Merge pull request #1 from Addepto/feat/docker-fastapi

Feat/docker fastapi

Author: filipmomot

.env_example

@@ -0,0 +1,5 @@
+API_PORT=9000
+DEBUG=True
+LOG_LEVEL=debug
+API_BASE_URL=http://localhost:9000
+API_KEY=xxx

.gitignore

@@ -4,4 +4,7 @@ data/
 *.ipynb_checkpoint*
 notebooks/test*
 
-.venv
+.venv
+.env
+
+.DS_store

README.md

@@ -1,7 +1,9 @@
 # graph_builder
-Graph builder package. 
+
+Graph builder package.
 
 ## Installation
+
 Following good practices we suggest you create a separate virtual environment for working with Graph builder package.
 
 Note that the graph_builder requires python 3.12 or higher.
@@ -14,6 +16,7 @@ poetry install
 
 Note that this is a short example taken from `examples/example1.ipynb`, for more information please
 refer to it.
+
 ```python
 from entity_graph.graph_extractor.entities_graph_extractor import EntitiesGraphExtractor
 
@@ -32,8 +35,30 @@ config = {
 # Load table to graph
 extractor.load_table_from_file(
     config,
-    "coffee_machines.pdf", 
-    "Machines", 
+    "coffee_machines.pdf",
+    "Machines",
     "instances",
 )
 ```
+
+## graph_builder FastApi
+
+In the folder containing the `docker-compose.yml` file, run the commands:
+
+```
+docker compose build
+```
+
+Once the image is built:
+
+```
+docker compose up
+```
+
+Make sure to create the `.env` file in the directory based on the `.env_example` file with the needed environmental values.
+
+## Important notes
+
+In the current version of the application, graphs are retained between requests but not preserved across API restarts.
+
+This means that each time the API is restarted, the graphs must be rebuilt.

docker-compose.yml

@@ -0,0 +1,24 @@
+services:
+  entity_graph_api:
+    build:
+      context: .
+      dockerfile: ./src/entity_graph_api/Dockerfile
+    restart: unless-stopped
+    env_file:
+      - .env
+    volumes:
+      - ./src:/app/src
+    ports:
+      - "${API_PORT:-8000}:8000"
+    environment:
+      API_PORT: ${API_PORT:-8000}
+      DEBUG: ${DEBUG:-true}
+      LOG_LEVEL: ${LOG_LEVEL}
+      PYTHONPATH: /app:/app/src
+    healthcheck:
+      test: ["CMD", "curl", "-f", "http://localhost:8000/health/live"]
+      interval: 1m
+      timeout: 10s
+      retries: 10
+      start_period: 30s
+    command: poetry run uvicorn entity_graph_api.main:app --host 0.0.0.0 --port 8000 --log-level ${LOG_LEVEL:-debug} --reload

poetry.lock

@@ -17,6 +17,9 @@ pymupdf = "^1.26.3"
 msoffcrypto-tool = "^5.4.2"
 xlrd = "^2.0.2"
 openpyxl = "^3.1.5"
+fastapi = "^0.116.1"
+uvicorn = "^0.35.0"
+python-multipart = "^0.0.20"
 
 
 [tool.poetry.group.dev.dependencies]

pyproject.toml

@@ -0,0 +1,54 @@
+########
+# BASE #
+########
+FROM python:3.12-slim
+
+WORKDIR /app/
+
+# Set environment variables - fixed Poetry configuration
+ENV PYTHONDONTWRITEBYTECODE=1 \
+    PYTHONUNBUFFERED=1 \
+    POETRY_VERSION=1.5.1 \
+    POETRY_VIRTUALENVS_CREATE=false \
+    POETRY_VIRTUALENVS_IN_PROJECT=false \
+    POETRY_NO_INTERACTION=1 \
+    POETRY_CACHE_DIR=/tmp/poetry_cache \
+    PYTHONPATH=/app
+
+# Install system dependencies
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    curl \
+    && rm -rf /var/lib/apt/lists/*
+
+# Install Poetry
+RUN pip install "poetry==$POETRY_VERSION"
+
+# Copy only requirements to cache them in docker layer
+COPY pyproject.toml ./
+# Copy poetry.lock if it exists
+COPY poetry.lock* ./
+
+# Install dependencies globally since we're not using virtualenvs
+RUN --mount=type=cache,target=$POETRY_CACHE_DIR poetry install --no-root --no-ansi --no-interaction
+
+ARG commit
+ARG version
+ARG build_time
+
+# Create build info
+RUN echo "COMMIT: ${commit:-unknown}" > build_info && \
+    echo "VERSION: ${version:-dev}" >> build_info && \
+    echo "BUILD_TIME: ${build_time:-$(date)}" >> build_info
+
+ENV COMMIT=${commit:-unknown}
+ENV VERSION=${version:-dev}
+ENV BUILD_TIME=${build_time:-$(date)}
+
+# Copy application code
+COPY . .
+
+# Install the local project package into the image so top-level packages
+# like `entity_graph` are available at runtime regardless of PYTHONPATH
+# set by docker-compose. We run a full poetry install here (no-root omitted
+# earlier to cache deps) which will install the package into the image.
+RUN --mount=type=cache,target=$POETRY_CACHE_DIR poetry install --no-ansi --no-interaction

src/entity_graph_api/Dockerfile

@@ -0,0 +1,61 @@
+import logging
+import logging.config
+import sys
+import warnings
+from enum import StrEnum
+from typing import Annotated
+
+from loguru import logger as logger_loguru
+from pydantic import AfterValidator, AnyUrl
+from pydantic_settings import BaseSettings, SettingsConfigDict
+from rich.console import Console
+
+from entity_graph_api.config_logger import config_logger
+
+console = Console()
+logger_loguru.level("INFO", color="<blue>")
+logger_loguru.level("DEBUG", color="<cyan>")
+
+def setup_logger(name: str) -> logging.Logger:
+
+    logging.config.dictConfig(config_logger)
+
+    warnings.filterwarnings("ignore")
+
+    logger_loguru.configure(
+        handlers=[
+            {
+                "sink": sys.stdout,
+                "level": CONFIG.LOGGING_LEVEL,
+            }
+        ]
+    )
+
+    return logger_loguru.bind(name=name)
+
+
+URL = Annotated[AnyUrl, AfterValidator(str)]
+
+UPLOAD_DIR = "/app/uploads"
+
+# Custom StrEnum is more handy than using integers that represent logging levels
+class LoggingLevels(StrEnum):
+    DEBUG = "DEBUG"
+    INFO = "INFO"
+    WARNING = "WARNING"
+    ERROR = "ERROR"
+    CRITICAL = "CRITICAL"
+
+class Settings(BaseSettings):
+    model_config = SettingsConfigDict(
+        env_file=(".env"), env_nested_delimiter="__", extra="ignore"
+    )
+
+    API_BASE_URL: URL | None = None
+
+    LOGGING_LEVEL: LoggingLevels = LoggingLevels.DEBUG
+
+    API_KEY: str | None = None
+
+
+CONFIG = Settings()

src/entity_graph_api/__init.__.py

@@ -0,0 +1,61 @@
+import importlib.util
+
+config_logger = {
+    "version": 1,
+    "disable_existing_loggers": False,
+    "root": {"handlers": ["console"], "level": "DEBUG"},
+    "handlers": {
+        "console": {
+            "formatter": "std_out",
+            "class": "logging.StreamHandler",
+            "level": "DEBUG",
+            "stream": "ext://sys.stdout",
+        },
+        "console_uvi": (
+            {
+                "formatter": "uvi",
+                "class": "logging.StreamHandler",
+                "level": "DEBUG",
+                "stream": "ext://sys.stdout",
+            }
+        ),
+    },
+    "formatters": {
+        "std_out": {
+            "format": "%(name)s : %(asctime)s : %(levelname)s : %(module)s : %(funcName)s : %(lineno)d : %(message)s",
+            "datefmt": "%d-%m-%Y %I:%M:%S",
+        },
+        "uvi": (
+            {
+                "()": ("uvicorn.logging.DefaultFormatter"),
+                "fmt": "%(levelprefix)s %(asctime)s - %(message)s",
+                "use_colors": True,
+            }
+        ),
+    },
+    "loggers": {
+        "entity_graph": {
+            "level": "DEBUG",
+        },
+        "openai": {
+            "level": "WARNING",
+        },
+        "watchfiles": {
+            "level": "WARNING",
+        },
+        "httpcore": {
+            "level": "WARNING",
+        },
+        "uvicorn": {
+            "level": "DEBUG",
+            "propagate": False,
+            "handlers": ["console_uvi"]
+        },
+    },
+}
+
+# Removing uvicorn-related logger config if there is no uvicorn module
+if not importlib.util.find_spec("uvicorn"):
+    config_logger["handlers"].pop("console_uvi")
+    config_logger["loggers"].pop("uvicorn")
+    config_logger["formatters"].pop("uvi")