Scheduled research persona generation
This commit is contained in:
ajaysi
98
backend/models/oauth_token_monitoring_models.py
Normal file
98
backend/models/oauth_token_monitoring_models.py
Normal file
@@ -0,0 +1,98 @@
|
||||
"""
|
||||
OAuth Token Monitoring Models
|
||||
Database models for tracking OAuth token status and monitoring tasks.
|
||||
"""
|
||||
|
||||
from sqlalchemy import Column, Integer, String, Text, DateTime, Boolean, JSON, Index, ForeignKey
|
||||
from sqlalchemy.orm import relationship
|
||||
from datetime import datetime
|
||||
|
||||
# Import the same Base from enhanced_strategy_models
|
||||
from models.enhanced_strategy_models import Base
|
||||
|
||||
|
||||
class OAuthTokenMonitoringTask(Base):
|
||||
"""
|
||||
Model for storing OAuth token monitoring tasks.
|
||||
|
||||
Tracks per-user, per-platform token monitoring with weekly checks.
|
||||
"""
|
||||
__tablename__ = "oauth_token_monitoring_tasks"
|
||||
|
||||
id = Column(Integer, primary_key=True, index=True)
|
||||
|
||||
# User and Platform Identification
|
||||
user_id = Column(String(255), nullable=False, index=True) # Clerk user ID (string)
|
||||
platform = Column(String(50), nullable=False) # 'gsc', 'bing', 'wordpress', 'wix'
|
||||
|
||||
# Task Status
|
||||
status = Column(String(50), default='active') # 'active', 'failed', 'paused'
|
||||
|
||||
# Execution Tracking
|
||||
last_check = Column(DateTime, nullable=True)
|
||||
last_success = Column(DateTime, nullable=True)
|
||||
last_failure = Column(DateTime, nullable=True)
|
||||
failure_reason = Column(Text, nullable=True)
|
||||
|
||||
# Scheduling
|
||||
next_check = Column(DateTime, nullable=True, index=True) # Next scheduled check time
|
||||
|
||||
# Metadata
|
||||
created_at = Column(DateTime, default=datetime.utcnow)
|
||||
updated_at = Column(DateTime, default=datetime.utcnow, onupdate=datetime.utcnow)
|
||||
|
||||
# Execution Logs Relationship
|
||||
execution_logs = relationship(
|
||||
"OAuthTokenExecutionLog",
|
||||
back_populates="task",
|
||||
cascade="all, delete-orphan"
|
||||
)
|
||||
|
||||
# Indexes for efficient queries
|
||||
__table_args__ = (
|
||||
Index('idx_user_platform', 'user_id', 'platform'),
|
||||
Index('idx_next_check', 'next_check'),
|
||||
Index('idx_status', 'status'),
|
||||
)
|
||||
|
||||
def __repr__(self):
|
||||
return f"<OAuthTokenMonitoringTask(id={self.id}, user_id={self.user_id}, platform={self.platform}, status={self.status})>"
|
||||
|
||||
|
||||
class OAuthTokenExecutionLog(Base):
|
||||
"""
|
||||
Model for storing OAuth token monitoring execution logs.
|
||||
|
||||
Tracks individual execution attempts with results and error details.
|
||||
"""
|
||||
__tablename__ = "oauth_token_execution_logs"
|
||||
|
||||
id = Column(Integer, primary_key=True, index=True)
|
||||
|
||||
# Task Reference
|
||||
task_id = Column(Integer, ForeignKey("oauth_token_monitoring_tasks.id"), nullable=False, index=True)
|
||||
|
||||
# Execution Details
|
||||
execution_date = Column(DateTime, default=datetime.utcnow, nullable=False)
|
||||
status = Column(String(50), nullable=False) # 'success', 'failed', 'skipped'
|
||||
|
||||
# Results
|
||||
result_data = Column(JSON, nullable=True) # Token status, expiration info, etc.
|
||||
error_message = Column(Text, nullable=True)
|
||||
execution_time_ms = Column(Integer, nullable=True)
|
||||
|
||||
# Metadata
|
||||
created_at = Column(DateTime, default=datetime.utcnow)
|
||||
|
||||
# Relationship to task
|
||||
task = relationship("OAuthTokenMonitoringTask", back_populates="execution_logs")
|
||||
|
||||
# Indexes for efficient queries
|
||||
__table_args__ = (
|
||||
Index('idx_task_execution_date', 'task_id', 'execution_date'),
|
||||
Index('idx_status', 'status'),
|
||||
)
|
||||
|
||||
def __repr__(self):
|
||||
return f"<OAuthTokenExecutionLog(id={self.id}, task_id={self.task_id}, status={self.status}, execution_date={self.execution_date})>"
|
||||
|
||||
@@ -157,12 +157,14 @@ class PersonaData(Base):
|
||||
id = Column(Integer, primary_key=True, autoincrement=True)
|
||||
session_id = Column(Integer, ForeignKey('onboarding_sessions.id', ondelete='CASCADE'), nullable=False)
|
||||
|
||||
# Persona generation results
|
||||
# Persona generation results
|
||||
core_persona = Column(JSON) # Core persona data (demographics, psychographics, etc.)
|
||||
platform_personas = Column(JSON) # Platform-specific personas (LinkedIn, Twitter, etc.)
|
||||
quality_metrics = Column(JSON) # Quality assessment metrics
|
||||
selected_platforms = Column(JSON) # Array of selected platforms
|
||||
|
||||
research_persona = Column(JSON, nullable=True) # AI-generated research persona with personalized defaults
|
||||
research_persona_generated_at = Column(DateTime, nullable=True) # Timestamp for 7-day TTL cache validation
|
||||
|
||||
# Metadata
|
||||
created_at = Column(DateTime, default=func.now())
|
||||
updated_at = Column(DateTime, default=func.now(), onupdate=func.now())
|
||||
@@ -182,6 +184,8 @@ class PersonaData(Base):
|
||||
'platform_personas': self.platform_personas,
|
||||
'quality_metrics': self.quality_metrics,
|
||||
'selected_platforms': self.selected_platforms,
|
||||
'research_persona': self.research_persona,
|
||||
'research_persona_generated_at': self.research_persona_generated_at.isoformat() if self.research_persona_generated_at else None,
|
||||
'created_at': self.created_at.isoformat() if self.created_at else None,
|
||||
'updated_at': self.updated_at.isoformat() if self.updated_at else None
|
||||
}
|
||||
110
backend/models/research_persona_models.py
Normal file
110
backend/models/research_persona_models.py
Normal file
@@ -0,0 +1,110 @@
|
||||
"""
|
||||
Research Persona Models
|
||||
Pydantic models for AI-generated research personas.
|
||||
"""
|
||||
|
||||
from typing import Dict, Any, List, Optional
|
||||
from pydantic import BaseModel, Field
|
||||
from datetime import datetime
|
||||
|
||||
|
||||
class ResearchPreset(BaseModel):
|
||||
"""Research preset configuration."""
|
||||
name: str
|
||||
keywords: str
|
||||
industry: str
|
||||
target_audience: str
|
||||
research_mode: str = Field(..., description="basic, comprehensive, or targeted")
|
||||
config: Dict[str, Any] = Field(default_factory=dict, description="Complete ResearchConfig object")
|
||||
description: Optional[str] = None
|
||||
icon: Optional[str] = None
|
||||
gradient: Optional[str] = None
|
||||
|
||||
|
||||
class ResearchPersona(BaseModel):
|
||||
"""AI-generated research persona providing personalized defaults and suggestions."""
|
||||
|
||||
# Smart Defaults
|
||||
default_industry: str = Field(..., description="Default industry from onboarding data")
|
||||
default_target_audience: str = Field(..., description="Default target audience from onboarding data")
|
||||
default_research_mode: str = Field(..., description="basic, comprehensive, or targeted")
|
||||
default_provider: str = Field(..., description="google or exa")
|
||||
|
||||
# Keyword Intelligence
|
||||
suggested_keywords: List[str] = Field(default_factory=list, description="8-12 relevant keywords")
|
||||
keyword_expansion_patterns: Dict[str, List[str]] = Field(
|
||||
default_factory=dict,
|
||||
description="Mapping of keywords to expanded, industry-specific terms"
|
||||
)
|
||||
|
||||
# Domain & Source Intelligence
|
||||
suggested_exa_domains: List[str] = Field(
|
||||
default_factory=list,
|
||||
description="4-6 authoritative domains for the industry"
|
||||
)
|
||||
suggested_exa_category: Optional[str] = Field(
|
||||
None,
|
||||
description="Suggested Exa category based on industry"
|
||||
)
|
||||
|
||||
# Query Enhancement Intelligence
|
||||
research_angles: List[str] = Field(
|
||||
default_factory=list,
|
||||
description="5-8 alternative research angles/focuses"
|
||||
)
|
||||
query_enhancement_rules: Dict[str, str] = Field(
|
||||
default_factory=dict,
|
||||
description="Templates for improving vague user queries"
|
||||
)
|
||||
|
||||
# Research History Insights
|
||||
recommended_presets: List[ResearchPreset] = Field(
|
||||
default_factory=list,
|
||||
description="3-5 personalized research preset templates"
|
||||
)
|
||||
|
||||
# Research Preferences
|
||||
research_preferences: Dict[str, Any] = Field(
|
||||
default_factory=dict,
|
||||
description="Structured research preferences from onboarding"
|
||||
)
|
||||
|
||||
# Metadata
|
||||
generated_at: Optional[str] = Field(None, description="ISO timestamp of generation")
|
||||
confidence_score: Optional[float] = Field(None, ge=0.0, le=1.0, description="Confidence score 0-1")
|
||||
version: Optional[str] = Field(None, description="Schema version")
|
||||
|
||||
class Config:
|
||||
json_schema_extra = {
|
||||
"example": {
|
||||
"default_industry": "Healthcare",
|
||||
"default_target_audience": "Medical professionals and healthcare administrators",
|
||||
"default_research_mode": "comprehensive",
|
||||
"default_provider": "exa",
|
||||
"suggested_keywords": ["telemedicine", "patient care", "healthcare technology"],
|
||||
"keyword_expansion_patterns": {
|
||||
"AI": ["healthcare AI", "medical AI", "clinical AI"],
|
||||
"tools": ["medical devices", "clinical tools"]
|
||||
},
|
||||
"suggested_exa_domains": ["pubmed.gov", "nejm.org", "thelancet.com"],
|
||||
"suggested_exa_category": "research paper",
|
||||
"research_angles": [
|
||||
"Compare telemedicine platforms",
|
||||
"Telemedicine ROI analysis",
|
||||
"Latest telemedicine trends"
|
||||
],
|
||||
"query_enhancement_rules": {
|
||||
"vague_ai": "Research: AI applications in Healthcare for Medical professionals",
|
||||
"vague_tools": "Compare top Healthcare tools"
|
||||
},
|
||||
"recommended_presets": [],
|
||||
"research_preferences": {
|
||||
"research_depth": "comprehensive",
|
||||
"content_types": ["blog", "article"]
|
||||
},
|
||||
"generated_at": "2024-01-01T00:00:00Z",
|
||||
"confidence_score": 0.85,
|
||||
"version": "1.0"
|
||||
}
|
||||
}
|
||||
|
||||
48
backend/models/scheduler_models.py
Normal file
48
backend/models/scheduler_models.py
Normal file
@@ -0,0 +1,48 @@
|
||||
"""
|
||||
Scheduler Event Models
|
||||
Models for tracking scheduler-level events and history.
|
||||
"""
|
||||
|
||||
from sqlalchemy import Column, Integer, String, Text, DateTime, JSON, Float
|
||||
from datetime import datetime
|
||||
|
||||
# Import the same Base from enhanced_strategy_models
|
||||
from models.enhanced_strategy_models import Base
|
||||
|
||||
|
||||
class SchedulerEventLog(Base):
|
||||
"""Model for storing scheduler-level events (check cycles, interval adjustments, etc.)"""
|
||||
__tablename__ = "scheduler_event_logs"
|
||||
|
||||
id = Column(Integer, primary_key=True, index=True)
|
||||
event_type = Column(String(50), nullable=False) # 'check_cycle', 'interval_adjustment', 'start', 'stop', 'job_scheduled', 'job_cancelled'
|
||||
event_date = Column(DateTime, default=datetime.utcnow, nullable=False, index=True)
|
||||
|
||||
# Event details
|
||||
check_cycle_number = Column(Integer, nullable=True) # For check_cycle events
|
||||
check_interval_minutes = Column(Integer, nullable=True) # Interval at time of event
|
||||
previous_interval_minutes = Column(Integer, nullable=True) # For interval_adjustment events
|
||||
new_interval_minutes = Column(Integer, nullable=True) # For interval_adjustment events
|
||||
|
||||
# Task execution summary for check cycles
|
||||
tasks_found = Column(Integer, nullable=True)
|
||||
tasks_executed = Column(Integer, nullable=True)
|
||||
tasks_failed = Column(Integer, nullable=True)
|
||||
tasks_by_type = Column(JSON, nullable=True) # {'monitoring_task': 5, ...}
|
||||
|
||||
# Job information
|
||||
job_id = Column(String(200), nullable=True) # For job_scheduled/cancelled events
|
||||
job_type = Column(String(50), nullable=True) # 'recurring', 'one_time'
|
||||
user_id = Column(String(200), nullable=True, index=True) # For user isolation
|
||||
|
||||
# Performance metrics
|
||||
check_duration_seconds = Column(Float, nullable=True) # How long the check cycle took
|
||||
active_strategies_count = Column(Integer, nullable=True)
|
||||
active_executions = Column(Integer, nullable=True)
|
||||
|
||||
# Additional context
|
||||
event_data = Column(JSON, nullable=True) # Additional event-specific data
|
||||
error_message = Column(Text, nullable=True) # For error events
|
||||
|
||||
created_at = Column(DateTime, default=datetime.utcnow)
|
||||
|
||||
Reference in New Issue
Block a user