[Feature Request]: Allow order of operations on prompt parser
Is there an existing issue for this?
- [X] I have searched the existing issues and checked the recent builds/commits
What would your feature do ?
Imagine the next prompt:
a cat AND a dog BREAK on a beach
Here this prompt can be parsed on 2 different ways:
1
a cat ━━━┓
┣━━┓
a dog ━━━┛ ┃
on a beach ━┻━━ OUTPUT
2
a cat ━━━━━━━━━━━┳━━ Output
a dog ━━━━━━━━┓ ┃
┣━━┛
on a beach ━━━┛
Both ways ends up with a different result. (drawing this ended up being more difficult than it looks)
My proposal is to do something similar to how math works, by a parenthesis-equivalent keywords, something like BEGIN/END so by doing that, now:
BEGIN a cat AND a dog END BREAK on a beach
and
a cat AND BEGIN a dog BREAK on a beach END
This is also relevant since other extensions create their own keywords and usually collide since there is no way to express order of operations. Having a native to the tool way of doing it might force the extension developers to respect them as well.
Proposed workflow
No change on workflow, just a couple of keywords.
Additional information
No response
[from:to:when] prompt syntax could be used for it.
for example [a cat:a dog:0.5] on a beach
see also https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Features#prompt-editing
[from:to:when]prompt syntax could be used for it.for example
[a cat:a dog:0.5] on a beachsee also https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Features#prompt-editing
That could be a good hack for some cases, but that's not the same.