Mirrors/talemate
1
0
Fork 0
mirror of https://github.com/vegu-ai/talemate.git synced 2025-12-16 11:47:48 +01:00
Code Issues Packages Projects Releases Wiki Activity

Add new properties to VisualPrompt for handling positive and negative prompts, including keywords and descriptive formats, and refactor prompt building logic.

Browse Source
This commit is contained in:
vegu-ai-tools
2025-11-22 17:28:42 +02:00
parent 16f135c8f0
commit dc3ba4a18a
2 changed files with 33 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
src/talemate/agents/visual/nodes.py
View File

@@ -287,6 +287,10 @@ class UnpackPrompt(Node):
self.add_output("instructions", socket_type="str")
self.add_output("positive_prompt", socket_type="str")
self.add_output("negative_prompt", socket_type="str")
self.add_output("positive_prompt_keywords", socket_type="str")
self.add_output("negative_prompt_keywords", socket_type="str")
self.add_output("positive_prompt_descriptive", socket_type="str")
self.add_output("negative_prompt_descriptive", socket_type="str")
async def run(self, state: GraphState):
prompt = self.normalized_input_value("prompt")
@@ -299,6 +303,10 @@ class UnpackPrompt(Node):
"parts": parts,
"positive_prompt": prompt.positive_prompt,
"negative_prompt": prompt.negative_prompt,
"positive_prompt_keywords": prompt.positive_prompt_keywords,
"negative_prompt_keywords": prompt.negative_prompt_keywords,
"positive_prompt_descriptive": prompt.positive_prompt_descriptive,
"negative_prompt_descriptive": prompt.negative_prompt_descriptive,
}
)

42
src/talemate/agents/visual/schema.py
View File

@@ -133,34 +133,42 @@ class VisualPrompt(pydantic.BaseModel):
@pydantic.computed_field
@property
def positive_prompt(self) -> str:
prompt: list[str] = []
if self.prompt_type == PROMPT_TYPE.KEYWORDS:
for part in self.parts:
prompt.extend(part.positive_keywords)
return ", ".join(dict.fromkeys(prompt))
elif self.prompt_type == PROMPT_TYPE.DESCRIPTIVE:
for part in self.parts:
if part.positive_descriptive:
prompt.append(part.positive_descriptive)
return "\n\n".join(prompt)
return ""
return self._build_prompt(self.prompt_type, True)
@pydantic.computed_field
@property
def negative_prompt(self) -> str:
return self._build_prompt(self.prompt_type, False)
@property
def positive_prompt_keywords(self) -> str:
return self._build_prompt(PROMPT_TYPE.KEYWORDS, True)
@property
def negative_prompt_keywords(self) -> str:
return self._build_prompt(PROMPT_TYPE.KEYWORDS, False)
@property
def positive_prompt_descriptive(self) -> str:
return self._build_prompt(PROMPT_TYPE.DESCRIPTIVE, True)
@property
def negative_prompt_descriptive(self) -> str:
return self._build_prompt(PROMPT_TYPE.DESCRIPTIVE, False)
def _build_prompt(self, prompt_type: PROMPT_TYPE, positive: bool) -> str:
prompt: list[str] = []
if self.prompt_type == PROMPT_TYPE.KEYWORDS:
if prompt_type == PROMPT_TYPE.KEYWORDS:
for part in self.parts:
prompt.extend(part.negative_keywords)
prompt.extend(part.positive_keywords if positive else part.negative_keywords)
return ", ".join(dict.fromkeys(prompt))
elif self.prompt_type == PROMPT_TYPE.DESCRIPTIVE:
elif prompt_type == PROMPT_TYPE.DESCRIPTIVE:
for part in self.parts:
if part.negative_descriptive:
prompt.append(part.negative_descriptive)
if part.positive_descriptive if positive else part.negative_descriptive:
prompt.append(part.positive_descriptive if positive else part.negative_descriptive)
return "\n\n".join(prompt)
return ""
class BackendStatus(pydantic.BaseModel):
type: BackendStatusType
message: str | None = None