After thinking about it, the auto-save toggle is meant for the *Editor* fields listed behind the Configure button. The auto-scroll toggle is not part of the Editor, and is more akin to a system setting, although it's placed in the main UI for convenience reasons related to its nature. As such, and especially considering it's a plugin, I lean towards decoupling auto-scroll from the auto-save settings, and just storing it independently.
There is an issue on the beta where if you use pipe ( | ) in the prompt to make a prompt matrix, the optional prompts are only applied when the last prompt in the matrix is used.
* Update README.md
* Update README.md
* Make on_sd_start.sh executable
* Merge pull request #542 from patriceac/patch-1
Fix restoration of model and VAE
* Merge pull request #541 from patriceac/patch-2
Fix restoration of parallel output setting
* Hypernetwork support
Adds support for hypernetworks. Hypernetworks are stored in /models/hypernetworks
* forgot to remove unused code
Co-authored-by: cmdr2 <secondary.cmdr2@gmail.com>
* New engine.js first draft.
* Small fixes...
* Bump version for cache...
* Improved cancellation code.
* Cleaning
* Wrong argument used in Task.waitUntil
* session_id needs to always match SD.sessionId
* Removed passing explicit Session ID from UI.
Use SD.sessionID to replace.
* Cleaning... Removed a disabled line and a hardcoded value.
* Fix return if tasks are still waiting.
* Added checkbox to reverse processing order.
* Fixed progress not displaying properly.
* Renamed reverse label.
* Only hide progress bar inside onCompleted.
* Thanks to rbertus2000 for helping testing and debugging!
* Resolve async promises when used optionally.
* when removed var should have used let, not const.
* Renamed getTaskErrorHandler to onTaskErrorHandler to better reflect actual implementation.
* Switched to the unsafer and less git friendly end of lines comma as requested in review.
* Raised SERVER_STATE_VALIDITY_DURATION to 90 seconds to match the changes to Beta.
* Added logging.
* Added one more hook before those inside the SD engine.
* Added selftest.plugin.js as part of core.
* Removed a tests that wasn't yet implemented...
* Groupped task stopping and abort in single function.
* Added optional test for plugins.
* Allow prompt text to be selected.
* Added comment.
* Improved isServerAvailable for better mobile usage and added comments for easier debugging.
* Comments...
* Normalized EVENT_STATUS_CHANGED to follow the same pattern as the other events.
* Disable plugins if editorModifierTagsList is not defined.
* Adds a new ServiceContainer to register IOC handlers.
* Added expect test for a missing dependency in a ServiceContainer
* Moved all event code in it's own sub class for easier reuse.
* Removed forgotten unused var...
* Allow getPrompts to be reused be plugins.
* Renamed EventSource to GenericEventSource to avoid redefining an existing class name.
* Added missing time argument to debounce
* Added output_quality to engine.js
* output_quality need to be an int.
* Fixed typo.
* Replaced the default euler_a by dpm2 to work with both SD1.# and SD2
* Remove generic completed tasks from plugins on generator complete.
* dpm2 starts at step 2, replaced with plms to start at step 1.
* Merge error
* Merge error
* changelog
Co-authored-by: Marc-Andre Ferland <madrang@gmail.com>
* fixed tools for image editor to be more modular and made cursor an actual cursor change
* fixed eraser cursor positioning
* updated opacity to not have a 100 option
* separated clear into an actions section
* added history support for image editor. ctrl-z and ctrl-y both work now
* removed extra console log debugging stuff
* updated buttons style
* updated the button ui on the main page as requested
* updated with a bunch of bugfixes
* safetensors support
Add support for checkpoints in safetensors format: https://github.com/huggingface/safetensors
This format shall be safer than pickle files
* pip install safetensors
* Fix upscaling when a source image is set
If you have an image selected (img2img) then clicking Upscale on another unrelated image, the image for img2img is used and you get something very unexpected.
* Fix for img2img and mask gens
With a pen, typing on a browser page, waiting a short moment, and then moving the pen scrolls the page.
Call event.preventDefault() to disable this default behaviour for events in the canvas area.
* started implementing hamunii's image editor, and added a hamunii theme
* fixed so active tab is main tab
* added some testing stuff for image ediotr
* re-implemented canvas drawing myself. just need to add layer stuff now
* moved everything to an image editor class and implement it so it actually works nicely now
* fixed a couple weird bugs and cleaned up the background image and sharpness stuff
* cleaned up a lot of stuff about the editor, added tools, buttons, made it mostly work in the current ui
* added inpainting support
* updated with more nice changes/updates to the inpainting and drawing editor
* made some more fixes and touchups to the image editor
* removed a bunch of semicolons
* remove old image inpainting system
* updated to work properly on mobile
* made a minor bugfix
* fixed img_size_box alignment
* Update index.html
Co-authored-by: cmdr2 <secondary.cmdr2@gmail.com>
Co-authored-by: cmdr2 <shashank.shekhar.global@gmail.com>
* Toggle image modifiers plugin
Right-click on image modifiers to temporarily turn them off without removing them. To quickly iterate and experiment with various combinations.
Please note this plugin required a minor tweak in getPrompts() to add support for image modifier inactive state.
* Fix tag matching
Co-authored-by: cmdr2 <secondary.cmdr2@gmail.com>
As per Discord conversation, this PR fixed the image modifiers behavior when a modifier appears more than once, and also fixes a regression introduced by ((weighted modifiers)).
Update the seed *before* starting the processing, so interrupting the processing retains the seed being used for the batch being currently processed.
The idea behind that is that if I like the gen I'm currently seeing and want to build on top of it, I can create a new task with the same seed without having to wait for the current task to complete.
When there are too many tasks and the top of the list is not visible, there is no visual feedback that a task has been successfully added to the queue.
Adding a subtle visual feedback on buttons upon click to reflect that the mouse event was taken into account.
