cross-posted from: https://lemmy.world/post/19286382

There has been a couple of Community Events in Casual Perchance since its implementation via Lemmy World posts.

What kind of events would you like to see or be part of?

Some of the old events were:

• Generator Jam - like a Game Jam but with generators based on a theme
  • Plain - create a generator that fits the specified theme
  • Useful Generators - generators that can be imported/useful for other generators (i.e. importable lists or plugins etc.)
  • Mad Science - not generators but anything made with Perchance Syntax.
• Contribute to a Generator - didn't get much traction, like a Generator Jam, but you allow others to give feedback/suggestions throughout the event to the generator you have submitted. (Might change this to just a Feedback/Suggestion Jam, submit your generators for feedback and suggestions and don't require to create a new generator)
• Image Challenge - Mostly AI Images generated with Perchance text-to-image-plugin based on a theme.

Also, feel free to give any feedback or suggestions about the Community Events. Some more specific feedback/suggestion categories are:

• Length of Events
• Type/Content of Events
• Engagement or Retainability of Participants

Note that you can host your own Community Events, see this post for an explanation.

Update 4: Fixed
Update 3: It seems that changing the 'src' of the iframe of the embedded seems to apply the changes. Clicking on the 'Go to Event' buttons on the home tab seems to revert the page to its previous versions. A video recording of the problem.
Update 2: Can confirm that the changes are reflected after a couple of minutes, for now.
Update: The embedded page has now reflected the changes. Will monitor if any other changes are immediately (or at least after a little bit of time) reflected.

I'm posting again to re-open a previous bug.

Previous Post: https://lemmy.world/post/15639946

Currently, the /hub where the /hub-events is embedded to, doesn't have the updated /hub-events page.

I've updated the code in the /hub-events last July 25. But currently, the change to it isn't reflected on the embedded /hub-events on the /hub at the Events tab.

Last time, it updated itself after a few minutes, but now two days have passed it hasn't updated yet.

There seems to be a overflow issue on the comments-plugin's settings and fullscreen buttons.

Not sure when this problem happened, but if you take a look at https://perchance.org/a-testing-gen which has an embedded Perchance page to https://perchance.org/b-testing-gen, the changes on the embedded generator /b-testing-gen, isn't applied on the iframe on /a-testing-gen.

I've tried disabling the cache on DevTools, reloaded both gens (and saved them again) and the changes on /b-testing-gen isn't applied on the /a-testing-gen.

Improved Make Table Plugin

This is a modification of the make-table-plugin with added features:

• Row Heights (option to change heights of each row)
• Row Alignments (option to change the alignment of the text in the row)
• Header Rows (makes the first row of the table bold, and an option to stylize it)
• Footer Rows (last row bold, and option to stylize it)
• Column Header (bold first column of the table, and option to stylize it)
• Caption (adds a Caption to the table, with option to place it at the top, and at the bottom, and to stylize it).

Here is an example of a table that can be made from it:

I've also created a Helper to help create tables with the plugin at Improved Make Table Plugin Helper.

Suppose we have the following list:

createInstance = {import:create-instance-plugin}

person
  name = [this.nameList]
  nameList
    Salman
    Manny
    Rhian
  age = {31-49}
  child = [this.child_obj]
  child_obj
    name = [this.nameList]
    nameList
      Anne
      Arram
      Amelia
    age = {3-17}

If we output:

[p = createInstance(person, "deep"), p.name] [p.age] [p.name] [p.age]

We can see that it has fixed the name and the age property on the instance. However, if we output:

[p = createInstance(person, "deep"), p.child.name] [p.child.age] [p.child.name] [p.child.age]

It would throw undefined. Now looking at similar list:

person
  name = {Salman|Manny|Rhian}
  age = {31-49}
  child = [this.child_obj]
  child_obj
    name = {Anne|Arram|Amelia}
    age = {3-17}

output
  [p = createInstance(person, "deep"), p.child.name] [p.child.age] [p.child.name] [p.child.age]

This would have the values fixed and working.

Looking at the code of the create-instance-plugin, it would only allow the 'deep' fixing of properties if there isn't any items/lists in it. What that means is the following:

child_obj
  name = [this.nameList]
  nameList
    Anne
    Arram
    Amelia
  age = {3-17}

This object/list, has a list ( nameList) within it as well as properties (name and age), while:

child_obj
  name = {Anne|Arram|Amelia}
  age = {3-17}

Only has properties. and based on Line 22 of the create-instance-plugin:

...
} else if(propValue.getPropertyKeys && propValue.getPropertyKeys.length > 0 && propValue.getLength === 0) {
...

If the list to be fixed has a list within it other than the properties i.e. propValue.getLength is not zero, then it wouldn't fix the properties within it.

The first child_obj has a nameList with it, and upon calling propValue.getLength it would have 1 while the second child_obj will return a propValue.getLength of 0 since it doesn't have any lists.

Thus, the solution for the problem is just removing the propValue.getLength === 0 check OR create another check without it.

TLDR; propValue.getLength === 0 at Line 22 of the create-instance-plugin can be removed to allow properties that rely on same level lists to be fixed. Here is the demo of the problem with a 'remixed' create instance plugin with the fix.

I've been made aware that the code I've written to access text-to-speech from ElevenLabs API is no longer working.

I've tested it and it seems that the CORS-Proxy that is being used in ai-character-chat currently doens't allow POST methods (which is being used to 'post' the text to be 'spoken' in ElevenLabs).

Not a major/priority issue but might be nice to be fixed. I also wonder how many are using text to speech (even just using the Speech Synthesis code) in the ai-character-chat....

Found out about this possible improvement to the URL Params Plugin in which you currently cannot pass through URL params in the null.perchance domain in an <iframe> i.e. it doesn't parse the passed through parameters.

I was able to pass through parameters in an iframe in the normal perchance domain, but without the null part, it has the navigation bar and seems to mess up the CSS. I'm currently applying it to 'navigate' from the 'Home' tab of the hub to the 'Events' tab, then select the appropriate events page using the URL params.

cross-posted from: https://lemmy.world/post/14744285

I've mentioned this on another post, but I'll create a new thread to hear other opinions.

I'm proposing to have the Community Events on Perchance Hub be hosted on this Lemmy Community.

To accomplish this, the Forms tab on the Perchance Hub would be changed to Events tab. It would host the Event posts from this community.

That is, anyone can post an Event thread here and it would be mirrored on the Perchance Hub.

Technical Notes: The posts here would be retrieved (via the Lemmy API) and the posts would be filtered to events only. The Event posts will then be parsed and neatly displayed on the Perchance Hub. This would mean that the posts would have a template to follow.

This is so that anyone is welcome to create Events and those Event makers would be the ones who will manage their own events (no longer needing to wait the Perchance Hub update).

This way, there could be different events running, managed by different people, with different topics, event types, and limitations/rules and to update the details on the Perchance Hub, the event creators can just edit the event posts and it would reflect on the Hub.

Not yet sure about what details would be on the Perchance Hub. But something like @BluePower's Upcoming Events Page might be a good format/structure to remix with (although the comments plugin on each event on Perchance Hub might not be used since the Event Post already have a comments/thread to talk with).

Technical Notes: The mirrored details can probably include a cover image, ability for the creators to change the colors on the Hub, specify if it is an 'image' event, 'generator/page' based event, etc., so the formatting on the Hub would be nice.

In Summary:

• Community Events will be hosted in https://lemmy.world/c/casual_perchance
• Events are now hosted/managed by Event Creators
  • The posts would have a (somewhat strict) template to nicely structure them on the mirror on the Perchance Hub.
  • Events can simultaneously ran by different people, with different topics, event types, limitations or rules.
  • Events can be updated by the Event Creator
• Event Details and Content / Template Format is not yet decided
• Format in the Perchance Hub is not yet designed.

Additional Notes:

• With this, I might remove the 'announcement modal' on the hub upon first visit (or when there is update to it) to remove the clutter.
• Might also implement a dark mode to the Hub. (will require quite an overhaul on the hub page)
• Might dynamically add the Event's banner on the Home Page for each event (with the most recent and about to end event on the top).
• I currently do not have a stable internet so these changes might take a while to implement xD

@[email protected] - pinging dev :)

Here is my implementation of 'remembering' the resized size of the textarea inputs. It is using the ResizeObserver (documentation) and will only observe textarea elements with resize style property enabled.

It is enabled with the @inputs:

[remember(root, "@inputs")]

EDIT: It would only save the resized size if you have inputs on the textarea.

EDIT 2: Fixed width responsiveness (for elements with width: 100%; max-width: ... ; resize: vertical where if the width gets lower than its maxed, it would save the width and fix the size without option to resize it back (since it is only resize: vertical). It now only saves the resizable direction i.e. height if resize is vertical and vice versa. For both or normal textarea it would save both width and height.

Here is the modified plugin: https://perchance.org/t2w8fixecn

Here is an example of it: https://perchance.org/5tfbmi0gtc

Due to the number of similar questions about the AI tools, here is a compiled list of frequently asked questions about the AI tools in Perchance.

Perchance AI FAQ

Please read it carefully and if you have any questions that isn't mentioned there, you can comment here.

On the ai-text-plugin, there is a way to get the inputs that were sent to the AI using the onStart(data) function:

onStart(data) - the code you put in this will run at the start of the generation process. You can access the inputs being used with data.inputs.instruction, data.inputs.startWith, etc.

I am wondering if we could also get the same for the text-to-image-plugin?

It can be either on the onStart(data) or on the onFinish(data) in which we can get the inputs/static values that were used for the response of the AI e.g. data.inputs.prompt, data.inputs.negativePrompt, etc.

With this, some features that can be implemented is a local history of past generations, in which the previous inputs were temporarily stored in the storage (or on the local storage so even if the page reloaded, the history is saved), then the users can revisit the previous generations (I have set up something similar like this on my text to image generators but they have some hacks to work).

This is not really a priority but might be good to have. Thanks!