I've built another mini generator inside of my experiment page, The View Counter Experiment, inspired off this generator that generates a color based on a random object using the power of AI for the day. I'm glad on how it turned out to be.
It's located in the AI Experiments tab on the right if you want to try that out.
It's text2image since the framework was originally used to create customizable text-to-image generators I think.
Thanks! I appreciate it! ๐
I'm going to do a few more updates and release a few more things before heading to the "revival" plan and eventually the big Generator Manager update.
There isn't much changed from the outlooks, but if you look closely there is only one "Start Learning" button below the example code section, there is no video tutorial section, there is no instruction on how to embed a generator (which I guess is still not a thing at that time) and there is even no "try this example" button that redirects to this example that was also linked in the old page somewhere nearby.
For me though, Perchance is really just a foundation for me to actually create things I like and engage with other creations since I discovered it back in 2022 when it's just plain simple (there was no hub, very simple welcome page, fewer plugins, simpler comments plugin UI, etc.) and there wasn't any AI image stuff or t2i generators yet. And that's the reason why the Generator Manager (my generator hub page) is here. I started to like it so much that I published my first own creation even published here on this platform. Love to see how it evolves, though, and glad this was explained in this Lemmy community!
That's a great answer! I've never thought of these solutions in a while, maybe even add a warning for that so other newbies could know easily.
You can use the shorthand exp ? if_true : if_false notation chained together instead of the if ... else notations. It's specially used for conditional inline value returns.
[s == "Wolf" ? wolf_gene : s == "Cat" ? cat_gene : s == "Fox" ? fox_gene : ""]
You can also do like this as well:
[({"Wolf": wolf_gene, "Cat": cat_gene, "Fox": fox_gene})[s]]
This is likely an error with the text-to-image-plugin interfering with some t2i-framework stuff or similar since it's trying to create the gallery iframe inside the generator, as I was getting the exact same error from the text2image-generator. This error normally doesn't happen, because the plugin is always set to access that variable (which points to a span element) under a timeout (which is right after the element was created). I don't know exactly what causes this, so I'm going to ping @[email protected] so they can address this issue in the meantime.
That error description on the second screenshot is something that I haven't caught before, I've just got one already, but a bit different. I initially thought it was a custom error from the generator, but turned out it was a new type of error from the Perchance engine.
Quick bug report, when I was playing with the testing panel, I tried to get to the very bottom of the long word list output by pressing Shift+Down on the output container, which pushes the entire container down that causes the input section to be cropped and really hard to get back in. The only ways to temporarily fix this issue is to press Shift+Tab while selecting the green output text thing, or just select the input box manually and type something on it.
Here's a short video demonstration on how I was able to catch the bug:
No problem, I appreciate that!
Another time to say, "long time not posting on this community", but here we go!
There is one question I've wanting to ask about the recent update about the dark mode on the main/official Perchance pages: is there a way to disable them only for the pages (i.e. keep the pages in light mode without having to turn off the dark mode completely)? Most of the time my eyes just don't get comfortable enough reading the text in the dark mode.
I'd also suggest putting a button to toggle the dark mode on/off just like we have on the Hub on these pages.
So in https://perchance.org/ai-text-to-image-generator (and other t2i-based image generators) you could choose an art style and then a few dropdowns come below the prompt description textbox allowing you to choose any additional effects (or built-in modifiers, I should say) to be added to the prompt.
Well, I've found a bug on these. When I choose one that has a dropdown menu (for example, "Professional Photo"), the dropdowns did come out, but when I reload the page, the option I chose earlier retained, however the dropdowns did not come out as if the default option was selected.
I also tested if this bug happened on other t2i-based AI image generators too, such as https://perchance.org/furry-ai and https://perchance.org/ai-image-lab, and it also did the same, so this is most likely a bug within the t2i framework.
Haven't been posting in this community for a while, but I'm here just to share a tip (I think there hasn't been anybody asking for this here yet) that someone on Discord has suggested on. This only works on computers with developer tools though, and I think you could do this in Firefox too.
Here's how to do it:
localStorage.previouslyUploadedFiles. It'll then give you a JSON of the upload list that you can copy the given string provided to a text file and transfer it to another computer you'd want to access the uploads for.
Right click and select "Copy string contents" ("Copy Message" in Firefox). Then, paste it into a text file somewhere that you can transfer through a media like a USB to be transferred into the other computer.
In another computer, repeat the first two steps above, and then write this:
localStorage.previouslyUploadedFiles = [upload list content from the text file]
[upload list content from the text file] part with the contents from the text file you've just transferred, and then hit Enter. This will magically set the upload list to the one you have in your primary computer.And that's it! The upload list isn't synced though, but at least could be useful for keeping stuff in. Feel free to add something missing here if you like! ๐
So I've just been working for the huge update for the Generator Manager since a couple days and I wanted to share one thing I've been working on in the middle of the making, that is the new error handling method on the generator hub page. Instead of just throwing the error onto the DevTools Console, it also throws the error into the Perchance error box inside the generator hub page, so it is much more convenient to see what's wrong with that feature in case I'm changing on something. And that is a custom Perchance error I just created.
This is accomplised using __perchanceError(), and I understand this is likely a "private" function or something that isn't guaranteed to be "backwards compatible" for long (as I understand that), but I wonder if this function was available for wide use and documented somewhere in the advanced tutorial, it'd be very good for someone to learn about. You can also put HTML into the custom error text to add some neat formatting to it.
This will be limited to the handling of generator cards and the recently updated section for now but in future updates these will be expanded to pretty much every single feature on the generator hub page.
Time for another trailer about the big Generator Manager 2.0.20.1 update! I have just started planning and getting to work on building the changes for the big update very soon, from the latest Preview update, and I hope that this goes well enough so that I could finish it in basically a few days from now. But there are three more off-Preview changes I want to mention here: new backgrounds and accent colors (of course), new sidebar items, and finally, the introduction of dark mode for the very first time ever!
And there's just so many more coming out for the huge update...
Mention the server like this: @perchance @lemmy.world (remove the spaces in-between so it would be an actual ping).