak9lQUmczJcvpXKDA0l8mQWNd+fe1GZ4grCiQ
WMbTKyVcqPt6ZMuxFNcYd5BRBhlm4Wxo9uoNSnq7W4o3X+mjYRwcbafkyjio
GGGyxV5o0p0.EWOKJJ426Rkuvvap6DaWro6Fkz3HPsnh+L5her0Rmj1YSPax
2R6ML4NbhGdEy5+4PpCT5JXon7jzrkrFNpqbUIgTjud07Z4EGWGmRAgLMCZ1
TR4oRV6I67X3QN2RSIdNU.Su44mBM8dgcQCyof6+C52zIzueKCcHb8HBGFeo
jrvBYRFCQ.WKje9oYHhd6erLXTc5pwrY7IFD40DYK2XERlUnju2IRvoIK0GL
R9fj9aH4TGH5bQJDieDnh+AbVdHktuT4BsLVa6cbsYvY.BmFcuR62tNA1Jjg
sfDCq42IF8m.VXZmnw5TLcZHsv8EggKbiZKWK6DfrM1SNGjf92O+ZZYAd.iI
wOE+BTdqUuJ
-----------end_max5_patcher-----------

[rodrigo.constanzo]

@weefuzzy

Ok. Not possible, I’m afraid, but what’s wrong with the unique names that come with putting no name in?

That works perfectly fine actually. Just didn’t know about getname when making this thread.

I don’t do a lot of jitter stuff really, so messages for dump-y objects weren’t a given for me.

[jamesbradbury]

Yes, to be fair to @rodrigo.constanzo I had the same epiphany at the hands of @weefuzzy at some point. It’s a convention that is quite useful, at least, you really are pleased when you know you need it and you are granted the wisdom to getstuff.

[marco.donnarumma]

hey @weefuzzy (long time no see!!)
sorry to resurrect this old thread, I’m using fluid.dataset in some abstractions in Pd and I’m having the exact same issue of @rodrigo.constanzo.
But, getname doesn’t seem to work in Pd (no method for ‘getname’). Is it a different query, or how else I can get the random name of the dataset?

Thanks!

[weefuzzy]

Hey @marco.donnarumma

Great to hear from you! Hope you’re thriving. Not on here so much at the moment, as I get used to a new gig…

Presently the PD objects don’t offer the same get<x> magic, because we were just piggy backing something built in to Max, and have a less feature-ful hand rolled attribute system in PD. No reason in principle (IIRC) that it couldn’t be added though.

If you want, I’ll put an feature request on Issues · flucoma/flucoma-pd · GitHub or, if you’re Githubbed up, feel welcome to do so…

O x

[marco.donnarumma]

Hey @weefuzzy , likewise! been following your work since Edinburgh

Ok, I see about the issue. I’ve added a feature request on github, thanks for the info!
This is kind of a showstopper for the instrument I’m developing at the moment, and I have a premiere with it on 22nd September. So I’m wondering, (also channeling @tremblap here) can I perhaps sponsor financially the implementation of this feature so as to speed it up?

It’s an idea I got from the Pd crowd, worked fairly well there as a method to keep development going.

[tremblap]

@marco.donnarumma

can I perhaps sponsor financially the implementation of this feature so as to speed it up?

It is a model we never tried… although I have bribed @weefuzzy with various quality drinks

[marco.donnarumma]

ah! good to know he is bribable

[tremblap]

May I dare say, since it is a public forum, only by dear friends and/or close, long-term, mischief partners I have evidences I am at least one of these

[tremblap]

As discussed on GitHub, the trick $0-variable works in Pd but we haven’t implemented the getname method indeed.

[marco.donnarumma]

thanks @tremblap for updating the post, and sorry for my wrong report on $0.