Pretty much I just grovelled through what I could identify and found things like ‘cfoulc cf’ and ‘kockie69 SquinkyVCV-Main’ as far as stuff I wanted to be able to include. I’ve already got SurgeXT, Fundamental and mine. Looks like the script is working, which is normal for baconpaul
Oh, no, wait, it’s absolutely not. It’s trying to do it as if it was on Paul’s computer. I may or may not be able to fix that yeah, it wants to be on Paul’s computer and it wants the SDK. False alarm, I got nothing
I exported the wrong dir, I just have to figure out whether it wants to point that at the program, or the SDK (which I do have)
Yeah, I’m nowhere, I get a lot of ‘already up to date’ and ‘make: jq: Command not found’. I found the slugs but can’t simply export RACK_DIR as the location of my Rack SDK, and I’m not competent to fix this. I will go without modules for those I can’t find in your google drive
chris: just install the x86 version and run under rosetta. ARM is still pretty dev-like and lots of the user gestures - like the library! - aren’t ready yet.
naw, I’m fine for now. I got Audible and Befaco and Surge XT and Airwindows and Fundamental, it should be fine for now
Correction, I got them and it turned the downloads into a folder, and updates grabs a new thing and re-does the folder, but it creates a plugin.so file and not a plugin.dylib, and nothing shows up. Can’t go no further right at the moment, see you folks later in the day. Paul, weren’t you off doing stuff today? Don’t worry yourself about this. It’s just Rackisms
yep I agree there is a DB with primary and unique keys somewhere, but it implies a date where someone else names a repo rack-modules again … easy enough to fix by a target git argument, if the git pull is not automated. https://library.vcvrack.com/KRTPlugina 404
Yep, but GitHub - netboy3/21kHz-rack-plugins at 0bee82247151e2fe2885f1e15fbbd74ddb4f48d0 isn’t named by slug, so if the optimization is load all, compile all and the slug is not used for uniqueness of the git clone then later additions can fail the process with a git error. OK, so not a massive security issue, but a pedantic edit minor.
Yep, it’s only manual downloading for development test building that might be an issue. Consider it closed.
The conversation really doesn’t belong on this thread but if you look at the repo I linked it has renamed the submodule to the slug name so the git modules status in the library really is the definitive way to map from git endpoint plus hash to slug and code
Maybe open a new thread if you still have questions on this though so we can keep this one on airwindows pls?
Ah, this “Lock effect type” is killing me. I’m not about suggesting to drop the function but @jinx6568 & @baconpaul could you please tell a bit about the considerations behind this feature? Why is it needed?
Would you please consider also a hotkey (i.e. ‘L’) to toggle between locked/unlocked?
My previous workflow:
I read about an interesting effect
I load it
Ctrl-R to check 3-6 variation how it sounds
repeat
My current workflow:
I read about an interesting effect
I load it
Ctrl-R to check some variation how it sounds
I swear cos I forgot to lock the effect
Ctrl-Z to get back my selected effect
I click twice to lock the effect in the menu
Ctrl-R to check 3-6 variation how it sounds
repeat
Or alternatively would you enable the effect loading (enable the selection in the context menu) even if the “lock effect” is set? It’s not realistic to fear that you right-click the module, move the mouse over an effect group, and left-click a new effect by mistake.
For what it’s worth, I would be okay with ‘Lock effect type’ going in any direction people like. I pretty much never randomize so it’s a moot point to me. Paul is likely to be interested in getting people’s opinions on what’s best, and is also interested in which things I can cope with and which things upset me. I don’t think there’s any possible way ‘Lock effect type’ could be, where I would mind. It’s a feature that could be in any condition because I won’t use it. So, let discussion ensue!
Yes so randomize without a lock, and then a fast undo-ish key or click to bring back “the effect” but keep the random parameters. Like a far search, but closer to a found. The lock is more like a found, and misses on an accidental far.
Yes, it’s the full undo has already been heard, but just an effect undo (or move forward based on last selected which can be double undone), has potential for finding due to randomization alot. So I’m saying I’d leave the lock off, and flick back to last often, although it would be a random “locked alike” version of last.
Let me contradict myself. Earlier I wrote that I’m not recommending the removal of “Lock Effect Type” but actually I want to get rid of it.
The more I think about it the more certain I am: it’s pointless. And once it’s in the context menu we can spend countless hours discussing how should it behave. (Ctrl-Z? Ctrl-I? Ctrl-R? Etc.)
But honestly, it serves no use at all. Let’s drop it.