Initial commit: breakpilot-core - Shared Infrastructure
Docker Compose with 24+ services: - PostgreSQL (PostGIS), Valkey, MinIO, Qdrant - Vault (PKI/TLS), Nginx (Reverse Proxy) - Backend Core API, Consent Service, Billing Service - RAG Service, Embedding Service - Gitea, Woodpecker CI/CD - Night Scheduler, Health Aggregator - Jitsi (Web/XMPP/JVB/Jicofo), Mailpit Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
Benjamin Boenisch
12
docs-src/backend/alerts_agent/models/__init__.py
Normal file
12
docs-src/backend/alerts_agent/models/__init__.py
Normal file
@@ -0,0 +1,12 @@
|
||||
"""Alert Agent Models."""
|
||||
|
||||
from .alert_item import AlertItem, AlertSource, AlertStatus
|
||||
from .relevance_profile import RelevanceProfile, PriorityItem
|
||||
|
||||
__all__ = [
|
||||
"AlertItem",
|
||||
"AlertSource",
|
||||
"AlertStatus",
|
||||
"RelevanceProfile",
|
||||
"PriorityItem",
|
||||
]
|
||||
174
docs-src/backend/alerts_agent/models/alert_item.py
Normal file
174
docs-src/backend/alerts_agent/models/alert_item.py
Normal file
@@ -0,0 +1,174 @@
|
||||
"""
|
||||
AlertItem Model.
|
||||
|
||||
Repräsentiert einen einzelnen Alert aus Google Alerts (RSS oder Email).
|
||||
"""
|
||||
|
||||
from dataclasses import dataclass, field
|
||||
from datetime import datetime
|
||||
from enum import Enum
|
||||
from typing import Optional
|
||||
import hashlib
|
||||
import uuid
|
||||
|
||||
|
||||
class AlertSource(str, Enum):
|
||||
"""Quelle des Alerts."""
|
||||
GOOGLE_ALERTS_RSS = "google_alerts_rss"
|
||||
GOOGLE_ALERTS_EMAIL = "google_alerts_email"
|
||||
MANUAL = "manual"
|
||||
|
||||
|
||||
class AlertStatus(str, Enum):
|
||||
"""Verarbeitungsstatus des Alerts."""
|
||||
NEW = "new"
|
||||
PROCESSED = "processed"
|
||||
DUPLICATE = "duplicate"
|
||||
SCORED = "scored"
|
||||
REVIEWED = "reviewed"
|
||||
ARCHIVED = "archived"
|
||||
|
||||
|
||||
@dataclass
|
||||
class AlertItem:
|
||||
"""Ein einzelner Alert-Eintrag."""
|
||||
|
||||
# Identifikation
|
||||
id: str = field(default_factory=lambda: str(uuid.uuid4()))
|
||||
|
||||
# Quelle
|
||||
source: AlertSource = AlertSource.GOOGLE_ALERTS_RSS
|
||||
topic_label: str = "" # z.B. "Schulrecht Bayern"
|
||||
feed_url: Optional[str] = None
|
||||
|
||||
# Content
|
||||
title: str = ""
|
||||
url: str = ""
|
||||
snippet: str = ""
|
||||
article_text: Optional[str] = None
|
||||
|
||||
# Metadaten
|
||||
lang: str = "de"
|
||||
published_at: Optional[datetime] = None
|
||||
fetched_at: datetime = field(default_factory=datetime.utcnow)
|
||||
|
||||
# Deduplication
|
||||
canonical_url: Optional[str] = None
|
||||
url_hash: Optional[str] = None
|
||||
content_hash: Optional[str] = None # SimHash für fuzzy matching
|
||||
|
||||
# Verarbeitung
|
||||
status: AlertStatus = AlertStatus.NEW
|
||||
cluster_id: Optional[str] = None
|
||||
|
||||
# Relevanz (nach Scoring)
|
||||
relevance_score: Optional[float] = None # 0.0 - 1.0
|
||||
relevance_decision: Optional[str] = None # KEEP, DROP, REVIEW
|
||||
relevance_reasons: list = field(default_factory=list)
|
||||
relevance_summary: Optional[str] = None
|
||||
|
||||
def __post_init__(self):
|
||||
"""Berechne Hashes nach Initialisierung."""
|
||||
if not self.url_hash and self.url:
|
||||
self.url_hash = self._compute_url_hash()
|
||||
if not self.canonical_url and self.url:
|
||||
self.canonical_url = self._normalize_url(self.url)
|
||||
|
||||
def _compute_url_hash(self) -> str:
|
||||
"""Berechne SHA256 Hash der URL."""
|
||||
normalized = self._normalize_url(self.url)
|
||||
return hashlib.sha256(normalized.encode()).hexdigest()[:16]
|
||||
|
||||
def _normalize_url(self, url: str) -> str:
|
||||
"""Normalisiere URL für Deduplizierung."""
|
||||
# Entferne Tracking-Parameter
|
||||
import urllib.parse
|
||||
parsed = urllib.parse.urlparse(url)
|
||||
|
||||
# Google News Redirect auflösen
|
||||
if "news.google.com" in parsed.netloc and "/articles/" in parsed.path:
|
||||
# news.google.com URLs enthalten die echte URL base64-kodiert
|
||||
# Hier nur Basic-Handling - echte Auflösung komplexer
|
||||
pass
|
||||
|
||||
# Tracking-Parameter entfernen
|
||||
tracking_params = {
|
||||
"utm_source", "utm_medium", "utm_campaign", "utm_content", "utm_term",
|
||||
"fbclid", "gclid", "ref", "source"
|
||||
}
|
||||
|
||||
query_params = urllib.parse.parse_qs(parsed.query)
|
||||
cleaned_params = {k: v for k, v in query_params.items()
|
||||
if k.lower() not in tracking_params}
|
||||
|
||||
cleaned_query = urllib.parse.urlencode(cleaned_params, doseq=True)
|
||||
|
||||
# Rekonstruiere URL ohne Fragment
|
||||
normalized = urllib.parse.urlunparse((
|
||||
parsed.scheme,
|
||||
parsed.netloc.lower(),
|
||||
parsed.path.rstrip("/"),
|
||||
parsed.params,
|
||||
cleaned_query,
|
||||
"" # No fragment
|
||||
))
|
||||
|
||||
return normalized
|
||||
|
||||
def compute_content_hash(self, text: Optional[str] = None) -> str:
|
||||
"""
|
||||
Berechne SimHash des Inhalts für Fuzzy-Matching.
|
||||
|
||||
SimHash erlaubt es, ähnliche Texte zu erkennen, auch wenn sie
|
||||
sich leicht unterscheiden (z.B. verschiedene Quellen zum selben Thema).
|
||||
"""
|
||||
from ..processing.dedup import compute_simhash
|
||||
|
||||
content = text or self.article_text or self.snippet or self.title
|
||||
if content:
|
||||
self.content_hash = compute_simhash(content)
|
||||
return self.content_hash or ""
|
||||
|
||||
def to_dict(self) -> dict:
|
||||
"""Konvertiere zu Dictionary für JSON/DB."""
|
||||
return {
|
||||
"id": self.id,
|
||||
"source": self.source.value,
|
||||
"topic_label": self.topic_label,
|
||||
"feed_url": self.feed_url,
|
||||
"title": self.title,
|
||||
"url": self.url,
|
||||
"snippet": self.snippet,
|
||||
"article_text": self.article_text,
|
||||
"lang": self.lang,
|
||||
"published_at": self.published_at.isoformat() if self.published_at else None,
|
||||
"fetched_at": self.fetched_at.isoformat() if self.fetched_at else None,
|
||||
"canonical_url": self.canonical_url,
|
||||
"url_hash": self.url_hash,
|
||||
"content_hash": self.content_hash,
|
||||
"status": self.status.value,
|
||||
"cluster_id": self.cluster_id,
|
||||
"relevance_score": self.relevance_score,
|
||||
"relevance_decision": self.relevance_decision,
|
||||
"relevance_reasons": self.relevance_reasons,
|
||||
"relevance_summary": self.relevance_summary,
|
||||
}
|
||||
|
||||
@classmethod
|
||||
def from_dict(cls, data: dict) -> "AlertItem":
|
||||
"""Erstelle AlertItem aus Dictionary."""
|
||||
# Parse Enums
|
||||
if "source" in data and isinstance(data["source"], str):
|
||||
data["source"] = AlertSource(data["source"])
|
||||
if "status" in data and isinstance(data["status"], str):
|
||||
data["status"] = AlertStatus(data["status"])
|
||||
|
||||
# Parse Timestamps
|
||||
for field_name in ["published_at", "fetched_at"]:
|
||||
if field_name in data and isinstance(data[field_name], str):
|
||||
data[field_name] = datetime.fromisoformat(data[field_name])
|
||||
|
||||
return cls(**data)
|
||||
|
||||
def __repr__(self) -> str:
|
||||
return f"AlertItem(id={self.id[:8]}, title='{self.title[:50]}...', status={self.status.value})"
|
||||
288
docs-src/backend/alerts_agent/models/relevance_profile.py
Normal file
288
docs-src/backend/alerts_agent/models/relevance_profile.py
Normal file
@@ -0,0 +1,288 @@
|
||||
"""
|
||||
RelevanceProfile Model.
|
||||
|
||||
Definiert das Relevanzprofil eines Nutzers für die Alerts-Filterung.
|
||||
Lernt über Zeit durch Feedback.
|
||||
"""
|
||||
|
||||
from dataclasses import dataclass, field
|
||||
from datetime import datetime
|
||||
from typing import Optional
|
||||
import uuid
|
||||
|
||||
|
||||
@dataclass
|
||||
class PriorityItem:
|
||||
"""Ein Prioritäts-Thema im Profil."""
|
||||
label: str # z.B. "Inklusion", "Datenschutz Schule"
|
||||
weight: float = 0.5 # 0.0 - 1.0, höher = wichtiger
|
||||
keywords: list = field(default_factory=list) # Zusätzliche Keywords
|
||||
description: Optional[str] = None # Kontext für LLM
|
||||
|
||||
def to_dict(self) -> dict:
|
||||
return {
|
||||
"label": self.label,
|
||||
"weight": self.weight,
|
||||
"keywords": self.keywords,
|
||||
"description": self.description,
|
||||
}
|
||||
|
||||
@classmethod
|
||||
def from_dict(cls, data: dict) -> "PriorityItem":
|
||||
return cls(**data)
|
||||
|
||||
|
||||
@dataclass
|
||||
class RelevanceProfile:
|
||||
"""
|
||||
Nutzerprofil für Relevanz-Scoring.
|
||||
|
||||
Das Profil wird verwendet, um Alerts auf Relevanz zu prüfen.
|
||||
Es enthält:
|
||||
- Prioritäten: Themen die wichtig sind (mit Gewichtung)
|
||||
- Ausschlüsse: Themen die ignoriert werden sollen
|
||||
- Positive Beispiele: URLs/Titel die relevant waren
|
||||
- Negative Beispiele: URLs/Titel die irrelevant waren
|
||||
- Policies: Zusätzliche Regeln (z.B. nur deutsche Quellen)
|
||||
"""
|
||||
|
||||
# Identifikation
|
||||
id: str = field(default_factory=lambda: str(uuid.uuid4()))
|
||||
user_id: Optional[str] = None # Falls benutzerspezifisch
|
||||
|
||||
# Relevanz-Kriterien
|
||||
priorities: list = field(default_factory=list) # List[PriorityItem]
|
||||
exclusions: list = field(default_factory=list) # Keywords zum Ausschließen
|
||||
|
||||
# Beispiele für Few-Shot Learning
|
||||
positive_examples: list = field(default_factory=list) # Relevante Alerts
|
||||
negative_examples: list = field(default_factory=list) # Irrelevante Alerts
|
||||
|
||||
# Policies
|
||||
policies: dict = field(default_factory=dict)
|
||||
|
||||
# Metadaten
|
||||
created_at: datetime = field(default_factory=datetime.utcnow)
|
||||
updated_at: datetime = field(default_factory=datetime.utcnow)
|
||||
|
||||
# Statistiken
|
||||
total_scored: int = 0
|
||||
total_kept: int = 0
|
||||
total_dropped: int = 0
|
||||
accuracy_estimate: Optional[float] = None # Geschätzte Genauigkeit
|
||||
|
||||
def add_priority(self, label: str, weight: float = 0.5, **kwargs) -> None:
|
||||
"""Füge ein Prioritäts-Thema hinzu."""
|
||||
self.priorities.append(PriorityItem(
|
||||
label=label,
|
||||
weight=weight,
|
||||
**kwargs
|
||||
))
|
||||
self.updated_at = datetime.utcnow()
|
||||
|
||||
def add_exclusion(self, keyword: str) -> None:
|
||||
"""Füge ein Ausschluss-Keyword hinzu."""
|
||||
if keyword not in self.exclusions:
|
||||
self.exclusions.append(keyword)
|
||||
self.updated_at = datetime.utcnow()
|
||||
|
||||
def add_positive_example(self, title: str, url: str, reason: str = "") -> None:
|
||||
"""Füge ein positives Beispiel hinzu (für Few-Shot Learning)."""
|
||||
self.positive_examples.append({
|
||||
"title": title,
|
||||
"url": url,
|
||||
"reason": reason,
|
||||
"added_at": datetime.utcnow().isoformat(),
|
||||
})
|
||||
# Begrenze auf letzte 20 Beispiele
|
||||
self.positive_examples = self.positive_examples[-20:]
|
||||
self.updated_at = datetime.utcnow()
|
||||
|
||||
def add_negative_example(self, title: str, url: str, reason: str = "") -> None:
|
||||
"""Füge ein negatives Beispiel hinzu."""
|
||||
self.negative_examples.append({
|
||||
"title": title,
|
||||
"url": url,
|
||||
"reason": reason,
|
||||
"added_at": datetime.utcnow().isoformat(),
|
||||
})
|
||||
# Begrenze auf letzte 20 Beispiele
|
||||
self.negative_examples = self.negative_examples[-20:]
|
||||
self.updated_at = datetime.utcnow()
|
||||
|
||||
def update_from_feedback(self, alert_title: str, alert_url: str,
|
||||
is_relevant: bool, reason: str = "") -> None:
|
||||
"""
|
||||
Aktualisiere Profil basierend auf Nutzer-Feedback.
|
||||
|
||||
Args:
|
||||
alert_title: Titel des Alerts
|
||||
alert_url: URL des Alerts
|
||||
is_relevant: True wenn der Nutzer den Alert als relevant markiert hat
|
||||
reason: Optional - Grund für die Entscheidung
|
||||
"""
|
||||
if is_relevant:
|
||||
self.add_positive_example(alert_title, alert_url, reason)
|
||||
self.total_kept += 1
|
||||
else:
|
||||
self.add_negative_example(alert_title, alert_url, reason)
|
||||
self.total_dropped += 1
|
||||
|
||||
self.total_scored += 1
|
||||
|
||||
# Aktualisiere Accuracy-Schätzung (vereinfacht)
|
||||
if self.total_scored > 10:
|
||||
# Hier könnte eine komplexere Berechnung erfolgen
|
||||
# basierend auf Vergleich von Vorhersage vs. tatsächlichem Feedback
|
||||
pass
|
||||
|
||||
def get_prompt_context(self) -> str:
|
||||
"""
|
||||
Generiere Kontext für LLM-Prompt.
|
||||
|
||||
Dieser Text wird in den System-Prompt des Relevanz-Scorers eingefügt.
|
||||
"""
|
||||
lines = ["## Relevanzprofil des Nutzers\n"]
|
||||
|
||||
# Prioritäten
|
||||
if self.priorities:
|
||||
lines.append("### Prioritäten (Themen von Interesse):")
|
||||
for p in self.priorities:
|
||||
if isinstance(p, dict):
|
||||
p = PriorityItem.from_dict(p)
|
||||
weight_label = "Sehr wichtig" if p.weight > 0.7 else "Wichtig" if p.weight > 0.4 else "Interessant"
|
||||
lines.append(f"- **{p.label}** ({weight_label})")
|
||||
if p.description:
|
||||
lines.append(f" {p.description}")
|
||||
if p.keywords:
|
||||
lines.append(f" Keywords: {', '.join(p.keywords)}")
|
||||
lines.append("")
|
||||
|
||||
# Ausschlüsse
|
||||
if self.exclusions:
|
||||
lines.append("### Ausschlüsse (ignorieren):")
|
||||
lines.append(f"Themen mit diesen Keywords: {', '.join(self.exclusions)}")
|
||||
lines.append("")
|
||||
|
||||
# Positive Beispiele
|
||||
if self.positive_examples:
|
||||
lines.append("### Beispiele für relevante Alerts:")
|
||||
for ex in self.positive_examples[-5:]: # Letzte 5
|
||||
lines.append(f"- \"{ex['title']}\"")
|
||||
if ex.get("reason"):
|
||||
lines.append(f" Grund: {ex['reason']}")
|
||||
lines.append("")
|
||||
|
||||
# Negative Beispiele
|
||||
if self.negative_examples:
|
||||
lines.append("### Beispiele für irrelevante Alerts:")
|
||||
for ex in self.negative_examples[-5:]: # Letzte 5
|
||||
lines.append(f"- \"{ex['title']}\"")
|
||||
if ex.get("reason"):
|
||||
lines.append(f" Grund: {ex['reason']}")
|
||||
lines.append("")
|
||||
|
||||
# Policies
|
||||
if self.policies:
|
||||
lines.append("### Zusätzliche Regeln:")
|
||||
for key, value in self.policies.items():
|
||||
lines.append(f"- {key}: {value}")
|
||||
|
||||
return "\n".join(lines)
|
||||
|
||||
def to_dict(self) -> dict:
|
||||
"""Konvertiere zu Dictionary."""
|
||||
return {
|
||||
"id": self.id,
|
||||
"user_id": self.user_id,
|
||||
"priorities": [p.to_dict() if isinstance(p, PriorityItem) else p
|
||||
for p in self.priorities],
|
||||
"exclusions": self.exclusions,
|
||||
"positive_examples": self.positive_examples,
|
||||
"negative_examples": self.negative_examples,
|
||||
"policies": self.policies,
|
||||
"created_at": self.created_at.isoformat(),
|
||||
"updated_at": self.updated_at.isoformat(),
|
||||
"total_scored": self.total_scored,
|
||||
"total_kept": self.total_kept,
|
||||
"total_dropped": self.total_dropped,
|
||||
"accuracy_estimate": self.accuracy_estimate,
|
||||
}
|
||||
|
||||
@classmethod
|
||||
def from_dict(cls, data: dict) -> "RelevanceProfile":
|
||||
"""Erstelle RelevanceProfile aus Dictionary."""
|
||||
# Parse Timestamps
|
||||
for field_name in ["created_at", "updated_at"]:
|
||||
if field_name in data and isinstance(data[field_name], str):
|
||||
data[field_name] = datetime.fromisoformat(data[field_name])
|
||||
|
||||
# Parse Priorities
|
||||
if "priorities" in data:
|
||||
data["priorities"] = [
|
||||
PriorityItem.from_dict(p) if isinstance(p, dict) else p
|
||||
for p in data["priorities"]
|
||||
]
|
||||
|
||||
return cls(**data)
|
||||
|
||||
@classmethod
|
||||
def create_default_education_profile(cls) -> "RelevanceProfile":
|
||||
"""
|
||||
Erstelle ein Standard-Profil für Bildungsthemen.
|
||||
|
||||
Dieses Profil ist für Lehrkräfte/Schulpersonal optimiert.
|
||||
"""
|
||||
profile = cls()
|
||||
|
||||
# Bildungs-relevante Prioritäten
|
||||
profile.add_priority(
|
||||
"Inklusion",
|
||||
weight=0.9,
|
||||
keywords=["inklusiv", "Förderbedarf", "Behinderung", "Barrierefreiheit"],
|
||||
description="Inklusive Bildung, Förderschulen, Nachteilsausgleich"
|
||||
)
|
||||
profile.add_priority(
|
||||
"Datenschutz Schule",
|
||||
weight=0.85,
|
||||
keywords=["DSGVO", "Schülerfotos", "Einwilligung", "personenbezogene Daten"],
|
||||
description="DSGVO in Schulen, Datenschutz bei Klassenfotos"
|
||||
)
|
||||
profile.add_priority(
|
||||
"Schulrecht Bayern",
|
||||
weight=0.8,
|
||||
keywords=["BayEUG", "Schulordnung", "Kultusministerium", "Bayern"],
|
||||
description="Bayerisches Schulrecht, Verordnungen"
|
||||
)
|
||||
profile.add_priority(
|
||||
"Digitalisierung Schule",
|
||||
weight=0.7,
|
||||
keywords=["DigitalPakt", "Tablet-Klasse", "Lernplattform"],
|
||||
description="Digitale Medien im Unterricht"
|
||||
)
|
||||
profile.add_priority(
|
||||
"Elternarbeit",
|
||||
weight=0.6,
|
||||
keywords=["Elternbeirat", "Elternabend", "Kommunikation"],
|
||||
description="Zusammenarbeit mit Eltern"
|
||||
)
|
||||
|
||||
# Standard-Ausschlüsse
|
||||
profile.exclusions = [
|
||||
"Stellenanzeige",
|
||||
"Praktikum gesucht",
|
||||
"Werbung",
|
||||
"Pressemitteilung", # Oft generisch
|
||||
]
|
||||
|
||||
# Policies
|
||||
profile.policies = {
|
||||
"prefer_german_sources": True,
|
||||
"max_age_days": 30, # Ältere Alerts ignorieren
|
||||
"min_content_length": 100, # Sehr kurze Snippets ignorieren
|
||||
}
|
||||
|
||||
return profile
|
||||
|
||||
def __repr__(self) -> str:
|
||||
return f"RelevanceProfile(id={self.id[:8]}, priorities={len(self.priorities)}, examples={len(self.positive_examples) + len(self.negative_examples)})"
|
||||
31
docs-src/backend/llm_gateway/models/__init__.py
Normal file
31
docs-src/backend/llm_gateway/models/__init__.py
Normal file
@@ -0,0 +1,31 @@
|
||||
"""
|
||||
Pydantic Models für OpenAI-kompatible API.
|
||||
"""
|
||||
|
||||
from .chat import (
|
||||
ChatMessage,
|
||||
ChatCompletionRequest,
|
||||
ChatCompletionResponse,
|
||||
ChatCompletionChunk,
|
||||
ChatChoice,
|
||||
ChatChoiceDelta,
|
||||
Usage,
|
||||
ToolCall,
|
||||
FunctionCall,
|
||||
Tool,
|
||||
ToolFunction,
|
||||
)
|
||||
|
||||
__all__ = [
|
||||
"ChatMessage",
|
||||
"ChatCompletionRequest",
|
||||
"ChatCompletionResponse",
|
||||
"ChatCompletionChunk",
|
||||
"ChatChoice",
|
||||
"ChatChoiceDelta",
|
||||
"Usage",
|
||||
"ToolCall",
|
||||
"FunctionCall",
|
||||
"Tool",
|
||||
"ToolFunction",
|
||||
]
|
||||
135
docs-src/backend/llm_gateway/models/chat.py
Normal file
135
docs-src/backend/llm_gateway/models/chat.py
Normal file
@@ -0,0 +1,135 @@
|
||||
"""
|
||||
OpenAI-kompatible Chat Completion Models.
|
||||
|
||||
Basiert auf OpenAI API Spezifikation:
|
||||
https://platform.openai.com/docs/api-reference/chat/create
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
from typing import Optional, Literal, Any, Union, List, Dict
|
||||
from pydantic import BaseModel, Field
|
||||
import time
|
||||
import uuid
|
||||
|
||||
|
||||
class FunctionCall(BaseModel):
|
||||
"""Function call in einer Tool-Anfrage."""
|
||||
name: str
|
||||
arguments: str # JSON string
|
||||
|
||||
|
||||
class ToolCall(BaseModel):
|
||||
"""Tool Call vom Modell."""
|
||||
id: str = Field(default_factory=lambda: f"call_{uuid.uuid4().hex[:12]}")
|
||||
type: Literal["function"] = "function"
|
||||
function: FunctionCall
|
||||
|
||||
|
||||
class ChatMessage(BaseModel):
|
||||
"""Eine Nachricht im Chat."""
|
||||
role: Literal["system", "user", "assistant", "tool"]
|
||||
content: Optional[str] = None
|
||||
name: Optional[str] = None
|
||||
tool_call_id: Optional[str] = None
|
||||
tool_calls: Optional[list[ToolCall]] = None
|
||||
|
||||
|
||||
class ToolFunction(BaseModel):
|
||||
"""Definition einer Tool-Funktion."""
|
||||
name: str
|
||||
description: Optional[str] = None
|
||||
parameters: dict[str, Any] = Field(default_factory=dict)
|
||||
|
||||
|
||||
class Tool(BaseModel):
|
||||
"""Tool-Definition für Function Calling."""
|
||||
type: Literal["function"] = "function"
|
||||
function: ToolFunction
|
||||
|
||||
|
||||
class RequestMetadata(BaseModel):
|
||||
"""Zusätzliche Metadaten für die Anfrage."""
|
||||
playbook_id: Optional[str] = None
|
||||
tenant_id: Optional[str] = None
|
||||
user_id: Optional[str] = None
|
||||
|
||||
|
||||
class ChatCompletionRequest(BaseModel):
|
||||
"""Request für Chat Completions."""
|
||||
model: str
|
||||
messages: list[ChatMessage]
|
||||
stream: bool = False
|
||||
temperature: Optional[float] = Field(default=0.7, ge=0, le=2)
|
||||
top_p: Optional[float] = Field(default=1.0, ge=0, le=1)
|
||||
max_tokens: Optional[int] = Field(default=None, ge=1)
|
||||
stop: Optional[Union[List[str], str]] = None
|
||||
presence_penalty: Optional[float] = Field(default=0, ge=-2, le=2)
|
||||
frequency_penalty: Optional[float] = Field(default=0, ge=-2, le=2)
|
||||
user: Optional[str] = None
|
||||
tools: Optional[list[Tool]] = None
|
||||
tool_choice: Optional[Union[str, Dict[str, Any]]] = None
|
||||
metadata: Optional[RequestMetadata] = None
|
||||
|
||||
|
||||
class ChatChoice(BaseModel):
|
||||
"""Ein Choice in der Response."""
|
||||
index: int = 0
|
||||
message: ChatMessage
|
||||
finish_reason: Optional[Literal["stop", "length", "tool_calls", "content_filter"]] = None
|
||||
|
||||
|
||||
class ChatChoiceDelta(BaseModel):
|
||||
"""Delta für Streaming Response."""
|
||||
role: Optional[str] = None
|
||||
content: Optional[str] = None
|
||||
tool_calls: Optional[list[ToolCall]] = None
|
||||
|
||||
|
||||
class StreamChoice(BaseModel):
|
||||
"""Choice in Streaming Response."""
|
||||
index: int = 0
|
||||
delta: ChatChoiceDelta
|
||||
finish_reason: Optional[Literal["stop", "length", "tool_calls", "content_filter"]] = None
|
||||
|
||||
|
||||
class Usage(BaseModel):
|
||||
"""Token Usage Statistiken."""
|
||||
prompt_tokens: int = 0
|
||||
completion_tokens: int = 0
|
||||
total_tokens: int = 0
|
||||
|
||||
|
||||
class ChatCompletionResponse(BaseModel):
|
||||
"""Response für Chat Completions (non-streaming)."""
|
||||
id: str = Field(default_factory=lambda: f"chatcmpl-{uuid.uuid4().hex[:12]}")
|
||||
object: Literal["chat.completion"] = "chat.completion"
|
||||
created: int = Field(default_factory=lambda: int(time.time()))
|
||||
model: str
|
||||
choices: list[ChatChoice]
|
||||
usage: Optional[Usage] = None
|
||||
|
||||
|
||||
class ChatCompletionChunk(BaseModel):
|
||||
"""Chunk für Streaming Response."""
|
||||
id: str = Field(default_factory=lambda: f"chatcmpl-{uuid.uuid4().hex[:12]}")
|
||||
object: Literal["chat.completion.chunk"] = "chat.completion.chunk"
|
||||
created: int = Field(default_factory=lambda: int(time.time()))
|
||||
model: str
|
||||
choices: list[StreamChoice]
|
||||
|
||||
|
||||
# Model Info
|
||||
class ModelInfo(BaseModel):
|
||||
"""Information über ein verfügbares Modell."""
|
||||
id: str
|
||||
object: Literal["model"] = "model"
|
||||
created: int = Field(default_factory=lambda: int(time.time()))
|
||||
owned_by: str = "breakpilot"
|
||||
description: Optional[str] = None
|
||||
context_length: int = 8192
|
||||
|
||||
|
||||
class ModelListResponse(BaseModel):
|
||||
"""Response für /v1/models."""
|
||||
object: Literal["list"] = "list"
|
||||
data: list[ModelInfo]
|
||||
155
docs-src/backend/migrations/add_abitur_docs_tables.sql
Normal file
155
docs-src/backend/migrations/add_abitur_docs_tables.sql
Normal file
@@ -0,0 +1,155 @@
|
||||
-- ============================================================================
|
||||
-- Abitur Documents Migration
|
||||
-- ============================================================================
|
||||
-- Creates tables for storing Abitur documents (NiBiS, etc.) persistently.
|
||||
-- Run with: psql -h localhost -U breakpilot -d breakpilot_dev -f add_abitur_docs_tables.sql
|
||||
--
|
||||
-- Tables created:
|
||||
-- 1. abitur_dokumente - Main document metadata
|
||||
-- 2. abitur_dokumente_chunks - Text chunks for RAG (optional, Qdrant primary)
|
||||
-- ============================================================================
|
||||
|
||||
-- ============================================================================
|
||||
-- ENUMS
|
||||
-- ============================================================================
|
||||
|
||||
DO $$ BEGIN
|
||||
CREATE TYPE bundesland_enum AS ENUM (
|
||||
'niedersachsen', 'bayern', 'baden_wuerttemberg', 'nordrhein_westfalen',
|
||||
'hessen', 'sachsen', 'thueringen', 'berlin', 'hamburg',
|
||||
'schleswig_holstein', 'bremen', 'brandenburg', 'mecklenburg_vorpommern',
|
||||
'sachsen_anhalt', 'rheinland_pfalz', 'saarland'
|
||||
);
|
||||
EXCEPTION WHEN duplicate_object THEN NULL;
|
||||
END $$;
|
||||
|
||||
DO $$ BEGIN
|
||||
CREATE TYPE abitur_fach_enum AS ENUM (
|
||||
'deutsch', 'englisch', 'mathematik', 'biologie', 'chemie', 'physik',
|
||||
'geschichte', 'erdkunde', 'politik_wirtschaft', 'franzoesisch', 'spanisch',
|
||||
'latein', 'griechisch', 'kunst', 'musik', 'sport', 'informatik',
|
||||
'ev_religion', 'kath_religion', 'werte_normen', 'brc', 'bvw',
|
||||
'ernaehrung', 'mechatronik', 'gesundheit_pflege', 'paedagogik_psychologie'
|
||||
);
|
||||
EXCEPTION WHEN duplicate_object THEN NULL;
|
||||
END $$;
|
||||
|
||||
DO $$ BEGIN
|
||||
CREATE TYPE abitur_niveau_enum AS ENUM ('eA', 'gA');
|
||||
EXCEPTION WHEN duplicate_object THEN NULL;
|
||||
END $$;
|
||||
|
||||
DO $$ BEGIN
|
||||
CREATE TYPE abitur_dok_typ_enum AS ENUM (
|
||||
'aufgabe', 'erwartungshorizont', 'deckblatt', 'material',
|
||||
'hoerverstehen', 'sprachmittlung', 'bewertungsbogen'
|
||||
);
|
||||
EXCEPTION WHEN duplicate_object THEN NULL;
|
||||
END $$;
|
||||
|
||||
DO $$ BEGIN
|
||||
CREATE TYPE abitur_status_enum AS ENUM (
|
||||
'pending', 'processing', 'recognized', 'confirmed', 'indexed', 'error'
|
||||
);
|
||||
EXCEPTION WHEN duplicate_object THEN NULL;
|
||||
END $$;
|
||||
|
||||
-- ============================================================================
|
||||
-- MAIN TABLE: abitur_dokumente
|
||||
-- ============================================================================
|
||||
|
||||
CREATE TABLE IF NOT EXISTS abitur_dokumente (
|
||||
id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
|
||||
|
||||
-- File info
|
||||
dateiname VARCHAR(500) NOT NULL,
|
||||
original_dateiname VARCHAR(500) NOT NULL,
|
||||
file_path VARCHAR(1000), -- MinIO path or local path
|
||||
file_size BIGINT DEFAULT 0,
|
||||
file_hash VARCHAR(64), -- SHA-256 for deduplication
|
||||
|
||||
-- Metadata
|
||||
bundesland bundesland_enum NOT NULL DEFAULT 'niedersachsen',
|
||||
fach abitur_fach_enum NOT NULL,
|
||||
jahr INTEGER NOT NULL CHECK (jahr >= 2000 AND jahr <= 2100),
|
||||
niveau abitur_niveau_enum NOT NULL DEFAULT 'eA',
|
||||
typ abitur_dok_typ_enum NOT NULL DEFAULT 'aufgabe',
|
||||
aufgaben_nummer VARCHAR(20), -- I, II, III, 1, 2, etc.
|
||||
variante VARCHAR(50), -- BG, Tech, Wirt, etc.
|
||||
|
||||
-- Processing status
|
||||
status abitur_status_enum NOT NULL DEFAULT 'pending',
|
||||
confidence REAL DEFAULT 0.0,
|
||||
|
||||
-- Vector store integration
|
||||
indexed BOOLEAN DEFAULT FALSE,
|
||||
vector_ids TEXT[], -- Qdrant vector IDs
|
||||
qdrant_collection VARCHAR(100) DEFAULT 'bp_nibis_eh',
|
||||
|
||||
-- Source tracking
|
||||
source_dir VARCHAR(500), -- Original source directory
|
||||
import_batch_id UUID, -- For batch imports
|
||||
|
||||
-- Timestamps
|
||||
created_at TIMESTAMPTZ DEFAULT NOW(),
|
||||
updated_at TIMESTAMPTZ DEFAULT NOW(),
|
||||
indexed_at TIMESTAMPTZ
|
||||
);
|
||||
|
||||
-- Indexes for common queries
|
||||
CREATE INDEX IF NOT EXISTS idx_abitur_dok_bundesland ON abitur_dokumente(bundesland);
|
||||
CREATE INDEX IF NOT EXISTS idx_abitur_dok_fach ON abitur_dokumente(fach);
|
||||
CREATE INDEX IF NOT EXISTS idx_abitur_dok_jahr ON abitur_dokumente(jahr);
|
||||
CREATE INDEX IF NOT EXISTS idx_abitur_dok_niveau ON abitur_dokumente(niveau);
|
||||
CREATE INDEX IF NOT EXISTS idx_abitur_dok_typ ON abitur_dokumente(typ);
|
||||
CREATE INDEX IF NOT EXISTS idx_abitur_dok_status ON abitur_dokumente(status);
|
||||
CREATE INDEX IF NOT EXISTS idx_abitur_dok_indexed ON abitur_dokumente(indexed);
|
||||
CREATE INDEX IF NOT EXISTS idx_abitur_dok_file_hash ON abitur_dokumente(file_hash);
|
||||
|
||||
-- Composite index for typical searches
|
||||
CREATE INDEX IF NOT EXISTS idx_abitur_dok_search
|
||||
ON abitur_dokumente(bundesland, fach, jahr, niveau);
|
||||
|
||||
-- ============================================================================
|
||||
-- TRIGGER: Auto-update updated_at
|
||||
-- ============================================================================
|
||||
|
||||
CREATE OR REPLACE FUNCTION update_abitur_dok_timestamp()
|
||||
RETURNS TRIGGER AS $$
|
||||
BEGIN
|
||||
NEW.updated_at = NOW();
|
||||
RETURN NEW;
|
||||
END;
|
||||
$$ LANGUAGE plpgsql;
|
||||
|
||||
DROP TRIGGER IF EXISTS trigger_abitur_dok_updated ON abitur_dokumente;
|
||||
CREATE TRIGGER trigger_abitur_dok_updated
|
||||
BEFORE UPDATE ON abitur_dokumente
|
||||
FOR EACH ROW
|
||||
EXECUTE FUNCTION update_abitur_dok_timestamp();
|
||||
|
||||
-- ============================================================================
|
||||
-- HELPER VIEWS
|
||||
-- ============================================================================
|
||||
|
||||
CREATE OR REPLACE VIEW v_abitur_dok_stats AS
|
||||
SELECT
|
||||
bundesland,
|
||||
fach,
|
||||
jahr,
|
||||
COUNT(*) as total,
|
||||
COUNT(*) FILTER (WHERE indexed = TRUE) as indexed_count,
|
||||
COUNT(*) FILTER (WHERE typ = 'aufgabe') as aufgaben_count,
|
||||
COUNT(*) FILTER (WHERE typ = 'erwartungshorizont') as ewh_count
|
||||
FROM abitur_dokumente
|
||||
GROUP BY bundesland, fach, jahr
|
||||
ORDER BY jahr DESC, fach;
|
||||
|
||||
-- ============================================================================
|
||||
-- SAMPLE DATA CHECK
|
||||
-- ============================================================================
|
||||
|
||||
-- Show table structure
|
||||
-- \d abitur_dokumente
|
||||
|
||||
SELECT 'Migration completed: abitur_dokumente table created' AS status;
|
||||
293
docs-src/backend/migrations/add_agent_core_tables.sql
Normal file
293
docs-src/backend/migrations/add_agent_core_tables.sql
Normal file
@@ -0,0 +1,293 @@
|
||||
-- Migration: Add Multi-Agent Architecture Tables
|
||||
-- Date: 2025-01-15
|
||||
-- Description: Creates tables for agent sessions, memory store, and message audit
|
||||
|
||||
-- Enable required extensions
|
||||
CREATE EXTENSION IF NOT EXISTS "uuid-ossp";
|
||||
CREATE EXTENSION IF NOT EXISTS "pgcrypto";
|
||||
|
||||
-- ============================================================================
|
||||
-- 1. Agent Sessions Table
|
||||
-- ============================================================================
|
||||
-- Stores agent session data with state, context, and checkpoints
|
||||
CREATE TABLE IF NOT EXISTS agent_sessions (
|
||||
id UUID PRIMARY KEY DEFAULT uuid_generate_v4(),
|
||||
agent_type VARCHAR(50) NOT NULL,
|
||||
user_id UUID REFERENCES users(id) ON DELETE SET NULL,
|
||||
state VARCHAR(20) NOT NULL DEFAULT 'active'
|
||||
CHECK (state IN ('active', 'paused', 'completed', 'failed', 'deleted')),
|
||||
context JSONB DEFAULT '{}'::jsonb,
|
||||
checkpoints JSONB DEFAULT '[]'::jsonb,
|
||||
metadata JSONB DEFAULT '{}'::jsonb,
|
||||
created_at TIMESTAMPTZ DEFAULT NOW(),
|
||||
updated_at TIMESTAMPTZ DEFAULT NOW(),
|
||||
last_heartbeat TIMESTAMPTZ DEFAULT NOW()
|
||||
);
|
||||
|
||||
-- Indexes for common queries
|
||||
CREATE INDEX IF NOT EXISTS idx_agent_sessions_user
|
||||
ON agent_sessions(user_id);
|
||||
CREATE INDEX IF NOT EXISTS idx_agent_sessions_state
|
||||
ON agent_sessions(state) WHERE state = 'active';
|
||||
CREATE INDEX IF NOT EXISTS idx_agent_sessions_agent_type
|
||||
ON agent_sessions(agent_type);
|
||||
CREATE INDEX IF NOT EXISTS idx_agent_sessions_heartbeat
|
||||
ON agent_sessions(last_heartbeat);
|
||||
CREATE INDEX IF NOT EXISTS idx_agent_sessions_created
|
||||
ON agent_sessions(created_at DESC);
|
||||
|
||||
-- GIN index for JSONB context queries
|
||||
CREATE INDEX IF NOT EXISTS idx_agent_sessions_context
|
||||
ON agent_sessions USING GIN (context jsonb_path_ops);
|
||||
|
||||
-- Comments for documentation
|
||||
COMMENT ON TABLE agent_sessions IS 'Stores agent session state and checkpoints for recovery';
|
||||
COMMENT ON COLUMN agent_sessions.agent_type IS 'Type: tutor-agent, grader-agent, quality-judge, alert-agent, orchestrator';
|
||||
COMMENT ON COLUMN agent_sessions.state IS 'Session state: active, paused, completed, failed, deleted';
|
||||
COMMENT ON COLUMN agent_sessions.context IS 'Session context data (entities, conversation state)';
|
||||
COMMENT ON COLUMN agent_sessions.checkpoints IS 'Recovery checkpoints as JSON array';
|
||||
COMMENT ON COLUMN agent_sessions.last_heartbeat IS 'Last heartbeat timestamp for liveness detection';
|
||||
|
||||
-- ============================================================================
|
||||
-- 2. Agent Memory Table
|
||||
-- ============================================================================
|
||||
-- Long-term memory store for agents with TTL support
|
||||
CREATE TABLE IF NOT EXISTS agent_memory (
|
||||
id UUID PRIMARY KEY DEFAULT uuid_generate_v4(),
|
||||
namespace VARCHAR(100) NOT NULL DEFAULT 'breakpilot',
|
||||
key VARCHAR(500) NOT NULL,
|
||||
value JSONB NOT NULL,
|
||||
agent_id VARCHAR(50) NOT NULL,
|
||||
access_count INTEGER DEFAULT 0,
|
||||
created_at TIMESTAMPTZ DEFAULT NOW(),
|
||||
updated_at TIMESTAMPTZ DEFAULT NOW(),
|
||||
last_accessed TIMESTAMPTZ,
|
||||
expires_at TIMESTAMPTZ,
|
||||
metadata JSONB DEFAULT '{}'::jsonb,
|
||||
|
||||
-- Unique constraint per namespace
|
||||
CONSTRAINT agent_memory_namespace_key_unique UNIQUE (namespace, key)
|
||||
);
|
||||
|
||||
-- Indexes for efficient queries
|
||||
CREATE INDEX IF NOT EXISTS idx_agent_memory_namespace
|
||||
ON agent_memory(namespace);
|
||||
CREATE INDEX IF NOT EXISTS idx_agent_memory_agent
|
||||
ON agent_memory(agent_id);
|
||||
CREATE INDEX IF NOT EXISTS idx_agent_memory_expires
|
||||
ON agent_memory(expires_at) WHERE expires_at IS NOT NULL;
|
||||
CREATE INDEX IF NOT EXISTS idx_agent_memory_key_pattern
|
||||
ON agent_memory(key varchar_pattern_ops);
|
||||
CREATE INDEX IF NOT EXISTS idx_agent_memory_access_count
|
||||
ON agent_memory(access_count DESC);
|
||||
|
||||
-- GIN index for value queries
|
||||
CREATE INDEX IF NOT EXISTS idx_agent_memory_value
|
||||
ON agent_memory USING GIN (value jsonb_path_ops);
|
||||
|
||||
-- Comments
|
||||
COMMENT ON TABLE agent_memory IS 'Long-term memory store for agents with TTL';
|
||||
COMMENT ON COLUMN agent_memory.namespace IS 'Namespace for isolation (default: breakpilot)';
|
||||
COMMENT ON COLUMN agent_memory.key IS 'Memory key (e.g., evaluation:math:student123)';
|
||||
COMMENT ON COLUMN agent_memory.value IS 'Stored value as JSONB';
|
||||
COMMENT ON COLUMN agent_memory.access_count IS 'Number of times this memory was accessed';
|
||||
COMMENT ON COLUMN agent_memory.expires_at IS 'When this memory expires (NULL = never)';
|
||||
|
||||
-- ============================================================================
|
||||
-- 3. Agent Messages Table (Audit Trail)
|
||||
-- ============================================================================
|
||||
-- Stores all inter-agent messages for audit and debugging
|
||||
CREATE TABLE IF NOT EXISTS agent_messages (
|
||||
id UUID PRIMARY KEY DEFAULT uuid_generate_v4(),
|
||||
sender VARCHAR(50) NOT NULL,
|
||||
receiver VARCHAR(50) NOT NULL,
|
||||
message_type VARCHAR(50) NOT NULL,
|
||||
payload JSONB NOT NULL,
|
||||
priority INTEGER DEFAULT 1 CHECK (priority BETWEEN 0 AND 3),
|
||||
correlation_id UUID,
|
||||
reply_to VARCHAR(50),
|
||||
created_at TIMESTAMPTZ DEFAULT NOW(),
|
||||
|
||||
-- Partition hint for future partitioning
|
||||
created_date DATE GENERATED ALWAYS AS (DATE(created_at)) STORED
|
||||
);
|
||||
|
||||
-- Indexes for message queries
|
||||
CREATE INDEX IF NOT EXISTS idx_agent_messages_correlation
|
||||
ON agent_messages(correlation_id) WHERE correlation_id IS NOT NULL;
|
||||
CREATE INDEX IF NOT EXISTS idx_agent_messages_sender
|
||||
ON agent_messages(sender);
|
||||
CREATE INDEX IF NOT EXISTS idx_agent_messages_receiver
|
||||
ON agent_messages(receiver);
|
||||
CREATE INDEX IF NOT EXISTS idx_agent_messages_type
|
||||
ON agent_messages(message_type);
|
||||
CREATE INDEX IF NOT EXISTS idx_agent_messages_created
|
||||
ON agent_messages(created_at DESC);
|
||||
CREATE INDEX IF NOT EXISTS idx_agent_messages_date
|
||||
ON agent_messages(created_date);
|
||||
|
||||
-- Comments
|
||||
COMMENT ON TABLE agent_messages IS 'Audit trail for inter-agent communication';
|
||||
COMMENT ON COLUMN agent_messages.priority IS '0=LOW, 1=NORMAL, 2=HIGH, 3=CRITICAL';
|
||||
COMMENT ON COLUMN agent_messages.correlation_id IS 'Links request/response pairs';
|
||||
COMMENT ON COLUMN agent_messages.created_date IS 'Partition column for future table partitioning';
|
||||
|
||||
-- ============================================================================
|
||||
-- 4. Helper Functions
|
||||
-- ============================================================================
|
||||
|
||||
-- Function to clean up expired memories
|
||||
CREATE OR REPLACE FUNCTION cleanup_expired_agent_memory()
|
||||
RETURNS INTEGER AS $$
|
||||
DECLARE
|
||||
deleted_count INTEGER;
|
||||
BEGIN
|
||||
DELETE FROM agent_memory
|
||||
WHERE expires_at IS NOT NULL AND expires_at < NOW();
|
||||
|
||||
GET DIAGNOSTICS deleted_count = ROW_COUNT;
|
||||
RETURN deleted_count;
|
||||
END;
|
||||
$$ LANGUAGE plpgsql;
|
||||
|
||||
COMMENT ON FUNCTION cleanup_expired_agent_memory() IS 'Removes expired memory entries, returns count';
|
||||
|
||||
-- Function to clean up stale sessions
|
||||
CREATE OR REPLACE FUNCTION cleanup_stale_agent_sessions(max_age_hours INTEGER DEFAULT 48)
|
||||
RETURNS INTEGER AS $$
|
||||
DECLARE
|
||||
updated_count INTEGER;
|
||||
BEGIN
|
||||
UPDATE agent_sessions
|
||||
SET state = 'failed',
|
||||
updated_at = NOW(),
|
||||
context = context || '{"failure_reason": "heartbeat_timeout"}'::jsonb
|
||||
WHERE state = 'active'
|
||||
AND last_heartbeat < NOW() - (max_age_hours || ' hours')::INTERVAL;
|
||||
|
||||
GET DIAGNOSTICS updated_count = ROW_COUNT;
|
||||
RETURN updated_count;
|
||||
END;
|
||||
$$ LANGUAGE plpgsql;
|
||||
|
||||
COMMENT ON FUNCTION cleanup_stale_agent_sessions(INTEGER) IS 'Marks stale sessions as failed, returns count';
|
||||
|
||||
-- Function to update session heartbeat
|
||||
CREATE OR REPLACE FUNCTION update_session_heartbeat(session_uuid UUID)
|
||||
RETURNS BOOLEAN AS $$
|
||||
BEGIN
|
||||
UPDATE agent_sessions
|
||||
SET last_heartbeat = NOW(), updated_at = NOW()
|
||||
WHERE id = session_uuid AND state = 'active';
|
||||
|
||||
RETURN FOUND;
|
||||
END;
|
||||
$$ LANGUAGE plpgsql;
|
||||
|
||||
COMMENT ON FUNCTION update_session_heartbeat(UUID) IS 'Updates session heartbeat, returns true if found';
|
||||
|
||||
-- ============================================================================
|
||||
-- 5. Triggers
|
||||
-- ============================================================================
|
||||
|
||||
-- Auto-update updated_at on agent_sessions
|
||||
CREATE OR REPLACE FUNCTION trigger_set_updated_at()
|
||||
RETURNS TRIGGER AS $$
|
||||
BEGIN
|
||||
NEW.updated_at = NOW();
|
||||
RETURN NEW;
|
||||
END;
|
||||
$$ LANGUAGE plpgsql;
|
||||
|
||||
DROP TRIGGER IF EXISTS agent_sessions_updated_at ON agent_sessions;
|
||||
CREATE TRIGGER agent_sessions_updated_at
|
||||
BEFORE UPDATE ON agent_sessions
|
||||
FOR EACH ROW
|
||||
EXECUTE FUNCTION trigger_set_updated_at();
|
||||
|
||||
DROP TRIGGER IF EXISTS agent_memory_updated_at ON agent_memory;
|
||||
CREATE TRIGGER agent_memory_updated_at
|
||||
BEFORE UPDATE ON agent_memory
|
||||
FOR EACH ROW
|
||||
EXECUTE FUNCTION trigger_set_updated_at();
|
||||
|
||||
-- ============================================================================
|
||||
-- 6. DSGVO Compliance: Audit Views
|
||||
-- ============================================================================
|
||||
|
||||
-- View for session audit without PII
|
||||
CREATE OR REPLACE VIEW v_agent_sessions_audit AS
|
||||
SELECT
|
||||
id,
|
||||
agent_type,
|
||||
-- Hash user_id for privacy
|
||||
CASE
|
||||
WHEN user_id IS NOT NULL
|
||||
THEN encode(digest(user_id::text, 'sha256'), 'hex')
|
||||
ELSE NULL
|
||||
END AS user_id_hash,
|
||||
state,
|
||||
-- Only expose non-sensitive context keys
|
||||
jsonb_build_object(
|
||||
'message_count', COALESCE((context->>'message_count')::int, 0),
|
||||
'intent_count', COALESCE(jsonb_array_length(context->'intent_history'), 0)
|
||||
) AS context_summary,
|
||||
jsonb_array_length(checkpoints) AS checkpoint_count,
|
||||
created_at,
|
||||
updated_at,
|
||||
last_heartbeat,
|
||||
EXTRACT(EPOCH FROM (updated_at - created_at)) AS session_duration_seconds
|
||||
FROM agent_sessions;
|
||||
|
||||
COMMENT ON VIEW v_agent_sessions_audit IS 'Privacy-safe view of agent sessions for auditing';
|
||||
|
||||
-- View for message audit
|
||||
CREATE OR REPLACE VIEW v_agent_messages_daily_stats AS
|
||||
SELECT
|
||||
created_date,
|
||||
sender,
|
||||
receiver,
|
||||
message_type,
|
||||
COUNT(*) AS message_count,
|
||||
AVG(priority) AS avg_priority
|
||||
FROM agent_messages
|
||||
GROUP BY created_date, sender, receiver, message_type;
|
||||
|
||||
COMMENT ON VIEW v_agent_messages_daily_stats IS 'Daily statistics for inter-agent messages';
|
||||
|
||||
-- ============================================================================
|
||||
-- 7. Sample Data for Testing (Optional - Comment out in production)
|
||||
-- ============================================================================
|
||||
/*
|
||||
-- Uncomment to insert sample data for testing
|
||||
|
||||
INSERT INTO agent_sessions (agent_type, state, context)
|
||||
VALUES
|
||||
('tutor-agent', 'active', '{"subject": "math", "grade": 10}'::jsonb),
|
||||
('grader-agent', 'active', '{"exam_type": "vorabitur"}'::jsonb);
|
||||
|
||||
INSERT INTO agent_memory (namespace, key, value, agent_id, expires_at)
|
||||
VALUES
|
||||
('breakpilot', 'test:memory:1', '{"test": true}'::jsonb, 'tutor-agent', NOW() + INTERVAL '30 days');
|
||||
*/
|
||||
|
||||
-- ============================================================================
|
||||
-- 8. Grants (Adjust based on your user/role setup)
|
||||
-- ============================================================================
|
||||
-- Uncomment and adjust for your environment
|
||||
/*
|
||||
GRANT SELECT, INSERT, UPDATE ON agent_sessions TO breakpilot_app;
|
||||
GRANT SELECT, INSERT, UPDATE, DELETE ON agent_memory TO breakpilot_app;
|
||||
GRANT SELECT, INSERT ON agent_messages TO breakpilot_app;
|
||||
GRANT EXECUTE ON FUNCTION cleanup_expired_agent_memory() TO breakpilot_app;
|
||||
GRANT EXECUTE ON FUNCTION cleanup_stale_agent_sessions(INTEGER) TO breakpilot_app;
|
||||
GRANT EXECUTE ON FUNCTION update_session_heartbeat(UUID) TO breakpilot_app;
|
||||
*/
|
||||
|
||||
-- ============================================================================
|
||||
-- Migration Complete
|
||||
-- ============================================================================
|
||||
-- To verify migration:
|
||||
-- \dt agent_*
|
||||
-- \df cleanup_*
|
||||
-- \dv v_agent_*
|
||||
807
docs-src/backend/migrations/add_compliance_tables.sql
Normal file
807
docs-src/backend/migrations/add_compliance_tables.sql
Normal file
@@ -0,0 +1,807 @@
|
||||
-- ============================================================================
|
||||
-- Compliance & Audit Framework Migration
|
||||
-- ============================================================================
|
||||
-- This migration creates all tables required for the Compliance module.
|
||||
-- Run with: psql -h localhost -U breakpilot -d breakpilot -f add_compliance_tables.sql
|
||||
--
|
||||
-- Tables created:
|
||||
-- 1. Core Compliance Framework (7 tables)
|
||||
-- 2. Service Module Registry (3 tables)
|
||||
-- 3. Audit Sessions & Sign-Off (2 tables)
|
||||
-- 4. ISO 27001 ISMS Models (11 tables)
|
||||
-- ============================================================================
|
||||
|
||||
-- ============================================================================
|
||||
-- ENUMS (PostgreSQL ENUM types)
|
||||
-- ============================================================================
|
||||
|
||||
DO $$ BEGIN
|
||||
CREATE TYPE regulation_type AS ENUM (
|
||||
'eu_regulation', 'eu_directive', 'de_law', 'bsi_standard', 'industry_standard'
|
||||
);
|
||||
EXCEPTION WHEN duplicate_object THEN NULL;
|
||||
END $$;
|
||||
|
||||
DO $$ BEGIN
|
||||
CREATE TYPE control_type AS ENUM ('preventive', 'detective', 'corrective');
|
||||
EXCEPTION WHEN duplicate_object THEN NULL;
|
||||
END $$;
|
||||
|
||||
DO $$ BEGIN
|
||||
CREATE TYPE control_domain AS ENUM (
|
||||
'gov', 'priv', 'iam', 'crypto', 'sdlc', 'ops', 'ai', 'cra', 'aud'
|
||||
);
|
||||
EXCEPTION WHEN duplicate_object THEN NULL;
|
||||
END $$;
|
||||
|
||||
DO $$ BEGIN
|
||||
CREATE TYPE control_status AS ENUM ('pass', 'partial', 'fail', 'n/a', 'planned');
|
||||
EXCEPTION WHEN duplicate_object THEN NULL;
|
||||
END $$;
|
||||
|
||||
DO $$ BEGIN
|
||||
CREATE TYPE risk_level AS ENUM ('low', 'medium', 'high', 'critical');
|
||||
EXCEPTION WHEN duplicate_object THEN NULL;
|
||||
END $$;
|
||||
|
||||
DO $$ BEGIN
|
||||
CREATE TYPE evidence_status AS ENUM ('valid', 'expired', 'pending', 'failed');
|
||||
EXCEPTION WHEN duplicate_object THEN NULL;
|
||||
END $$;
|
||||
|
||||
DO $$ BEGIN
|
||||
CREATE TYPE export_status AS ENUM ('pending', 'generating', 'completed', 'failed');
|
||||
EXCEPTION WHEN duplicate_object THEN NULL;
|
||||
END $$;
|
||||
|
||||
DO $$ BEGIN
|
||||
CREATE TYPE service_type AS ENUM (
|
||||
'backend', 'database', 'ai', 'communication', 'storage',
|
||||
'infrastructure', 'monitoring', 'security'
|
||||
);
|
||||
EXCEPTION WHEN duplicate_object THEN NULL;
|
||||
END $$;
|
||||
|
||||
DO $$ BEGIN
|
||||
CREATE TYPE relevance_level AS ENUM ('critical', 'high', 'medium', 'low');
|
||||
EXCEPTION WHEN duplicate_object THEN NULL;
|
||||
END $$;
|
||||
|
||||
DO $$ BEGIN
|
||||
CREATE TYPE audit_result AS ENUM (
|
||||
'compliant', 'compliant_notes', 'non_compliant', 'not_applicable', 'pending'
|
||||
);
|
||||
EXCEPTION WHEN duplicate_object THEN NULL;
|
||||
END $$;
|
||||
|
||||
DO $$ BEGIN
|
||||
CREATE TYPE audit_session_status AS ENUM ('draft', 'in_progress', 'completed', 'archived');
|
||||
EXCEPTION WHEN duplicate_object THEN NULL;
|
||||
END $$;
|
||||
|
||||
DO $$ BEGIN
|
||||
CREATE TYPE approval_status AS ENUM ('draft', 'under_review', 'approved', 'superseded');
|
||||
EXCEPTION WHEN duplicate_object THEN NULL;
|
||||
END $$;
|
||||
|
||||
DO $$ BEGIN
|
||||
CREATE TYPE finding_type AS ENUM ('major', 'minor', 'ofi', 'positive');
|
||||
EXCEPTION WHEN duplicate_object THEN NULL;
|
||||
END $$;
|
||||
|
||||
DO $$ BEGIN
|
||||
CREATE TYPE finding_status AS ENUM (
|
||||
'open', 'in_progress', 'capa_pending', 'verification_pending', 'verified', 'closed'
|
||||
);
|
||||
EXCEPTION WHEN duplicate_object THEN NULL;
|
||||
END $$;
|
||||
|
||||
DO $$ BEGIN
|
||||
CREATE TYPE capa_type AS ENUM ('corrective', 'preventive', 'both');
|
||||
EXCEPTION WHEN duplicate_object THEN NULL;
|
||||
END $$;
|
||||
|
||||
|
||||
-- ============================================================================
|
||||
-- CORE COMPLIANCE TABLES
|
||||
-- ============================================================================
|
||||
|
||||
-- Table 1: compliance_regulations
|
||||
CREATE TABLE IF NOT EXISTS compliance_regulations (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
code VARCHAR(20) UNIQUE NOT NULL,
|
||||
name VARCHAR(200) NOT NULL,
|
||||
full_name TEXT,
|
||||
regulation_type regulation_type NOT NULL,
|
||||
source_url VARCHAR(500),
|
||||
local_pdf_path VARCHAR(500),
|
||||
effective_date DATE,
|
||||
description TEXT,
|
||||
is_active BOOLEAN DEFAULT TRUE,
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
|
||||
CREATE INDEX IF NOT EXISTS ix_regulations_code ON compliance_regulations(code);
|
||||
|
||||
-- Table 2: compliance_requirements
|
||||
CREATE TABLE IF NOT EXISTS compliance_requirements (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
regulation_id VARCHAR(36) NOT NULL REFERENCES compliance_regulations(id),
|
||||
article VARCHAR(50) NOT NULL,
|
||||
paragraph VARCHAR(20),
|
||||
requirement_id_external VARCHAR(50),
|
||||
title VARCHAR(300) NOT NULL,
|
||||
description TEXT,
|
||||
requirement_text TEXT,
|
||||
breakpilot_interpretation TEXT,
|
||||
implementation_status VARCHAR(30) DEFAULT 'not_started',
|
||||
implementation_details TEXT,
|
||||
code_references JSONB,
|
||||
documentation_links JSONB,
|
||||
evidence_description TEXT,
|
||||
evidence_artifacts JSONB,
|
||||
auditor_notes TEXT,
|
||||
audit_status VARCHAR(30) DEFAULT 'pending',
|
||||
last_audit_date TIMESTAMP,
|
||||
last_auditor VARCHAR(100),
|
||||
is_applicable BOOLEAN DEFAULT TRUE,
|
||||
applicability_reason TEXT,
|
||||
priority INTEGER DEFAULT 2,
|
||||
source_page INTEGER,
|
||||
source_section VARCHAR(100),
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
|
||||
CREATE INDEX IF NOT EXISTS ix_requirement_regulation_article ON compliance_requirements(regulation_id, article);
|
||||
CREATE INDEX IF NOT EXISTS ix_requirement_audit_status ON compliance_requirements(audit_status);
|
||||
CREATE INDEX IF NOT EXISTS ix_requirement_impl_status ON compliance_requirements(implementation_status);
|
||||
|
||||
-- Table 3: compliance_controls
|
||||
CREATE TABLE IF NOT EXISTS compliance_controls (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
control_id VARCHAR(20) UNIQUE NOT NULL,
|
||||
domain control_domain NOT NULL,
|
||||
control_type control_type NOT NULL,
|
||||
title VARCHAR(300) NOT NULL,
|
||||
description TEXT,
|
||||
pass_criteria TEXT NOT NULL,
|
||||
implementation_guidance TEXT,
|
||||
code_reference VARCHAR(500),
|
||||
documentation_url VARCHAR(500),
|
||||
is_automated BOOLEAN DEFAULT FALSE,
|
||||
automation_tool VARCHAR(100),
|
||||
automation_config JSONB,
|
||||
status control_status DEFAULT 'planned',
|
||||
status_notes TEXT,
|
||||
owner VARCHAR(100),
|
||||
review_frequency_days INTEGER DEFAULT 90,
|
||||
last_reviewed_at TIMESTAMP,
|
||||
next_review_at TIMESTAMP,
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
|
||||
CREATE INDEX IF NOT EXISTS ix_control_id ON compliance_controls(control_id);
|
||||
CREATE INDEX IF NOT EXISTS ix_control_domain_status ON compliance_controls(domain, status);
|
||||
|
||||
-- Table 4: compliance_control_mappings
|
||||
CREATE TABLE IF NOT EXISTS compliance_control_mappings (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
requirement_id VARCHAR(36) NOT NULL REFERENCES compliance_requirements(id),
|
||||
control_id VARCHAR(36) NOT NULL REFERENCES compliance_controls(id),
|
||||
coverage_level VARCHAR(20) DEFAULT 'full',
|
||||
notes TEXT,
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
|
||||
CREATE UNIQUE INDEX IF NOT EXISTS ix_mapping_req_ctrl ON compliance_control_mappings(requirement_id, control_id);
|
||||
|
||||
-- Table 5: compliance_evidence
|
||||
CREATE TABLE IF NOT EXISTS compliance_evidence (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
control_id VARCHAR(36) NOT NULL REFERENCES compliance_controls(id),
|
||||
evidence_type VARCHAR(50) NOT NULL,
|
||||
title VARCHAR(300) NOT NULL,
|
||||
description TEXT,
|
||||
artifact_path VARCHAR(500),
|
||||
artifact_url VARCHAR(500),
|
||||
artifact_hash VARCHAR(64),
|
||||
file_size_bytes INTEGER,
|
||||
mime_type VARCHAR(100),
|
||||
valid_from TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
valid_until TIMESTAMP,
|
||||
status evidence_status DEFAULT 'valid',
|
||||
source VARCHAR(100),
|
||||
ci_job_id VARCHAR(100),
|
||||
uploaded_by VARCHAR(100),
|
||||
collected_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
|
||||
CREATE INDEX IF NOT EXISTS ix_evidence_control_type ON compliance_evidence(control_id, evidence_type);
|
||||
CREATE INDEX IF NOT EXISTS ix_evidence_status ON compliance_evidence(status);
|
||||
|
||||
-- Table 6: compliance_risks
|
||||
CREATE TABLE IF NOT EXISTS compliance_risks (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
risk_id VARCHAR(20) UNIQUE NOT NULL,
|
||||
title VARCHAR(300) NOT NULL,
|
||||
description TEXT,
|
||||
category VARCHAR(50) NOT NULL,
|
||||
likelihood INTEGER NOT NULL,
|
||||
impact INTEGER NOT NULL,
|
||||
inherent_risk risk_level NOT NULL,
|
||||
mitigating_controls JSONB,
|
||||
residual_likelihood INTEGER,
|
||||
residual_impact INTEGER,
|
||||
residual_risk risk_level,
|
||||
owner VARCHAR(100),
|
||||
status VARCHAR(20) DEFAULT 'open',
|
||||
treatment_plan TEXT,
|
||||
identified_date DATE DEFAULT CURRENT_DATE,
|
||||
review_date DATE,
|
||||
last_assessed_at TIMESTAMP,
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
|
||||
CREATE INDEX IF NOT EXISTS ix_risk_id ON compliance_risks(risk_id);
|
||||
CREATE INDEX IF NOT EXISTS ix_risk_category_status ON compliance_risks(category, status);
|
||||
CREATE INDEX IF NOT EXISTS ix_risk_inherent ON compliance_risks(inherent_risk);
|
||||
|
||||
-- Table 7: compliance_audit_exports
|
||||
CREATE TABLE IF NOT EXISTS compliance_audit_exports (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
export_type VARCHAR(50) NOT NULL,
|
||||
export_name VARCHAR(200),
|
||||
included_regulations JSONB,
|
||||
included_domains JSONB,
|
||||
date_range_start DATE,
|
||||
date_range_end DATE,
|
||||
requested_by VARCHAR(100) NOT NULL,
|
||||
requested_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
completed_at TIMESTAMP,
|
||||
file_path VARCHAR(500),
|
||||
file_hash VARCHAR(64),
|
||||
file_size_bytes INTEGER,
|
||||
status export_status DEFAULT 'pending',
|
||||
error_message TEXT,
|
||||
total_controls INTEGER,
|
||||
total_evidence INTEGER,
|
||||
compliance_score FLOAT,
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
|
||||
|
||||
-- ============================================================================
|
||||
-- SERVICE MODULE REGISTRY TABLES
|
||||
-- ============================================================================
|
||||
|
||||
-- Table 8: compliance_service_modules
|
||||
CREATE TABLE IF NOT EXISTS compliance_service_modules (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
name VARCHAR(100) UNIQUE NOT NULL,
|
||||
display_name VARCHAR(200) NOT NULL,
|
||||
description TEXT,
|
||||
service_type service_type NOT NULL,
|
||||
port INTEGER,
|
||||
technology_stack JSONB,
|
||||
repository_path VARCHAR(500),
|
||||
docker_image VARCHAR(200),
|
||||
data_categories JSONB,
|
||||
processes_pii BOOLEAN DEFAULT FALSE,
|
||||
processes_health_data BOOLEAN DEFAULT FALSE,
|
||||
ai_components BOOLEAN DEFAULT FALSE,
|
||||
is_active BOOLEAN DEFAULT TRUE,
|
||||
criticality VARCHAR(20) DEFAULT 'medium',
|
||||
compliance_score FLOAT,
|
||||
last_compliance_check TIMESTAMP,
|
||||
owner_team VARCHAR(100),
|
||||
owner_contact VARCHAR(200),
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
|
||||
CREATE INDEX IF NOT EXISTS ix_module_name ON compliance_service_modules(name);
|
||||
CREATE INDEX IF NOT EXISTS ix_module_type_active ON compliance_service_modules(service_type, is_active);
|
||||
|
||||
-- Table 9: compliance_module_regulations
|
||||
CREATE TABLE IF NOT EXISTS compliance_module_regulations (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
module_id VARCHAR(36) NOT NULL REFERENCES compliance_service_modules(id),
|
||||
regulation_id VARCHAR(36) NOT NULL REFERENCES compliance_regulations(id),
|
||||
relevance_level relevance_level DEFAULT 'medium',
|
||||
notes TEXT,
|
||||
applicable_articles JSONB,
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
|
||||
CREATE UNIQUE INDEX IF NOT EXISTS ix_module_regulation ON compliance_module_regulations(module_id, regulation_id);
|
||||
|
||||
-- Table 10: compliance_module_risks
|
||||
CREATE TABLE IF NOT EXISTS compliance_module_risks (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
module_id VARCHAR(36) NOT NULL REFERENCES compliance_service_modules(id),
|
||||
risk_id VARCHAR(36) NOT NULL REFERENCES compliance_risks(id),
|
||||
module_likelihood INTEGER,
|
||||
module_impact INTEGER,
|
||||
module_risk_level risk_level,
|
||||
assessment_notes TEXT,
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
|
||||
CREATE UNIQUE INDEX IF NOT EXISTS ix_module_risk ON compliance_module_risks(module_id, risk_id);
|
||||
|
||||
|
||||
-- ============================================================================
|
||||
-- AUDIT SESSION & SIGN-OFF TABLES
|
||||
-- ============================================================================
|
||||
|
||||
-- Table 11: compliance_audit_sessions
|
||||
CREATE TABLE IF NOT EXISTS compliance_audit_sessions (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
name VARCHAR(200) NOT NULL,
|
||||
description TEXT,
|
||||
auditor_name VARCHAR(100) NOT NULL,
|
||||
auditor_email VARCHAR(200),
|
||||
auditor_organization VARCHAR(200),
|
||||
status audit_session_status DEFAULT 'draft',
|
||||
regulation_ids JSONB,
|
||||
total_items INTEGER DEFAULT 0,
|
||||
completed_items INTEGER DEFAULT 0,
|
||||
compliant_count INTEGER DEFAULT 0,
|
||||
non_compliant_count INTEGER DEFAULT 0,
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
started_at TIMESTAMP,
|
||||
completed_at TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
|
||||
CREATE INDEX IF NOT EXISTS ix_audit_session_status ON compliance_audit_sessions(status);
|
||||
CREATE INDEX IF NOT EXISTS ix_audit_session_auditor ON compliance_audit_sessions(auditor_name);
|
||||
|
||||
-- Table 12: compliance_audit_signoffs
|
||||
CREATE TABLE IF NOT EXISTS compliance_audit_signoffs (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
session_id VARCHAR(36) NOT NULL REFERENCES compliance_audit_sessions(id),
|
||||
requirement_id VARCHAR(36) NOT NULL REFERENCES compliance_requirements(id),
|
||||
result audit_result DEFAULT 'pending',
|
||||
notes TEXT,
|
||||
evidence_ids JSONB,
|
||||
signature_hash VARCHAR(64),
|
||||
signed_at TIMESTAMP,
|
||||
signed_by VARCHAR(100),
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
|
||||
CREATE UNIQUE INDEX IF NOT EXISTS ix_signoff_session_requirement ON compliance_audit_signoffs(session_id, requirement_id);
|
||||
CREATE INDEX IF NOT EXISTS ix_signoff_result ON compliance_audit_signoffs(result);
|
||||
|
||||
|
||||
-- ============================================================================
|
||||
-- ISO 27001 ISMS TABLES
|
||||
-- ============================================================================
|
||||
|
||||
-- Table 13: compliance_isms_scope
|
||||
CREATE TABLE IF NOT EXISTS compliance_isms_scope (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
version VARCHAR(20) NOT NULL DEFAULT '1.0',
|
||||
scope_statement TEXT NOT NULL,
|
||||
included_locations JSONB,
|
||||
included_processes JSONB,
|
||||
included_services JSONB,
|
||||
excluded_items JSONB,
|
||||
exclusion_justification TEXT,
|
||||
organizational_boundary TEXT,
|
||||
physical_boundary TEXT,
|
||||
technical_boundary TEXT,
|
||||
status approval_status DEFAULT 'draft',
|
||||
approved_by VARCHAR(100),
|
||||
approved_at TIMESTAMP,
|
||||
approval_signature VARCHAR(64),
|
||||
effective_date DATE,
|
||||
review_date DATE,
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
created_by VARCHAR(100),
|
||||
updated_by VARCHAR(100)
|
||||
);
|
||||
|
||||
CREATE INDEX IF NOT EXISTS ix_isms_scope_status ON compliance_isms_scope(status);
|
||||
|
||||
-- Table 14: compliance_isms_context
|
||||
CREATE TABLE IF NOT EXISTS compliance_isms_context (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
version VARCHAR(20) NOT NULL DEFAULT '1.0',
|
||||
internal_issues JSONB,
|
||||
external_issues JSONB,
|
||||
interested_parties JSONB,
|
||||
regulatory_requirements JSONB,
|
||||
contractual_requirements JSONB,
|
||||
swot_strengths JSONB,
|
||||
swot_weaknesses JSONB,
|
||||
swot_opportunities JSONB,
|
||||
swot_threats JSONB,
|
||||
status approval_status DEFAULT 'draft',
|
||||
approved_by VARCHAR(100),
|
||||
approved_at TIMESTAMP,
|
||||
last_reviewed_at TIMESTAMP,
|
||||
next_review_date DATE,
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
|
||||
-- Table 15: compliance_isms_policies
|
||||
CREATE TABLE IF NOT EXISTS compliance_isms_policies (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
policy_id VARCHAR(30) UNIQUE NOT NULL,
|
||||
title VARCHAR(200) NOT NULL,
|
||||
policy_type VARCHAR(50) NOT NULL,
|
||||
description TEXT,
|
||||
policy_text TEXT NOT NULL,
|
||||
applies_to JSONB,
|
||||
version VARCHAR(20) NOT NULL DEFAULT '1.0',
|
||||
status approval_status DEFAULT 'draft',
|
||||
authored_by VARCHAR(100),
|
||||
reviewed_by VARCHAR(100),
|
||||
approved_by VARCHAR(100),
|
||||
approved_at TIMESTAMP,
|
||||
approval_signature VARCHAR(64),
|
||||
effective_date DATE,
|
||||
review_frequency_months INTEGER DEFAULT 12,
|
||||
next_review_date DATE,
|
||||
parent_policy_id VARCHAR(36) REFERENCES compliance_isms_policies(id),
|
||||
related_controls JSONB,
|
||||
document_path VARCHAR(500),
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
|
||||
CREATE INDEX IF NOT EXISTS ix_policy_id ON compliance_isms_policies(policy_id);
|
||||
CREATE INDEX IF NOT EXISTS ix_policy_type_status ON compliance_isms_policies(policy_type, status);
|
||||
|
||||
-- Table 16: compliance_security_objectives
|
||||
CREATE TABLE IF NOT EXISTS compliance_security_objectives (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
objective_id VARCHAR(30) UNIQUE NOT NULL,
|
||||
title VARCHAR(200) NOT NULL,
|
||||
description TEXT,
|
||||
category VARCHAR(50),
|
||||
specific TEXT,
|
||||
measurable TEXT,
|
||||
achievable TEXT,
|
||||
relevant TEXT,
|
||||
time_bound TEXT,
|
||||
kpi_name VARCHAR(100),
|
||||
kpi_target VARCHAR(100),
|
||||
kpi_current VARCHAR(100),
|
||||
kpi_unit VARCHAR(50),
|
||||
measurement_frequency VARCHAR(50),
|
||||
owner VARCHAR(100),
|
||||
accountable VARCHAR(100),
|
||||
status VARCHAR(30) DEFAULT 'active',
|
||||
progress_percentage INTEGER DEFAULT 0,
|
||||
target_date DATE,
|
||||
achieved_date DATE,
|
||||
related_controls JSONB,
|
||||
related_risks JSONB,
|
||||
approved_by VARCHAR(100),
|
||||
approved_at TIMESTAMP,
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
|
||||
CREATE INDEX IF NOT EXISTS ix_objective_id ON compliance_security_objectives(objective_id);
|
||||
CREATE INDEX IF NOT EXISTS ix_objective_status ON compliance_security_objectives(status);
|
||||
CREATE INDEX IF NOT EXISTS ix_objective_category ON compliance_security_objectives(category);
|
||||
|
||||
-- Table 17: compliance_soa (Statement of Applicability)
|
||||
CREATE TABLE IF NOT EXISTS compliance_soa (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
annex_a_control VARCHAR(20) NOT NULL,
|
||||
annex_a_title VARCHAR(300) NOT NULL,
|
||||
annex_a_category VARCHAR(100),
|
||||
is_applicable BOOLEAN NOT NULL,
|
||||
applicability_justification TEXT NOT NULL,
|
||||
implementation_status VARCHAR(30) DEFAULT 'planned',
|
||||
implementation_notes TEXT,
|
||||
breakpilot_control_ids JSONB,
|
||||
coverage_level VARCHAR(20) DEFAULT 'full',
|
||||
evidence_description TEXT,
|
||||
evidence_ids JSONB,
|
||||
risk_assessment_notes TEXT,
|
||||
compensating_controls TEXT,
|
||||
reviewed_by VARCHAR(100),
|
||||
reviewed_at TIMESTAMP,
|
||||
approved_by VARCHAR(100),
|
||||
approved_at TIMESTAMP,
|
||||
version VARCHAR(20) DEFAULT '1.0',
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
|
||||
CREATE UNIQUE INDEX IF NOT EXISTS ix_soa_annex_control ON compliance_soa(annex_a_control);
|
||||
CREATE INDEX IF NOT EXISTS ix_soa_applicable ON compliance_soa(is_applicable);
|
||||
CREATE INDEX IF NOT EXISTS ix_soa_status ON compliance_soa(implementation_status);
|
||||
|
||||
-- Table 18: compliance_internal_audits (MUST be before audit_findings due to FK)
|
||||
CREATE TABLE IF NOT EXISTS compliance_internal_audits (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
audit_id VARCHAR(30) UNIQUE NOT NULL,
|
||||
title VARCHAR(200) NOT NULL,
|
||||
audit_type VARCHAR(50) NOT NULL,
|
||||
scope_description TEXT NOT NULL,
|
||||
iso_chapters_covered JSONB,
|
||||
annex_a_controls_covered JSONB,
|
||||
processes_covered JSONB,
|
||||
departments_covered JSONB,
|
||||
criteria TEXT,
|
||||
planned_date DATE NOT NULL,
|
||||
actual_start_date DATE,
|
||||
actual_end_date DATE,
|
||||
lead_auditor VARCHAR(100) NOT NULL,
|
||||
audit_team JSONB,
|
||||
auditee_representatives JSONB,
|
||||
status VARCHAR(30) DEFAULT 'planned',
|
||||
total_findings INTEGER DEFAULT 0,
|
||||
major_findings INTEGER DEFAULT 0,
|
||||
minor_findings INTEGER DEFAULT 0,
|
||||
ofi_count INTEGER DEFAULT 0,
|
||||
positive_observations INTEGER DEFAULT 0,
|
||||
audit_conclusion TEXT,
|
||||
overall_assessment VARCHAR(30),
|
||||
report_date DATE,
|
||||
report_document_path VARCHAR(500),
|
||||
report_approved_by VARCHAR(100),
|
||||
report_approved_at TIMESTAMP,
|
||||
follow_up_audit_required BOOLEAN DEFAULT FALSE,
|
||||
follow_up_audit_id VARCHAR(36),
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
|
||||
CREATE INDEX IF NOT EXISTS ix_internal_audit_id ON compliance_internal_audits(audit_id);
|
||||
CREATE INDEX IF NOT EXISTS ix_internal_audit_date ON compliance_internal_audits(planned_date);
|
||||
CREATE INDEX IF NOT EXISTS ix_internal_audit_status ON compliance_internal_audits(status);
|
||||
|
||||
-- Table 19: compliance_audit_findings
|
||||
CREATE TABLE IF NOT EXISTS compliance_audit_findings (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
finding_id VARCHAR(30) UNIQUE NOT NULL,
|
||||
audit_session_id VARCHAR(36) REFERENCES compliance_audit_sessions(id),
|
||||
internal_audit_id VARCHAR(36) REFERENCES compliance_internal_audits(id),
|
||||
finding_type finding_type NOT NULL,
|
||||
iso_chapter VARCHAR(20),
|
||||
annex_a_control VARCHAR(20),
|
||||
title VARCHAR(300) NOT NULL,
|
||||
description TEXT NOT NULL,
|
||||
objective_evidence TEXT NOT NULL,
|
||||
root_cause TEXT,
|
||||
root_cause_method VARCHAR(50),
|
||||
impact_description TEXT,
|
||||
affected_processes JSONB,
|
||||
affected_assets JSONB,
|
||||
status finding_status DEFAULT 'open',
|
||||
owner VARCHAR(100),
|
||||
auditor VARCHAR(100),
|
||||
identified_date DATE NOT NULL DEFAULT CURRENT_DATE,
|
||||
due_date DATE,
|
||||
closed_date DATE,
|
||||
verification_method TEXT,
|
||||
verified_by VARCHAR(100),
|
||||
verified_at TIMESTAMP,
|
||||
verification_evidence TEXT,
|
||||
closure_notes TEXT,
|
||||
closed_by VARCHAR(100),
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
|
||||
CREATE INDEX IF NOT EXISTS ix_finding_id ON compliance_audit_findings(finding_id);
|
||||
CREATE INDEX IF NOT EXISTS ix_finding_type_status ON compliance_audit_findings(finding_type, status);
|
||||
CREATE INDEX IF NOT EXISTS ix_finding_due_date ON compliance_audit_findings(due_date);
|
||||
|
||||
-- Table 20: compliance_corrective_actions
|
||||
CREATE TABLE IF NOT EXISTS compliance_corrective_actions (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
capa_id VARCHAR(30) UNIQUE NOT NULL,
|
||||
finding_id VARCHAR(36) NOT NULL REFERENCES compliance_audit_findings(id),
|
||||
capa_type capa_type NOT NULL,
|
||||
title VARCHAR(300) NOT NULL,
|
||||
description TEXT NOT NULL,
|
||||
expected_outcome TEXT,
|
||||
assigned_to VARCHAR(100) NOT NULL,
|
||||
approved_by VARCHAR(100),
|
||||
planned_start DATE,
|
||||
planned_completion DATE NOT NULL,
|
||||
actual_completion DATE,
|
||||
status VARCHAR(30) DEFAULT 'planned',
|
||||
progress_percentage INTEGER DEFAULT 0,
|
||||
estimated_effort_hours INTEGER,
|
||||
actual_effort_hours INTEGER,
|
||||
resources_required TEXT,
|
||||
implementation_evidence TEXT,
|
||||
evidence_ids JSONB,
|
||||
effectiveness_criteria TEXT,
|
||||
effectiveness_verified BOOLEAN DEFAULT FALSE,
|
||||
effectiveness_verification_date DATE,
|
||||
effectiveness_notes TEXT,
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
|
||||
CREATE INDEX IF NOT EXISTS ix_capa_id ON compliance_corrective_actions(capa_id);
|
||||
CREATE INDEX IF NOT EXISTS ix_capa_status ON compliance_corrective_actions(status);
|
||||
CREATE INDEX IF NOT EXISTS ix_capa_due ON compliance_corrective_actions(planned_completion);
|
||||
|
||||
-- Table 21: compliance_management_reviews
|
||||
CREATE TABLE IF NOT EXISTS compliance_management_reviews (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
review_id VARCHAR(30) UNIQUE NOT NULL,
|
||||
title VARCHAR(200) NOT NULL,
|
||||
review_date DATE NOT NULL,
|
||||
review_period_start DATE,
|
||||
review_period_end DATE,
|
||||
chairperson VARCHAR(100) NOT NULL,
|
||||
attendees JSONB,
|
||||
input_previous_actions TEXT,
|
||||
input_isms_changes TEXT,
|
||||
input_security_performance TEXT,
|
||||
input_interested_party_feedback TEXT,
|
||||
input_risk_assessment_results TEXT,
|
||||
input_improvement_opportunities TEXT,
|
||||
input_policy_effectiveness TEXT,
|
||||
input_objective_achievement TEXT,
|
||||
input_resource_adequacy TEXT,
|
||||
output_improvement_decisions TEXT,
|
||||
output_isms_changes TEXT,
|
||||
output_resource_needs TEXT,
|
||||
action_items JSONB,
|
||||
isms_effectiveness_rating VARCHAR(20),
|
||||
key_decisions TEXT,
|
||||
status VARCHAR(30) DEFAULT 'draft',
|
||||
approved_by VARCHAR(100),
|
||||
approved_at TIMESTAMP,
|
||||
minutes_document_path VARCHAR(500),
|
||||
next_review_date DATE,
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
|
||||
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
|
||||
CREATE INDEX IF NOT EXISTS ix_mgmt_review_id ON compliance_management_reviews(review_id);
|
||||
CREATE INDEX IF NOT EXISTS ix_mgmt_review_date ON compliance_management_reviews(review_date);
|
||||
CREATE INDEX IF NOT EXISTS ix_mgmt_review_status ON compliance_management_reviews(status);
|
||||
|
||||
-- Table 22: compliance_audit_trail
|
||||
CREATE TABLE IF NOT EXISTS compliance_audit_trail (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
entity_type VARCHAR(50) NOT NULL,
|
||||
entity_id VARCHAR(36) NOT NULL,
|
||||
entity_name VARCHAR(200),
|
||||
action VARCHAR(20) NOT NULL,
|
||||
field_changed VARCHAR(100),
|
||||
old_value TEXT,
|
||||
new_value TEXT,
|
||||
change_summary TEXT,
|
||||
performed_by VARCHAR(100) NOT NULL,
|
||||
performed_at TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,
|
||||
ip_address VARCHAR(45),
|
||||
user_agent VARCHAR(500),
|
||||
session_id VARCHAR(100),
|
||||
checksum VARCHAR(64),
|
||||
created_at TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
|
||||
CREATE INDEX IF NOT EXISTS ix_audit_trail_entity ON compliance_audit_trail(entity_type, entity_id);
|
||||
CREATE INDEX IF NOT EXISTS ix_audit_trail_time ON compliance_audit_trail(performed_at);
|
||||
CREATE INDEX IF NOT EXISTS ix_audit_trail_user ON compliance_audit_trail(performed_by);
|
||||
|
||||
-- Table 23: compliance_isms_readiness
|
||||
CREATE TABLE IF NOT EXISTS compliance_isms_readiness (
|
||||
id VARCHAR(36) PRIMARY KEY,
|
||||
check_date TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,
|
||||
triggered_by VARCHAR(100),
|
||||
overall_status VARCHAR(20) NOT NULL,
|
||||
certification_possible BOOLEAN NOT NULL,
|
||||
chapter_4_status VARCHAR(20),
|
||||
chapter_5_status VARCHAR(20),
|
||||
chapter_6_status VARCHAR(20),
|
||||
chapter_7_status VARCHAR(20),
|
||||
chapter_8_status VARCHAR(20),
|
||||
chapter_9_status VARCHAR(20),
|
||||
chapter_10_status VARCHAR(20),
|
||||
potential_majors JSONB,
|
||||
potential_minors JSONB,
|
||||
improvement_opportunities JSONB,
|
||||
readiness_score FLOAT,
|
||||
documentation_score FLOAT,
|
||||
implementation_score FLOAT,
|
||||
evidence_score FLOAT,
|
||||
priority_actions JSONB,
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
|
||||
CREATE INDEX IF NOT EXISTS ix_readiness_date ON compliance_isms_readiness(check_date);
|
||||
CREATE INDEX IF NOT EXISTS ix_readiness_status ON compliance_isms_readiness(overall_status);
|
||||
|
||||
|
||||
-- ============================================================================
|
||||
-- UPDATE TIMESTAMPS TRIGGER
|
||||
-- ============================================================================
|
||||
|
||||
CREATE OR REPLACE FUNCTION update_compliance_timestamp()
|
||||
RETURNS TRIGGER AS $$
|
||||
BEGIN
|
||||
NEW.updated_at = CURRENT_TIMESTAMP;
|
||||
RETURN NEW;
|
||||
END;
|
||||
$$ LANGUAGE plpgsql;
|
||||
|
||||
-- Apply trigger to all compliance tables with updated_at
|
||||
DO $$
|
||||
DECLARE
|
||||
t TEXT;
|
||||
BEGIN
|
||||
FOR t IN
|
||||
SELECT table_name
|
||||
FROM information_schema.columns
|
||||
WHERE table_schema = 'public'
|
||||
AND column_name = 'updated_at'
|
||||
AND table_name LIKE 'compliance_%'
|
||||
LOOP
|
||||
EXECUTE format('
|
||||
DROP TRIGGER IF EXISTS trigger_%I_updated_at ON %I;
|
||||
CREATE TRIGGER trigger_%I_updated_at
|
||||
BEFORE UPDATE ON %I
|
||||
FOR EACH ROW EXECUTE FUNCTION update_compliance_timestamp();
|
||||
', t, t, t, t);
|
||||
END LOOP;
|
||||
END $$;
|
||||
|
||||
|
||||
-- ============================================================================
|
||||
-- CLEANUP FUNCTIONS
|
||||
-- ============================================================================
|
||||
|
||||
-- Function to cleanup expired evidence
|
||||
CREATE OR REPLACE FUNCTION cleanup_expired_compliance_evidence()
|
||||
RETURNS INTEGER AS $$
|
||||
DECLARE
|
||||
deleted_count INTEGER;
|
||||
BEGIN
|
||||
UPDATE compliance_evidence
|
||||
SET status = 'expired'
|
||||
WHERE valid_until < CURRENT_TIMESTAMP
|
||||
AND status = 'valid';
|
||||
|
||||
GET DIAGNOSTICS deleted_count = ROW_COUNT;
|
||||
RETURN deleted_count;
|
||||
END;
|
||||
$$ LANGUAGE plpgsql;
|
||||
|
||||
|
||||
-- ============================================================================
|
||||
-- GRANT PERMISSIONS
|
||||
-- ============================================================================
|
||||
|
||||
-- Grant permissions to the application user (adjust username as needed)
|
||||
-- GRANT ALL PRIVILEGES ON ALL TABLES IN SCHEMA public TO breakpilot;
|
||||
-- GRANT ALL PRIVILEGES ON ALL SEQUENCES IN SCHEMA public TO breakpilot;
|
||||
|
||||
|
||||
-- ============================================================================
|
||||
-- MIGRATION COMPLETE
|
||||
-- ============================================================================
|
||||
|
||||
SELECT 'Compliance migration completed. ' ||
|
||||
(SELECT COUNT(*) FROM information_schema.tables
|
||||
WHERE table_schema = 'public' AND table_name LIKE 'compliance_%') ||
|
||||
' tables created.' AS status;
|
||||
241
docs-src/backend/migrations/add_game_tables.sql
Normal file
241
docs-src/backend/migrations/add_game_tables.sql
Normal file
@@ -0,0 +1,241 @@
|
||||
-- ==============================================
|
||||
-- Breakpilot Drive - Game Tables Migration
|
||||
-- ==============================================
|
||||
-- Run this migration to add game-related tables to PostgreSQL.
|
||||
--
|
||||
-- Execute with:
|
||||
-- psql -h localhost -U breakpilot -d breakpilot -f add_game_tables.sql
|
||||
|
||||
-- Enable UUID extension if not already enabled
|
||||
CREATE EXTENSION IF NOT EXISTS "uuid-ossp";
|
||||
|
||||
-- ==============================================
|
||||
-- Student Learning State
|
||||
-- ==============================================
|
||||
-- Tracks the learning progress of each student across subjects.
|
||||
-- This is the core table for adaptive difficulty.
|
||||
|
||||
CREATE TABLE IF NOT EXISTS student_learning_state (
|
||||
id UUID PRIMARY KEY DEFAULT uuid_generate_v4(),
|
||||
student_id UUID NOT NULL,
|
||||
overall_level INTEGER DEFAULT 3 CHECK (overall_level >= 1 AND overall_level <= 5),
|
||||
math_level DECIMAL(3,2) DEFAULT 3.0 CHECK (math_level >= 1.0 AND math_level <= 5.0),
|
||||
german_level DECIMAL(3,2) DEFAULT 3.0 CHECK (german_level >= 1.0 AND german_level <= 5.0),
|
||||
english_level DECIMAL(3,2) DEFAULT 3.0 CHECK (english_level >= 1.0 AND english_level <= 5.0),
|
||||
total_play_time_minutes INTEGER DEFAULT 0,
|
||||
total_sessions INTEGER DEFAULT 0,
|
||||
questions_answered INTEGER DEFAULT 0,
|
||||
questions_correct INTEGER DEFAULT 0,
|
||||
created_at TIMESTAMPTZ DEFAULT NOW(),
|
||||
updated_at TIMESTAMPTZ DEFAULT NOW(),
|
||||
CONSTRAINT unique_student_learning UNIQUE(student_id)
|
||||
);
|
||||
|
||||
-- Index for fast lookups by student_id
|
||||
CREATE INDEX IF NOT EXISTS idx_learning_state_student ON student_learning_state(student_id);
|
||||
|
||||
-- Comment on table
|
||||
COMMENT ON TABLE student_learning_state IS 'Tracks learning progress for Breakpilot Drive game';
|
||||
COMMENT ON COLUMN student_learning_state.overall_level IS 'Overall difficulty level 1-5 (1=Beginner/Grade 2-3, 5=Expert/Grade 6+)';
|
||||
COMMENT ON COLUMN student_learning_state.math_level IS 'Math subject proficiency level';
|
||||
COMMENT ON COLUMN student_learning_state.german_level IS 'German subject proficiency level';
|
||||
COMMENT ON COLUMN student_learning_state.english_level IS 'English subject proficiency level';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Game Sessions
|
||||
-- ==============================================
|
||||
-- Records each game session played by a student.
|
||||
|
||||
CREATE TABLE IF NOT EXISTS game_sessions (
|
||||
id UUID PRIMARY KEY DEFAULT uuid_generate_v4(),
|
||||
student_id UUID NOT NULL,
|
||||
game_mode VARCHAR(20) NOT NULL CHECK (game_mode IN ('video', 'audio')),
|
||||
duration_seconds INTEGER NOT NULL CHECK (duration_seconds >= 0),
|
||||
distance_traveled DECIMAL(10,2),
|
||||
score INTEGER NOT NULL DEFAULT 0,
|
||||
questions_answered INTEGER DEFAULT 0,
|
||||
questions_correct INTEGER DEFAULT 0,
|
||||
difficulty_level INTEGER NOT NULL CHECK (difficulty_level >= 1 AND difficulty_level <= 5),
|
||||
started_at TIMESTAMPTZ NOT NULL,
|
||||
ended_at TIMESTAMPTZ DEFAULT NOW(),
|
||||
metadata JSONB,
|
||||
created_at TIMESTAMPTZ DEFAULT NOW()
|
||||
);
|
||||
|
||||
-- Indexes for common queries
|
||||
CREATE INDEX IF NOT EXISTS idx_game_sessions_student ON game_sessions(student_id);
|
||||
CREATE INDEX IF NOT EXISTS idx_game_sessions_date ON game_sessions(ended_at);
|
||||
CREATE INDEX IF NOT EXISTS idx_game_sessions_score ON game_sessions(score DESC);
|
||||
|
||||
-- Comment on table
|
||||
COMMENT ON TABLE game_sessions IS 'Records individual game sessions for Breakpilot Drive';
|
||||
COMMENT ON COLUMN game_sessions.game_mode IS 'Game mode: video (visual) or audio (voice-guided)';
|
||||
COMMENT ON COLUMN game_sessions.distance_traveled IS 'Distance traveled in game units';
|
||||
COMMENT ON COLUMN game_sessions.metadata IS 'Additional session data in JSON format';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Game Quiz Answers
|
||||
-- ==============================================
|
||||
-- Tracks individual quiz answers for detailed analytics.
|
||||
|
||||
CREATE TABLE IF NOT EXISTS game_quiz_answers (
|
||||
id UUID PRIMARY KEY DEFAULT uuid_generate_v4(),
|
||||
session_id UUID REFERENCES game_sessions(id) ON DELETE CASCADE,
|
||||
question_id VARCHAR(100) NOT NULL,
|
||||
subject VARCHAR(50) NOT NULL CHECK (subject IN ('math', 'german', 'english', 'general')),
|
||||
difficulty INTEGER NOT NULL CHECK (difficulty >= 1 AND difficulty <= 5),
|
||||
is_correct BOOLEAN NOT NULL,
|
||||
answer_time_ms INTEGER CHECK (answer_time_ms >= 0),
|
||||
created_at TIMESTAMPTZ DEFAULT NOW()
|
||||
);
|
||||
|
||||
-- Indexes for analytics queries
|
||||
CREATE INDEX IF NOT EXISTS idx_quiz_answers_session ON game_quiz_answers(session_id);
|
||||
CREATE INDEX IF NOT EXISTS idx_quiz_answers_subject ON game_quiz_answers(subject);
|
||||
CREATE INDEX IF NOT EXISTS idx_quiz_answers_correct ON game_quiz_answers(is_correct);
|
||||
|
||||
-- Comment on table
|
||||
COMMENT ON TABLE game_quiz_answers IS 'Individual quiz answer records for learning analytics';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Trigger: Update updated_at timestamp
|
||||
-- ==============================================
|
||||
CREATE OR REPLACE FUNCTION update_updated_at_column()
|
||||
RETURNS TRIGGER AS $$
|
||||
BEGIN
|
||||
NEW.updated_at = NOW();
|
||||
RETURN NEW;
|
||||
END;
|
||||
$$ language 'plpgsql';
|
||||
|
||||
-- Apply trigger to student_learning_state
|
||||
DROP TRIGGER IF EXISTS update_student_learning_state_updated_at ON student_learning_state;
|
||||
CREATE TRIGGER update_student_learning_state_updated_at
|
||||
BEFORE UPDATE ON student_learning_state
|
||||
FOR EACH ROW
|
||||
EXECUTE FUNCTION update_updated_at_column();
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- View: Student Summary Statistics
|
||||
-- ==============================================
|
||||
CREATE OR REPLACE VIEW game_student_summary AS
|
||||
SELECT
|
||||
sls.student_id,
|
||||
sls.overall_level,
|
||||
sls.math_level,
|
||||
sls.german_level,
|
||||
sls.english_level,
|
||||
sls.total_play_time_minutes,
|
||||
sls.total_sessions,
|
||||
sls.questions_answered,
|
||||
sls.questions_correct,
|
||||
CASE WHEN sls.questions_answered > 0
|
||||
THEN ROUND((sls.questions_correct::DECIMAL / sls.questions_answered) * 100, 1)
|
||||
ELSE 0 END as accuracy_percent,
|
||||
COALESCE(recent.recent_score, 0) as recent_score,
|
||||
COALESCE(recent.recent_sessions, 0) as sessions_last_7_days
|
||||
FROM student_learning_state sls
|
||||
LEFT JOIN (
|
||||
SELECT
|
||||
student_id,
|
||||
SUM(score) as recent_score,
|
||||
COUNT(*) as recent_sessions
|
||||
FROM game_sessions
|
||||
WHERE ended_at > NOW() - INTERVAL '7 days'
|
||||
GROUP BY student_id
|
||||
) recent ON sls.student_id = recent.student_id;
|
||||
|
||||
COMMENT ON VIEW game_student_summary IS 'Summary statistics for each student including recent activity';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- View: Daily Leaderboard
|
||||
-- ==============================================
|
||||
CREATE OR REPLACE VIEW game_daily_leaderboard AS
|
||||
SELECT
|
||||
student_id,
|
||||
SUM(score) as total_score,
|
||||
COUNT(*) as session_count,
|
||||
SUM(questions_correct) as total_correct,
|
||||
SUM(questions_answered) as total_questions,
|
||||
RANK() OVER (ORDER BY SUM(score) DESC) as rank
|
||||
FROM game_sessions
|
||||
WHERE ended_at > NOW() - INTERVAL '1 day'
|
||||
GROUP BY student_id
|
||||
ORDER BY total_score DESC;
|
||||
|
||||
COMMENT ON VIEW game_daily_leaderboard IS 'Daily leaderboard for Breakpilot Drive';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Function: Calculate Level Adjustment
|
||||
-- ==============================================
|
||||
-- Returns the recommended level adjustment based on recent performance.
|
||||
-- Returns: -1 (decrease), 0 (keep), 1 (increase)
|
||||
|
||||
CREATE OR REPLACE FUNCTION calculate_level_adjustment(p_student_id UUID)
|
||||
RETURNS INTEGER AS $$
|
||||
DECLARE
|
||||
v_recent_accuracy DECIMAL;
|
||||
v_recent_questions INTEGER;
|
||||
BEGIN
|
||||
-- Get accuracy from last 10 questions
|
||||
SELECT
|
||||
CASE WHEN COUNT(*) > 0
|
||||
THEN SUM(CASE WHEN is_correct THEN 1 ELSE 0 END)::DECIMAL / COUNT(*)
|
||||
ELSE 0 END,
|
||||
COUNT(*)
|
||||
INTO v_recent_accuracy, v_recent_questions
|
||||
FROM (
|
||||
SELECT is_correct
|
||||
FROM game_quiz_answers qa
|
||||
JOIN game_sessions gs ON qa.session_id = gs.id
|
||||
WHERE gs.student_id = p_student_id
|
||||
ORDER BY qa.created_at DESC
|
||||
LIMIT 10
|
||||
) recent;
|
||||
|
||||
-- Need at least 5 questions for adjustment
|
||||
IF v_recent_questions < 5 THEN
|
||||
RETURN 0;
|
||||
END IF;
|
||||
|
||||
-- High accuracy (>=80%) -> increase level
|
||||
IF v_recent_accuracy >= 0.8 THEN
|
||||
RETURN 1;
|
||||
END IF;
|
||||
|
||||
-- Low accuracy (<40%) -> decrease level
|
||||
IF v_recent_accuracy < 0.4 THEN
|
||||
RETURN -1;
|
||||
END IF;
|
||||
|
||||
-- Keep current level
|
||||
RETURN 0;
|
||||
END;
|
||||
$$ LANGUAGE plpgsql;
|
||||
|
||||
COMMENT ON FUNCTION calculate_level_adjustment IS 'Calculates recommended difficulty adjustment based on recent performance';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Grant Permissions (adjust user as needed)
|
||||
-- ==============================================
|
||||
-- GRANT ALL PRIVILEGES ON ALL TABLES IN SCHEMA public TO breakpilot;
|
||||
-- GRANT ALL PRIVILEGES ON ALL SEQUENCES IN SCHEMA public TO breakpilot;
|
||||
-- GRANT EXECUTE ON ALL FUNCTIONS IN SCHEMA public TO breakpilot;
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Migration Complete Message
|
||||
-- ==============================================
|
||||
DO $$
|
||||
BEGIN
|
||||
RAISE NOTICE 'Breakpilot Drive game tables created successfully!';
|
||||
RAISE NOTICE 'Tables: student_learning_state, game_sessions, game_quiz_answers';
|
||||
RAISE NOTICE 'Views: game_student_summary, game_daily_leaderboard';
|
||||
END $$;
|
||||
@@ -0,0 +1,409 @@
|
||||
-- ==============================================
|
||||
-- Jitsi Recordings & Transcription Tables Migration
|
||||
-- ==============================================
|
||||
-- Run this migration to add recording and transcription tables.
|
||||
--
|
||||
-- Execute with:
|
||||
-- psql -h localhost -U breakpilot -d breakpilot_db -f add_recording_transcription_tables.sql
|
||||
|
||||
-- Enable UUID extension if not already enabled
|
||||
CREATE EXTENSION IF NOT EXISTS "uuid-ossp";
|
||||
|
||||
-- ==============================================
|
||||
-- Meeting Recording Consents (DSGVO)
|
||||
-- ==============================================
|
||||
-- Tracks consent for meeting recordings.
|
||||
-- All participants must consent before recording starts.
|
||||
|
||||
CREATE TABLE IF NOT EXISTS meeting_recording_consents (
|
||||
id UUID PRIMARY KEY DEFAULT uuid_generate_v4(),
|
||||
meeting_id VARCHAR(255) NOT NULL,
|
||||
user_id UUID,
|
||||
consent_type VARCHAR(50) NOT NULL CHECK (consent_type IN ('opt_in', 'announced', 'implicit')),
|
||||
all_participants_consented BOOLEAN DEFAULT FALSE,
|
||||
participant_count INTEGER DEFAULT 0,
|
||||
consented_count INTEGER DEFAULT 0,
|
||||
consented_at TIMESTAMPTZ,
|
||||
withdrawn_at TIMESTAMPTZ,
|
||||
created_at TIMESTAMPTZ DEFAULT NOW(),
|
||||
updated_at TIMESTAMPTZ DEFAULT NOW()
|
||||
);
|
||||
|
||||
-- Indexes
|
||||
CREATE INDEX IF NOT EXISTS idx_recording_consents_meeting ON meeting_recording_consents(meeting_id);
|
||||
CREATE INDEX IF NOT EXISTS idx_recording_consents_user ON meeting_recording_consents(user_id);
|
||||
|
||||
-- Comments
|
||||
COMMENT ON TABLE meeting_recording_consents IS 'DSGVO-compliant consent tracking for meeting recordings';
|
||||
COMMENT ON COLUMN meeting_recording_consents.consent_type IS 'Type: opt_in (explicit), announced (verbally announced), implicit (policy-based)';
|
||||
COMMENT ON COLUMN meeting_recording_consents.withdrawn_at IS 'Set when consent is withdrawn (soft delete)';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Recordings
|
||||
-- ==============================================
|
||||
-- Stores metadata for recorded meetings.
|
||||
|
||||
CREATE TABLE IF NOT EXISTS recordings (
|
||||
id UUID PRIMARY KEY DEFAULT uuid_generate_v4(),
|
||||
meeting_id VARCHAR(255) NOT NULL,
|
||||
jibri_session_id VARCHAR(255),
|
||||
title VARCHAR(500),
|
||||
storage_path VARCHAR(1000) NOT NULL,
|
||||
audio_path VARCHAR(1000),
|
||||
file_size_bytes BIGINT,
|
||||
duration_seconds INTEGER,
|
||||
participant_count INTEGER DEFAULT 0,
|
||||
status VARCHAR(50) NOT NULL DEFAULT 'uploaded' CHECK (status IN ('uploaded', 'processing', 'ready', 'failed', 'deleted')),
|
||||
created_by UUID,
|
||||
recorded_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
|
||||
retention_days INTEGER DEFAULT 365,
|
||||
deleted_at TIMESTAMPTZ,
|
||||
created_at TIMESTAMPTZ DEFAULT NOW(),
|
||||
updated_at TIMESTAMPTZ DEFAULT NOW()
|
||||
);
|
||||
|
||||
-- Indexes
|
||||
CREATE INDEX IF NOT EXISTS idx_recordings_meeting ON recordings(meeting_id);
|
||||
CREATE INDEX IF NOT EXISTS idx_recordings_status ON recordings(status);
|
||||
CREATE INDEX IF NOT EXISTS idx_recordings_created_by ON recordings(created_by);
|
||||
CREATE INDEX IF NOT EXISTS idx_recordings_recorded_at ON recordings(recorded_at);
|
||||
|
||||
-- Comments
|
||||
COMMENT ON TABLE recordings IS 'Jitsi meeting recordings stored in MinIO';
|
||||
COMMENT ON COLUMN recordings.storage_path IS 'Path in MinIO bucket: recordings/{recording_name}/video.mp4';
|
||||
COMMENT ON COLUMN recordings.audio_path IS 'Extracted audio for transcription: recordings/{recording_name}/audio.wav';
|
||||
COMMENT ON COLUMN recordings.retention_days IS 'Days until automatic deletion (DSGVO compliance)';
|
||||
COMMENT ON COLUMN recordings.deleted_at IS 'Soft delete timestamp for DSGVO audit trail';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Transcriptions
|
||||
-- ==============================================
|
||||
-- Stores transcription metadata and full text.
|
||||
|
||||
CREATE TABLE IF NOT EXISTS transcriptions (
|
||||
id UUID PRIMARY KEY DEFAULT uuid_generate_v4(),
|
||||
recording_id UUID NOT NULL REFERENCES recordings(id) ON DELETE CASCADE,
|
||||
language VARCHAR(10) NOT NULL DEFAULT 'de',
|
||||
model VARCHAR(100) NOT NULL DEFAULT 'large-v3',
|
||||
status VARCHAR(50) NOT NULL DEFAULT 'pending' CHECK (status IN ('pending', 'queued', 'processing', 'completed', 'failed')),
|
||||
full_text TEXT,
|
||||
word_count INTEGER DEFAULT 0,
|
||||
confidence_score FLOAT,
|
||||
vtt_path VARCHAR(1000),
|
||||
srt_path VARCHAR(1000),
|
||||
json_path VARCHAR(1000),
|
||||
error_message TEXT,
|
||||
processing_started_at TIMESTAMPTZ,
|
||||
processing_completed_at TIMESTAMPTZ,
|
||||
processing_duration_seconds INTEGER,
|
||||
created_at TIMESTAMPTZ DEFAULT NOW(),
|
||||
updated_at TIMESTAMPTZ DEFAULT NOW()
|
||||
);
|
||||
|
||||
-- Indexes
|
||||
CREATE INDEX IF NOT EXISTS idx_transcriptions_recording ON transcriptions(recording_id);
|
||||
CREATE INDEX IF NOT EXISTS idx_transcriptions_status ON transcriptions(status);
|
||||
CREATE INDEX IF NOT EXISTS idx_transcriptions_language ON transcriptions(language);
|
||||
|
||||
-- Full-text search index for transcription content
|
||||
CREATE INDEX IF NOT EXISTS idx_transcriptions_fulltext ON transcriptions USING gin(to_tsvector('german', COALESCE(full_text, '')));
|
||||
|
||||
-- Comments
|
||||
COMMENT ON TABLE transcriptions IS 'Whisper transcriptions with speaker diarization';
|
||||
COMMENT ON COLUMN transcriptions.model IS 'Whisper model used: tiny, base, small, medium, large-v3';
|
||||
COMMENT ON COLUMN transcriptions.vtt_path IS 'WebVTT subtitle file path in MinIO';
|
||||
COMMENT ON COLUMN transcriptions.srt_path IS 'SRT subtitle file path in MinIO';
|
||||
COMMENT ON COLUMN transcriptions.json_path IS 'Full JSON with segments and speakers in MinIO';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Transcription Segments
|
||||
-- ==============================================
|
||||
-- Individual speech segments with speaker identification.
|
||||
|
||||
CREATE TABLE IF NOT EXISTS transcription_segments (
|
||||
id UUID PRIMARY KEY DEFAULT uuid_generate_v4(),
|
||||
transcription_id UUID NOT NULL REFERENCES transcriptions(id) ON DELETE CASCADE,
|
||||
segment_index INTEGER NOT NULL,
|
||||
start_time_ms INTEGER NOT NULL,
|
||||
end_time_ms INTEGER NOT NULL,
|
||||
text TEXT NOT NULL,
|
||||
speaker_id VARCHAR(50),
|
||||
speaker_name VARCHAR(255),
|
||||
confidence FLOAT,
|
||||
created_at TIMESTAMPTZ DEFAULT NOW()
|
||||
);
|
||||
|
||||
-- Indexes
|
||||
CREATE INDEX IF NOT EXISTS idx_segments_transcription ON transcription_segments(transcription_id);
|
||||
CREATE INDEX IF NOT EXISTS idx_segments_speaker ON transcription_segments(speaker_id);
|
||||
CREATE INDEX IF NOT EXISTS idx_segments_time ON transcription_segments(start_time_ms, end_time_ms);
|
||||
|
||||
-- Full-text search on segments
|
||||
CREATE INDEX IF NOT EXISTS idx_segments_fulltext ON transcription_segments USING gin(to_tsvector('german', text));
|
||||
|
||||
-- Comments
|
||||
COMMENT ON TABLE transcription_segments IS 'Individual speech segments with timestamps and speaker IDs';
|
||||
COMMENT ON COLUMN transcription_segments.speaker_id IS 'pyannote speaker ID: SPEAKER_00, SPEAKER_01, etc.';
|
||||
COMMENT ON COLUMN transcription_segments.speaker_name IS 'Optionally mapped to actual participant name';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Recording Audit Log (DSGVO)
|
||||
-- ==============================================
|
||||
-- Tracks all access and modifications for compliance.
|
||||
|
||||
CREATE TABLE IF NOT EXISTS recording_audit_log (
|
||||
id UUID PRIMARY KEY DEFAULT uuid_generate_v4(),
|
||||
recording_id UUID,
|
||||
transcription_id UUID,
|
||||
user_id UUID,
|
||||
action VARCHAR(100) NOT NULL CHECK (action IN (
|
||||
'created', 'viewed', 'downloaded', 'shared',
|
||||
'transcription_started', 'transcription_completed',
|
||||
'deleted', 'retention_expired', 'consent_withdrawn'
|
||||
)),
|
||||
ip_address INET,
|
||||
user_agent TEXT,
|
||||
metadata JSONB,
|
||||
created_at TIMESTAMPTZ DEFAULT NOW()
|
||||
);
|
||||
|
||||
-- Indexes
|
||||
CREATE INDEX IF NOT EXISTS idx_audit_recording ON recording_audit_log(recording_id);
|
||||
CREATE INDEX IF NOT EXISTS idx_audit_transcription ON recording_audit_log(transcription_id);
|
||||
CREATE INDEX IF NOT EXISTS idx_audit_user ON recording_audit_log(user_id);
|
||||
CREATE INDEX IF NOT EXISTS idx_audit_action ON recording_audit_log(action);
|
||||
CREATE INDEX IF NOT EXISTS idx_audit_created ON recording_audit_log(created_at);
|
||||
|
||||
-- Comments
|
||||
COMMENT ON TABLE recording_audit_log IS 'DSGVO audit trail for all recording access';
|
||||
COMMENT ON COLUMN recording_audit_log.action IS 'Type of action performed';
|
||||
COMMENT ON COLUMN recording_audit_log.metadata IS 'Additional context (e.g., reason for deletion)';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Transcription Queue (RQ Job Tracking)
|
||||
-- ==============================================
|
||||
-- Tracks pending and completed transcription jobs.
|
||||
|
||||
CREATE TABLE IF NOT EXISTS transcription_queue (
|
||||
id UUID PRIMARY KEY DEFAULT uuid_generate_v4(),
|
||||
transcription_id UUID NOT NULL REFERENCES transcriptions(id) ON DELETE CASCADE,
|
||||
job_id VARCHAR(255),
|
||||
priority INTEGER DEFAULT 0,
|
||||
status VARCHAR(50) NOT NULL DEFAULT 'pending' CHECK (status IN ('pending', 'queued', 'processing', 'completed', 'failed', 'cancelled')),
|
||||
worker_id VARCHAR(255),
|
||||
attempts INTEGER DEFAULT 0,
|
||||
max_attempts INTEGER DEFAULT 3,
|
||||
error_message TEXT,
|
||||
queued_at TIMESTAMPTZ,
|
||||
started_at TIMESTAMPTZ,
|
||||
completed_at TIMESTAMPTZ,
|
||||
created_at TIMESTAMPTZ DEFAULT NOW(),
|
||||
updated_at TIMESTAMPTZ DEFAULT NOW()
|
||||
);
|
||||
|
||||
-- Indexes
|
||||
CREATE INDEX IF NOT EXISTS idx_queue_status ON transcription_queue(status);
|
||||
CREATE INDEX IF NOT EXISTS idx_queue_priority ON transcription_queue(priority DESC, created_at ASC);
|
||||
CREATE INDEX IF NOT EXISTS idx_queue_job ON transcription_queue(job_id);
|
||||
|
||||
-- Comments
|
||||
COMMENT ON TABLE transcription_queue IS 'RQ job queue tracking for transcription workers';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Trigger: Update updated_at timestamp
|
||||
-- ==============================================
|
||||
CREATE OR REPLACE FUNCTION update_recording_updated_at()
|
||||
RETURNS TRIGGER AS $$
|
||||
BEGIN
|
||||
NEW.updated_at = NOW();
|
||||
RETURN NEW;
|
||||
END;
|
||||
$$ LANGUAGE 'plpgsql';
|
||||
|
||||
-- Apply triggers
|
||||
DROP TRIGGER IF EXISTS update_recordings_updated_at ON recordings;
|
||||
CREATE TRIGGER update_recordings_updated_at
|
||||
BEFORE UPDATE ON recordings
|
||||
FOR EACH ROW
|
||||
EXECUTE FUNCTION update_recording_updated_at();
|
||||
|
||||
DROP TRIGGER IF EXISTS update_transcriptions_updated_at ON transcriptions;
|
||||
CREATE TRIGGER update_transcriptions_updated_at
|
||||
BEFORE UPDATE ON transcriptions
|
||||
FOR EACH ROW
|
||||
EXECUTE FUNCTION update_recording_updated_at();
|
||||
|
||||
DROP TRIGGER IF EXISTS update_consents_updated_at ON meeting_recording_consents;
|
||||
CREATE TRIGGER update_consents_updated_at
|
||||
BEFORE UPDATE ON meeting_recording_consents
|
||||
FOR EACH ROW
|
||||
EXECUTE FUNCTION update_recording_updated_at();
|
||||
|
||||
DROP TRIGGER IF EXISTS update_queue_updated_at ON transcription_queue;
|
||||
CREATE TRIGGER update_queue_updated_at
|
||||
BEFORE UPDATE ON transcription_queue
|
||||
FOR EACH ROW
|
||||
EXECUTE FUNCTION update_recording_updated_at();
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- View: Recording Overview
|
||||
-- ==============================================
|
||||
CREATE OR REPLACE VIEW recording_overview AS
|
||||
SELECT
|
||||
r.id,
|
||||
r.meeting_id,
|
||||
r.title,
|
||||
r.status as recording_status,
|
||||
r.duration_seconds,
|
||||
r.participant_count,
|
||||
r.recorded_at,
|
||||
r.retention_days,
|
||||
r.recorded_at + (r.retention_days || ' days')::INTERVAL as retention_expires_at,
|
||||
t.id as transcription_id,
|
||||
t.status as transcription_status,
|
||||
t.language,
|
||||
t.word_count,
|
||||
t.confidence_score,
|
||||
c.all_participants_consented,
|
||||
c.consent_type
|
||||
FROM recordings r
|
||||
LEFT JOIN transcriptions t ON t.recording_id = r.id
|
||||
LEFT JOIN meeting_recording_consents c ON c.meeting_id = r.meeting_id
|
||||
WHERE r.deleted_at IS NULL;
|
||||
|
||||
COMMENT ON VIEW recording_overview IS 'Combined view of recordings with transcription and consent status';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- View: Pending Transcriptions
|
||||
-- ==============================================
|
||||
CREATE OR REPLACE VIEW pending_transcriptions AS
|
||||
SELECT
|
||||
t.id,
|
||||
t.recording_id,
|
||||
r.storage_path,
|
||||
r.audio_path,
|
||||
t.language,
|
||||
t.model,
|
||||
q.priority,
|
||||
q.attempts,
|
||||
q.max_attempts,
|
||||
q.created_at as queued_at
|
||||
FROM transcriptions t
|
||||
JOIN recordings r ON r.id = t.recording_id
|
||||
LEFT JOIN transcription_queue q ON q.transcription_id = t.id
|
||||
WHERE t.status IN ('pending', 'queued')
|
||||
AND r.status = 'uploaded'
|
||||
AND r.deleted_at IS NULL
|
||||
ORDER BY q.priority DESC, q.created_at ASC;
|
||||
|
||||
COMMENT ON VIEW pending_transcriptions IS 'Queue of transcriptions waiting to be processed';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Function: Search Transcripts
|
||||
-- ==============================================
|
||||
CREATE OR REPLACE FUNCTION search_transcripts(
|
||||
p_query TEXT,
|
||||
p_language VARCHAR(10) DEFAULT 'de',
|
||||
p_limit INTEGER DEFAULT 20
|
||||
)
|
||||
RETURNS TABLE (
|
||||
transcription_id UUID,
|
||||
recording_id UUID,
|
||||
meeting_id VARCHAR(255),
|
||||
title VARCHAR(500),
|
||||
segment_id UUID,
|
||||
segment_text TEXT,
|
||||
start_time_ms INTEGER,
|
||||
end_time_ms INTEGER,
|
||||
speaker_id VARCHAR(50),
|
||||
relevance FLOAT
|
||||
) AS $$
|
||||
BEGIN
|
||||
RETURN QUERY
|
||||
SELECT
|
||||
t.id as transcription_id,
|
||||
r.id as recording_id,
|
||||
r.meeting_id,
|
||||
r.title,
|
||||
s.id as segment_id,
|
||||
s.text as segment_text,
|
||||
s.start_time_ms,
|
||||
s.end_time_ms,
|
||||
s.speaker_id,
|
||||
ts_rank(to_tsvector('german', s.text), plainto_tsquery('german', p_query))::FLOAT as relevance
|
||||
FROM transcription_segments s
|
||||
JOIN transcriptions t ON t.id = s.transcription_id
|
||||
JOIN recordings r ON r.id = t.recording_id
|
||||
WHERE t.language = p_language
|
||||
AND t.status = 'completed'
|
||||
AND r.deleted_at IS NULL
|
||||
AND to_tsvector('german', s.text) @@ plainto_tsquery('german', p_query)
|
||||
ORDER BY relevance DESC
|
||||
LIMIT p_limit;
|
||||
END;
|
||||
$$ LANGUAGE plpgsql;
|
||||
|
||||
COMMENT ON FUNCTION search_transcripts IS 'Full-text search across all transcription segments';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Function: Cleanup Expired Recordings
|
||||
-- ==============================================
|
||||
-- Call periodically to soft-delete recordings past retention.
|
||||
|
||||
CREATE OR REPLACE FUNCTION cleanup_expired_recordings()
|
||||
RETURNS INTEGER AS $$
|
||||
DECLARE
|
||||
v_count INTEGER;
|
||||
BEGIN
|
||||
WITH expired AS (
|
||||
UPDATE recordings
|
||||
SET status = 'deleted',
|
||||
deleted_at = NOW()
|
||||
WHERE deleted_at IS NULL
|
||||
AND status != 'deleted'
|
||||
AND recorded_at + (retention_days || ' days')::INTERVAL < NOW()
|
||||
RETURNING id
|
||||
)
|
||||
SELECT COUNT(*) INTO v_count FROM expired;
|
||||
|
||||
-- Log the cleanup action
|
||||
IF v_count > 0 THEN
|
||||
INSERT INTO recording_audit_log (action, metadata)
|
||||
VALUES ('retention_expired', jsonb_build_object('count', v_count, 'timestamp', NOW()));
|
||||
END IF;
|
||||
|
||||
RETURN v_count;
|
||||
END;
|
||||
$$ LANGUAGE plpgsql;
|
||||
|
||||
COMMENT ON FUNCTION cleanup_expired_recordings IS 'Soft-deletes recordings past their retention period';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Grant Permissions
|
||||
-- ==============================================
|
||||
-- GRANT ALL PRIVILEGES ON ALL TABLES IN SCHEMA public TO breakpilot;
|
||||
-- GRANT ALL PRIVILEGES ON ALL SEQUENCES IN SCHEMA public TO breakpilot;
|
||||
-- GRANT EXECUTE ON ALL FUNCTIONS IN SCHEMA public TO breakpilot;
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Migration Complete Message
|
||||
-- ==============================================
|
||||
DO $$
|
||||
BEGIN
|
||||
RAISE NOTICE 'Recording & Transcription tables created successfully!';
|
||||
RAISE NOTICE 'Tables: meeting_recording_consents, recordings, transcriptions, transcription_segments, recording_audit_log, transcription_queue';
|
||||
RAISE NOTICE 'Views: recording_overview, pending_transcriptions';
|
||||
RAISE NOTICE 'Functions: search_transcripts, cleanup_expired_recordings';
|
||||
END $$;
|
||||
410
docs-src/backend/migrations/add_unit_tables.sql
Normal file
410
docs-src/backend/migrations/add_unit_tables.sql
Normal file
@@ -0,0 +1,410 @@
|
||||
-- ==============================================
|
||||
-- Breakpilot Drive - Educational Unit Tables Migration
|
||||
-- ==============================================
|
||||
-- Adds tables for the contextual learning unit system.
|
||||
-- Supports FlightPath and StationLoop templates.
|
||||
--
|
||||
-- Execute with:
|
||||
-- psql -h localhost -U breakpilot -d breakpilot -f add_unit_tables.sql
|
||||
|
||||
-- Enable UUID extension if not already enabled
|
||||
CREATE EXTENSION IF NOT EXISTS "uuid-ossp";
|
||||
|
||||
-- ==============================================
|
||||
-- Unit Definitions
|
||||
-- ==============================================
|
||||
-- Stores the configuration for each learning unit.
|
||||
-- JSON definition contains stops, interactions, vocab, etc.
|
||||
|
||||
CREATE TABLE IF NOT EXISTS unit_definitions (
|
||||
unit_id VARCHAR(100) PRIMARY KEY,
|
||||
template VARCHAR(50) NOT NULL CHECK (template IN ('flight_path', 'station_loop')),
|
||||
version VARCHAR(20) NOT NULL,
|
||||
locale VARCHAR(10)[] DEFAULT ARRAY['de-DE'],
|
||||
grade_band VARCHAR(10)[] DEFAULT ARRAY['5', '6'],
|
||||
duration_minutes INTEGER NOT NULL CHECK (duration_minutes >= 3 AND duration_minutes <= 20),
|
||||
difficulty VARCHAR(20) DEFAULT 'base' CHECK (difficulty IN ('base', 'advanced')),
|
||||
definition JSONB NOT NULL,
|
||||
is_published BOOLEAN DEFAULT false,
|
||||
created_at TIMESTAMPTZ DEFAULT NOW(),
|
||||
updated_at TIMESTAMPTZ DEFAULT NOW()
|
||||
);
|
||||
|
||||
-- Index for filtering
|
||||
CREATE INDEX IF NOT EXISTS idx_unit_definitions_template ON unit_definitions(template);
|
||||
CREATE INDEX IF NOT EXISTS idx_unit_definitions_published ON unit_definitions(is_published);
|
||||
CREATE INDEX IF NOT EXISTS idx_unit_definitions_locale ON unit_definitions USING GIN(locale);
|
||||
CREATE INDEX IF NOT EXISTS idx_unit_definitions_grade ON unit_definitions USING GIN(grade_band);
|
||||
|
||||
-- Comments
|
||||
COMMENT ON TABLE unit_definitions IS 'Stores unit configurations for contextual learning experiences';
|
||||
COMMENT ON COLUMN unit_definitions.template IS 'Unit template type: flight_path (linear) or station_loop (hub-based)';
|
||||
COMMENT ON COLUMN unit_definitions.definition IS 'Complete JSON definition including stops, interactions, vocab, etc.';
|
||||
COMMENT ON COLUMN unit_definitions.grade_band IS 'Target grade levels (e.g., ["5", "6", "7"])';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Unit Sessions
|
||||
-- ==============================================
|
||||
-- Records each unit session played by a student.
|
||||
|
||||
CREATE TABLE IF NOT EXISTS unit_sessions (
|
||||
session_id UUID PRIMARY KEY DEFAULT uuid_generate_v4(),
|
||||
unit_id VARCHAR(100) NOT NULL REFERENCES unit_definitions(unit_id),
|
||||
student_id UUID NOT NULL,
|
||||
locale VARCHAR(10) DEFAULT 'de-DE',
|
||||
difficulty VARCHAR(20) DEFAULT 'base',
|
||||
started_at TIMESTAMPTZ DEFAULT NOW(),
|
||||
completed_at TIMESTAMPTZ,
|
||||
aborted_at TIMESTAMPTZ,
|
||||
duration_seconds INTEGER,
|
||||
completion_rate DECIMAL(3,2) CHECK (completion_rate >= 0 AND completion_rate <= 1),
|
||||
precheck_score DECIMAL(3,2) CHECK (precheck_score >= 0 AND precheck_score <= 1),
|
||||
postcheck_score DECIMAL(3,2) CHECK (postcheck_score >= 0 AND postcheck_score <= 1),
|
||||
stops_completed INTEGER DEFAULT 0,
|
||||
total_stops INTEGER DEFAULT 0,
|
||||
session_token VARCHAR(500),
|
||||
metadata JSONB,
|
||||
created_at TIMESTAMPTZ DEFAULT NOW(),
|
||||
updated_at TIMESTAMPTZ DEFAULT NOW()
|
||||
);
|
||||
|
||||
-- Indexes for common queries
|
||||
CREATE INDEX IF NOT EXISTS idx_unit_sessions_student ON unit_sessions(student_id);
|
||||
CREATE INDEX IF NOT EXISTS idx_unit_sessions_unit ON unit_sessions(unit_id);
|
||||
CREATE INDEX IF NOT EXISTS idx_unit_sessions_completed ON unit_sessions(completed_at);
|
||||
CREATE INDEX IF NOT EXISTS idx_unit_sessions_started ON unit_sessions(started_at DESC);
|
||||
|
||||
-- Comments
|
||||
COMMENT ON TABLE unit_sessions IS 'Records individual unit playthrough sessions';
|
||||
COMMENT ON COLUMN unit_sessions.completion_rate IS 'Percentage of stops completed (0.0 to 1.0)';
|
||||
COMMENT ON COLUMN unit_sessions.precheck_score IS 'Score from pre-unit diagnostic quiz (0.0 to 1.0)';
|
||||
COMMENT ON COLUMN unit_sessions.postcheck_score IS 'Score from post-unit diagnostic quiz (0.0 to 1.0)';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Unit Telemetry Events
|
||||
-- ==============================================
|
||||
-- Stores detailed telemetry events from unit sessions.
|
||||
|
||||
CREATE TABLE IF NOT EXISTS unit_telemetry (
|
||||
id BIGSERIAL PRIMARY KEY,
|
||||
session_id UUID NOT NULL REFERENCES unit_sessions(session_id) ON DELETE CASCADE,
|
||||
event_type VARCHAR(50) NOT NULL,
|
||||
stop_id VARCHAR(100),
|
||||
event_timestamp TIMESTAMPTZ DEFAULT NOW(),
|
||||
metrics JSONB,
|
||||
created_at TIMESTAMPTZ DEFAULT NOW()
|
||||
);
|
||||
|
||||
-- Indexes for analytics
|
||||
CREATE INDEX IF NOT EXISTS idx_unit_telemetry_session ON unit_telemetry(session_id);
|
||||
CREATE INDEX IF NOT EXISTS idx_unit_telemetry_type ON unit_telemetry(event_type);
|
||||
CREATE INDEX IF NOT EXISTS idx_unit_telemetry_stop ON unit_telemetry(stop_id);
|
||||
CREATE INDEX IF NOT EXISTS idx_unit_telemetry_timestamp ON unit_telemetry(event_timestamp);
|
||||
|
||||
-- Partitioning hint (for production with high volume)
|
||||
-- Consider partitioning by created_at for older data cleanup
|
||||
|
||||
-- Comments
|
||||
COMMENT ON TABLE unit_telemetry IS 'Detailed telemetry events from unit sessions';
|
||||
COMMENT ON COLUMN unit_telemetry.event_type IS 'Event type: stop_completed, hint_used, state_change, etc.';
|
||||
COMMENT ON COLUMN unit_telemetry.metrics IS 'Event-specific metrics in JSON format';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Unit Stop Metrics
|
||||
-- ==============================================
|
||||
-- Aggregated metrics per stop per session.
|
||||
|
||||
CREATE TABLE IF NOT EXISTS unit_stop_metrics (
|
||||
id BIGSERIAL PRIMARY KEY,
|
||||
session_id UUID NOT NULL REFERENCES unit_sessions(session_id) ON DELETE CASCADE,
|
||||
stop_id VARCHAR(100) NOT NULL,
|
||||
completed BOOLEAN DEFAULT false,
|
||||
success BOOLEAN,
|
||||
attempts INTEGER DEFAULT 0,
|
||||
time_seconds DECIMAL(10,2),
|
||||
hints_used TEXT[],
|
||||
completed_at TIMESTAMPTZ,
|
||||
created_at TIMESTAMPTZ DEFAULT NOW(),
|
||||
CONSTRAINT unique_session_stop UNIQUE(session_id, stop_id)
|
||||
);
|
||||
|
||||
-- Indexes
|
||||
CREATE INDEX IF NOT EXISTS idx_stop_metrics_session ON unit_stop_metrics(session_id);
|
||||
CREATE INDEX IF NOT EXISTS idx_stop_metrics_stop ON unit_stop_metrics(stop_id);
|
||||
CREATE INDEX IF NOT EXISTS idx_stop_metrics_success ON unit_stop_metrics(success);
|
||||
|
||||
-- Comments
|
||||
COMMENT ON TABLE unit_stop_metrics IS 'Aggregated metrics for each stop in a unit session';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Unit Misconceptions
|
||||
-- ==============================================
|
||||
-- Tracks detected misconceptions per student.
|
||||
|
||||
CREATE TABLE IF NOT EXISTS unit_misconceptions (
|
||||
id BIGSERIAL PRIMARY KEY,
|
||||
student_id UUID NOT NULL,
|
||||
unit_id VARCHAR(100) NOT NULL REFERENCES unit_definitions(unit_id),
|
||||
misconception_id VARCHAR(100) NOT NULL,
|
||||
stop_id VARCHAR(100),
|
||||
detected_at TIMESTAMPTZ DEFAULT NOW(),
|
||||
addressed BOOLEAN DEFAULT false,
|
||||
addressed_at TIMESTAMPTZ,
|
||||
session_id UUID REFERENCES unit_sessions(session_id),
|
||||
CONSTRAINT unique_student_misconception UNIQUE(student_id, unit_id, misconception_id)
|
||||
);
|
||||
|
||||
-- Indexes
|
||||
CREATE INDEX IF NOT EXISTS idx_misconceptions_student ON unit_misconceptions(student_id);
|
||||
CREATE INDEX IF NOT EXISTS idx_misconceptions_unit ON unit_misconceptions(unit_id);
|
||||
CREATE INDEX IF NOT EXISTS idx_misconceptions_addressed ON unit_misconceptions(addressed);
|
||||
|
||||
-- Comments
|
||||
COMMENT ON TABLE unit_misconceptions IS 'Tracks detected misconceptions for targeted remediation';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Trigger: Update updated_at timestamp
|
||||
-- ==============================================
|
||||
-- Reuse existing function if available, otherwise create
|
||||
DO $$
|
||||
BEGIN
|
||||
IF NOT EXISTS (SELECT 1 FROM pg_proc WHERE proname = 'update_updated_at_column') THEN
|
||||
CREATE FUNCTION update_updated_at_column()
|
||||
RETURNS TRIGGER AS $func$
|
||||
BEGIN
|
||||
NEW.updated_at = NOW();
|
||||
RETURN NEW;
|
||||
END;
|
||||
$func$ LANGUAGE plpgsql;
|
||||
END IF;
|
||||
END $$;
|
||||
|
||||
-- Apply triggers
|
||||
DROP TRIGGER IF EXISTS update_unit_definitions_updated_at ON unit_definitions;
|
||||
CREATE TRIGGER update_unit_definitions_updated_at
|
||||
BEFORE UPDATE ON unit_definitions
|
||||
FOR EACH ROW
|
||||
EXECUTE FUNCTION update_updated_at_column();
|
||||
|
||||
DROP TRIGGER IF EXISTS update_unit_sessions_updated_at ON unit_sessions;
|
||||
CREATE TRIGGER update_unit_sessions_updated_at
|
||||
BEFORE UPDATE ON unit_sessions
|
||||
FOR EACH ROW
|
||||
EXECUTE FUNCTION update_updated_at_column();
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- View: Unit Session Summary
|
||||
-- ==============================================
|
||||
CREATE OR REPLACE VIEW unit_session_summary AS
|
||||
SELECT
|
||||
us.session_id,
|
||||
us.unit_id,
|
||||
ud.template,
|
||||
us.student_id,
|
||||
us.started_at,
|
||||
us.completed_at,
|
||||
us.duration_seconds,
|
||||
us.completion_rate,
|
||||
us.precheck_score,
|
||||
us.postcheck_score,
|
||||
CASE
|
||||
WHEN us.precheck_score IS NOT NULL AND us.postcheck_score IS NOT NULL
|
||||
THEN us.postcheck_score - us.precheck_score
|
||||
ELSE NULL
|
||||
END as learning_gain,
|
||||
us.stops_completed,
|
||||
us.total_stops,
|
||||
CASE WHEN us.completed_at IS NOT NULL THEN true ELSE false END as is_completed
|
||||
FROM unit_sessions us
|
||||
JOIN unit_definitions ud ON us.unit_id = ud.unit_id;
|
||||
|
||||
COMMENT ON VIEW unit_session_summary IS 'Summary view of unit sessions with learning gain calculation';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- View: Unit Analytics by Student
|
||||
-- ==============================================
|
||||
CREATE OR REPLACE VIEW unit_student_analytics AS
|
||||
SELECT
|
||||
student_id,
|
||||
COUNT(DISTINCT unit_id) as units_attempted,
|
||||
COUNT(*) as total_sessions,
|
||||
COUNT(*) FILTER (WHERE completed_at IS NOT NULL) as completed_sessions,
|
||||
AVG(completion_rate) as avg_completion_rate,
|
||||
AVG(precheck_score) as avg_precheck_score,
|
||||
AVG(postcheck_score) as avg_postcheck_score,
|
||||
AVG(CASE
|
||||
WHEN precheck_score IS NOT NULL AND postcheck_score IS NOT NULL
|
||||
THEN postcheck_score - precheck_score
|
||||
ELSE NULL
|
||||
END) as avg_learning_gain,
|
||||
SUM(duration_seconds) / 60 as total_minutes_played,
|
||||
MAX(completed_at) as last_completed_at
|
||||
FROM unit_sessions
|
||||
GROUP BY student_id;
|
||||
|
||||
COMMENT ON VIEW unit_student_analytics IS 'Aggregated analytics per student across all units';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- View: Unit Performance
|
||||
-- ==============================================
|
||||
CREATE OR REPLACE VIEW unit_performance AS
|
||||
SELECT
|
||||
ud.unit_id,
|
||||
ud.template,
|
||||
ud.difficulty,
|
||||
COUNT(us.session_id) as total_sessions,
|
||||
COUNT(*) FILTER (WHERE us.completed_at IS NOT NULL) as completed_sessions,
|
||||
ROUND(
|
||||
COUNT(*) FILTER (WHERE us.completed_at IS NOT NULL)::DECIMAL /
|
||||
NULLIF(COUNT(us.session_id), 0) * 100,
|
||||
1
|
||||
) as completion_percent,
|
||||
AVG(us.duration_seconds) / 60 as avg_duration_minutes,
|
||||
AVG(us.completion_rate) as avg_completion_rate,
|
||||
AVG(CASE
|
||||
WHEN us.precheck_score IS NOT NULL AND us.postcheck_score IS NOT NULL
|
||||
THEN us.postcheck_score - us.precheck_score
|
||||
ELSE NULL
|
||||
END) as avg_learning_gain
|
||||
FROM unit_definitions ud
|
||||
LEFT JOIN unit_sessions us ON ud.unit_id = us.unit_id
|
||||
GROUP BY ud.unit_id, ud.template, ud.difficulty;
|
||||
|
||||
COMMENT ON VIEW unit_performance IS 'Performance metrics per unit for content optimization';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Function: Get Recommended Units for Student
|
||||
-- ==============================================
|
||||
CREATE OR REPLACE FUNCTION get_recommended_units(
|
||||
p_student_id UUID,
|
||||
p_grade VARCHAR(10) DEFAULT NULL,
|
||||
p_locale VARCHAR(10) DEFAULT 'de-DE',
|
||||
p_limit INTEGER DEFAULT 5
|
||||
)
|
||||
RETURNS TABLE (
|
||||
unit_id VARCHAR(100),
|
||||
template VARCHAR(50),
|
||||
difficulty VARCHAR(20),
|
||||
reason TEXT
|
||||
) AS $$
|
||||
BEGIN
|
||||
RETURN QUERY
|
||||
SELECT
|
||||
ud.unit_id,
|
||||
ud.template,
|
||||
ud.difficulty,
|
||||
CASE
|
||||
-- Never played
|
||||
WHEN NOT EXISTS (
|
||||
SELECT 1 FROM unit_sessions us
|
||||
WHERE us.student_id = p_student_id AND us.unit_id = ud.unit_id
|
||||
) THEN 'Neu: Noch nicht gespielt'
|
||||
-- Played but not completed
|
||||
WHEN NOT EXISTS (
|
||||
SELECT 1 FROM unit_sessions us
|
||||
WHERE us.student_id = p_student_id
|
||||
AND us.unit_id = ud.unit_id
|
||||
AND us.completed_at IS NOT NULL
|
||||
) THEN 'Fortsetzen: Noch nicht abgeschlossen'
|
||||
-- Completed with low postcheck score
|
||||
WHEN EXISTS (
|
||||
SELECT 1 FROM unit_sessions us
|
||||
WHERE us.student_id = p_student_id
|
||||
AND us.unit_id = ud.unit_id
|
||||
AND us.postcheck_score < 0.6
|
||||
) THEN 'Wiederholen: Verständnis vertiefen'
|
||||
ELSE 'Abgeschlossen'
|
||||
END as reason
|
||||
FROM unit_definitions ud
|
||||
WHERE ud.is_published = true
|
||||
AND (p_locale = ANY(ud.locale) OR p_locale IS NULL)
|
||||
AND (p_grade = ANY(ud.grade_band) OR p_grade IS NULL)
|
||||
ORDER BY
|
||||
-- Prioritize: new > incomplete > low score > completed
|
||||
CASE
|
||||
WHEN NOT EXISTS (
|
||||
SELECT 1 FROM unit_sessions us
|
||||
WHERE us.student_id = p_student_id AND us.unit_id = ud.unit_id
|
||||
) THEN 1
|
||||
WHEN NOT EXISTS (
|
||||
SELECT 1 FROM unit_sessions us
|
||||
WHERE us.student_id = p_student_id
|
||||
AND us.unit_id = ud.unit_id
|
||||
AND us.completed_at IS NOT NULL
|
||||
) THEN 2
|
||||
WHEN EXISTS (
|
||||
SELECT 1 FROM unit_sessions us
|
||||
WHERE us.student_id = p_student_id
|
||||
AND us.unit_id = ud.unit_id
|
||||
AND us.postcheck_score < 0.6
|
||||
) THEN 3
|
||||
ELSE 4
|
||||
END,
|
||||
ud.unit_id
|
||||
LIMIT p_limit;
|
||||
END;
|
||||
$$ LANGUAGE plpgsql;
|
||||
|
||||
COMMENT ON FUNCTION get_recommended_units IS 'Returns recommended units for a student based on completion status';
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Sample Data: Demo Unit Definition
|
||||
-- ==============================================
|
||||
-- Insert a demo unit for testing (will be replaced by real content)
|
||||
|
||||
INSERT INTO unit_definitions (unit_id, template, version, locale, grade_band, duration_minutes, difficulty, definition, is_published)
|
||||
VALUES (
|
||||
'demo_unit_v1',
|
||||
'flight_path',
|
||||
'1.0.0',
|
||||
ARRAY['de-DE'],
|
||||
ARRAY['5', '6', '7'],
|
||||
5,
|
||||
'base',
|
||||
'{
|
||||
"unit_id": "demo_unit_v1",
|
||||
"template": "flight_path",
|
||||
"version": "1.0.0",
|
||||
"learning_objectives": ["Demo: Grundfunktion testen", "Demo: Navigation verstehen"],
|
||||
"stops": [
|
||||
{"stop_id": "stop_1", "label": {"de-DE": "Start"}, "interaction": {"type": "aim_and_pass"}},
|
||||
{"stop_id": "stop_2", "label": {"de-DE": "Mitte"}, "interaction": {"type": "aim_and_pass"}},
|
||||
{"stop_id": "stop_3", "label": {"de-DE": "Ende"}, "interaction": {"type": "aim_and_pass"}}
|
||||
],
|
||||
"teacher_controls": {"allow_skip": true, "allow_replay": true}
|
||||
}'::jsonb,
|
||||
true
|
||||
)
|
||||
ON CONFLICT (unit_id) DO UPDATE SET
|
||||
definition = EXCLUDED.definition,
|
||||
updated_at = NOW();
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Grant Permissions (adjust user as needed)
|
||||
-- ==============================================
|
||||
-- GRANT ALL PRIVILEGES ON ALL TABLES IN SCHEMA public TO breakpilot;
|
||||
-- GRANT ALL PRIVILEGES ON ALL SEQUENCES IN SCHEMA public TO breakpilot;
|
||||
-- GRANT EXECUTE ON ALL FUNCTIONS IN SCHEMA public TO breakpilot;
|
||||
|
||||
|
||||
-- ==============================================
|
||||
-- Migration Complete Message
|
||||
-- ==============================================
|
||||
DO $$
|
||||
BEGIN
|
||||
RAISE NOTICE 'Breakpilot Drive unit tables created successfully!';
|
||||
RAISE NOTICE 'Tables: unit_definitions, unit_sessions, unit_telemetry, unit_stop_metrics, unit_misconceptions';
|
||||
RAISE NOTICE 'Views: unit_session_summary, unit_student_analytics, unit_performance';
|
||||
RAISE NOTICE 'Functions: get_recommended_units';
|
||||
END $$;
|
||||
Reference in New Issue
Block a user