Restructure: Move services from root to unified repo

Moved updated services from /home/sam/development/ root into aboutme_chat_demo/:
- knowledge_service/ (with ChromaDB, gitea_scraper, FastAPI)
- langgraph_service/ (with LangGraph agent orchestration)
- airflow/ (with DAGs for scheduled ingestion)

All services now in single repo location.
Modular docker-compose files per service maintained.
Removed duplicate nested directories.
Updated files reflect latest working versions.
This commit is contained in:
2026-02-28 14:51:37 +11:00
parent 628ba96998
commit 76f7367e2f
7 changed files with 99 additions and 84 deletions

View File

@@ -5,10 +5,8 @@ Runs daily to fetch new/updated repos and ingest into ChromaDB.
from datetime import datetime, timedelta
from airflow import DAG
from airflow.operators.python import PythonOperator
from airflow.providers.http.operators.http import SimpleHttpOperator
import os
import sys
import json
# Add knowledge_service to path for imports
sys.path.insert(0, '/opt/airflow/dags/repo')
@@ -115,30 +113,26 @@ with DAG(
'gitea_daily_ingestion',
default_args=default_args,
description='Daily ingestion of Gitea repositories into knowledge base',
schedule_interval=timedelta(days=1), # Run daily
schedule_interval=timedelta(days=1),
start_date=datetime(2024, 1, 1),
catchup=False,
tags=['gitea', 'ingestion', 'knowledge'],
) as dag:
# Task 1: Fetch repository list
fetch_repos_task = PythonOperator(
task_id='fetch_repos',
python_callable=fetch_gitea_repos,
)
# Task 2: Fetch README content
fetch_readmes_task = PythonOperator(
task_id='fetch_readmes',
python_callable=fetch_readmes,
)
# Task 3: Ingest into ChromaDB
ingest_task = PythonOperator(
task_id='ingest_to_chroma',
python_callable=ingest_to_chroma,
)
# Define task dependencies
fetch_repos_task >> fetch_readmes_task >> ingest_task