I've been meaning to bring up this issue for years, but never manage to remember: I can never seem to manage to hide the name of an array. Nothing changes at all if I right-click on the array and select "Hide object name and arguments"...the name just stays there:

what am I missing?

@yannseznec I think what you actually hide in this case when you click the 'hide object name' checkbox is the name of the subpatch which contains the array, which has no name. This is why you get two dialogs when go to the array properties, the one is the properties for the array, the other is the properties for the subpatch.

ah ha that makes sense. so...is there no option for hiding the name of the array?

@yannseznec Can you post the part of the patch?
Normally yes.... the array name can be hidden.... and the canvas name too if it is being shown through a GOP.
I am fairly certain that the graph (subpatch) is always named "subpatch" and is hidden.
David.

@yannseznec As far as I know the only options are naming the array with unprinting characters, which is annoying for many reasons, or hide the name with a canvas or the like which won't work in your situation. But you could use data structures and [plot] instead and throw together your own array abstraction.

sure, the whole patch is here: https://github.com/yannseznec/ys.granular

@whale-av so if the array name can normally be hidden...how do I do that? I couldn't find anything in any of the reference files about that. or, in fact, changing colour or size of the black border, or label font size or colour.

@yannseznec Sorry...... I was wrong..... not a good day........
David.

You might be able to accomplish something like this by putting a data structure in a graph manually yourself.. I'm inexperienced w/ that but I think it is possible..

@yannseznec with the later versions of Pd it is possible to change the color of the graph-array which can allow a black canvas background to hide the unwanted array name.
you would need the place the canvas first then cut and paste the graphs and the bang over the top and send the color message to the graph-array

@Balwyn oh hey that's a pretty clever trick! I'll do that for now.

in the longer run, I feel like it would make sense for the border, background, and label to all have editable colors as well! maybe I'll stick a feature request on the main Pd code repo one day.

@yannseznec or put a white canvas on top of it.

... but as soon as the array is recreated for example by [;arrayname resize $1( , the name appears in front of the canvas again.

EDIT: recreated! not redrawn.

in response to this thread I thought I would give graphing audio in a data structure a go.
This is my result
scope-trace.pd
Edited 11/08/2022 fixed an error where I used x y x1 y1 instead of x1 y1 x2 y2 in the template which caused the need to adjust the y position of the plot by -1.5. the file and pic have been updated. I've also added a zero to the end of [plot vector c3 1 y] to centre the y axis

maybe more useful to the @yannseznec ys.granular.pd would be this kind of plot
This one fits the whole table to the 1024 element display, is visually scalable and has a milliseconds played cursor

Edited 11/08/2022 fixed an error where I used x y x1 y1 instead of x1 y1 x2 y2 in the template which caused the need to adjust the y position of the plot by -1.5. the file and pic have been updated. I've also added a zero to the end of [plot vector c3 1 y] to centre the y axis
here is the new pd file wave-display.pd

wow looks amazing! I will definitely have a play with this.

@yannseznec There is another method to hide the names of arrays. Rename the arrays with symbols containing only blanks. There are at least two different blank characters that can be used in Pd: ASCII decimals 32 and 160. So different names can be created by different length or different combinations of 32 and 160 blanks.

It would probably necessary to keep track of the names somehow, maybe by storing them in a list and get them by index when needed. And the patch could get somewhat un-intuitive. But it is a possible way.

If all related array objects stay unchanged, it would be enough to set them up with receives, so on load they could all receive the individual name of the according array.

The method is also described and used in this thread: https://forum.pdpatchrepo.info/topic/10854/xy-abstraction-to-get-mouse-click-and-drag-coordinates-vanilla/5

Data structure arrays have the benefit that they cannot be changed by mouse if this is wanted. So this could be nice for just displaying data.

@ingox 32 no longer works, with the addition of the list box 0bj in pd 52 all spaces (ascii 32) are escaped so symbols with spaces can be differentiated from lists, they made this behavior universal. 160 does work, as does '194 160' the unicode non printing character. I did not know about 160 on its own, that save me some typing. They also snuck in an edit method for arrays along with a few other new methods in 52.

"New method for garrays to set colors, line width, visibility, display style and disable mouse editing (documented in 2.control.examples/16.more.arrays.pd). Also added more flags to the "plot" object that plots arrays in data structures (by Christof Ressi)."

But I think @Balwyn is well on the way to making the old fashioned vanilla array a thing of the past.

@oid Thanks, didn't know that. But the xy patch in the other thread still works without changes in Pd 0.52 using 32 and 160 ASCII spaces in array names and no escaping is shown.

Well, i have to say in my example the arrays are being created by dynamic patching. Renaming arrays via message works, but it will not be displayed. This is probably a bug.

@ingox I think this might be system and or locale dependent, I get escapes with 32 160. Plain 160 is probably the safest way to go for portability since it is a non-breaking space in both UTF-8 and Unicode and should work on most system and still work if pd goes unicode. Apprently 194 160 is UTF-8 and not unicode as I had said, but it is the same character as 160 (a non-breaking space). I half suspect that 32 160 is being seen as a double space by pd on some systems and the 32 is dropped as pd tends to drop double spaces leaving you with just the 160s and no escaped 32s, but the other half of me suspects I am overlooking something.