Add elastic service and code of llm-pdf app

.dockerignore

@@ -0,0 +1,37 @@
+# Ignore Python cache files
+**/__pycache__
+
+# Ignore pytest cache files
+.pytest_cache/
+
+# Ignore ruff files
+.ruff_cache/
+
+# Ignore git files
+.gitignore
+.github/
+
+# Ignore Dockerfile and .dockerignore
+Dockerfile
+.dockerignore
+
+# Ignore other unnecessary files
+*.pyc
+*.pyo
+*.pyd
+*~
+
+# Ignore makefile
+Makefile
+
+# Environment variables
+.envrc
+
+# Fly.io
+fly.toml
+
+# Elastic container config
+elastic/
+
+# temporary data
+tmp/

.envrc-example

@@ -0,0 +1,4 @@
+export STREAMLIT_SERVER_PORT=8000
+export ELASTIC_URL='http://elacticserver:9200'
+export OPENAI_API_KEY='<<apikey here>>'
+export STREAMLIT_SERVER_COOKIE_SECRET='<<uuid here>>'

.github/workflows/fly.yml

@@ -0,0 +1,43 @@
+name: Fly Deploy
+on:
+  push:
+    branches:
+      - 'master'
+env:
+  FLY_API_TOKEN: ${{ secrets.FLY_API_TOKEN }}
+jobs:
+  test:
+    name: Run tests
+    runs-on: ubuntu-latest
+    env:
+      STREAMLIT_SERVER_PORT: 8080
+      STREAMLIT_SERVER_COOKIE_SECRET: 00000000-0000-0000-0000-000000000000
+      OPENAI_API_KEY: nnn
+    steps:   
+      - uses: actions/checkout@v4
+      - name: Restore asdf cache
+        id: asdf-cache
+        uses: actions/cache@v4
+        with:
+          path: ~/.asdf
+          key: ${{ runner.os }}-asdf-${{ hashFiles('.tool-versions') }}
+      - name: Install asdf
+        uses: asdf-vm/actions/install@v3
+      - name: Restore poetry cache
+        id: poetry-cache
+        uses: actions/cache@v4
+        with:
+          path: ~/.cache/pypoetry
+          key: ${{ runner.os }}-poetry-${{ hashFiles('pyproject.toml') }}
+      - name: Install Dependencies
+        run: make deps
+      - name: Run Linter
+        run: make lint
+  deploy:
+      name: Deploy to fly.io
+      runs-on: ubuntu-latest
+      needs: test
+      steps:
+        - uses: actions/checkout@v4
+        - uses: superfly/flyctl-actions/setup-flyctl@master
+        - run: flyctl deploy --remote-only

.gitignore

@@ -0,0 +1,49 @@
+# .gitignore
+__pycache__/
+.pytest_cache/
+.ruff_cache/
+*.pyc
+*.pyo
+*.pyd
+*.pyc
+*.pyo
+*.pyd
+*.egg-info/
+dist/
+build/
+*.egg
+*.log
+*.sql
+*.sqlite
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+
+# IDEs
+.idea/
+*.swp
+*.swo
+*.swn
+*.pyc
+*.pyo
+*.pyd
+*.egg-info/
+dist/
+build/
+*.egg
+*.log
+*.sql
+*.sqlite
+
+# Poetry
+/poetry.lock
+
+# Environment variables
+.envrc
+
+# temporary data
+tmp/

.streamlit/config.toml

@@ -0,0 +1,2 @@
+[server]
+maxUploadSize = 3

.tool-versions

@@ -0,0 +1,3 @@
+python 3.11.2
+poetry 1.8.2
+direnv 2.32.1

Dockerfile

@@ -0,0 +1,16 @@
+# Use an official Python runtime as a parent image
+FROM python:3.11-slim-buster
+
+# Set the working directory in the container to /app
+WORKDIR /app
+
+# Add the current directory contents into the container at /app
+ADD . /app
+
+# Install any needed packages specified in pyproject.toml
+RUN pip install poetry
+RUN poetry config virtualenvs.create false
+RUN poetry install --no-dev
+
+# Run app.py when the container launches
+CMD poetry run streamlit run app.py

Makefile

@@ -0,0 +1,22 @@
+.PHONY: deps lint shell server server_headless test
+
+deps:
+	poetry install
+
+lint:
+	poetry run ruff check . 
+
+shell:
+	poetry run python
+
+server:
+	poetry run streamlit run app.py
+
+server_headless:
+	poetry run streamlit run app.py --browser.serverAddress 0.0.0.0 --server.headless true
+
+test:
+	poetry run -- ptw -- -s -vv $(args)
+
+test_once:
+	poetry run pytest -s

app.py

@@ -0,0 +1,5 @@
+from src.llm_pdf import llm_pdf_app
+
+
+if __name__ == "__main__":
+    llm_pdf_app()

elastic/Dockerfile

@@ -0,0 +1,10 @@
+# Use the official Elasticsearch image
+FROM docker.elastic.co/elasticsearch/elasticsearch:8.4.3
+
+# Set environment variables
+ENV discovery.type=single-node
+ENV xpack.security.enabled=false
+
+# Expose ports
+EXPOSE 9200
+EXPOSE 9300

elastic/fly.toml

@@ -0,0 +1,25 @@
+# fly.toml app configuration file generated for llm-pdf-elastic on 2024-06-20T15:19:26+02:00
+#
+# See https://fly.io/docs/reference/configuration/ for information about how to use this file.
+#
+
+app = 'llm-pdf-elastic'
+primary_region = 'ams'
+
+[build]
+
+[http_service]
+  internal_port = 9200
+  auto_stop_machines = true
+  auto_start_machines = true
+  min_machines_running = 1
+  processes = ['app']
+
+  [http_service.concurrency]
+    type = 'requests'
+    soft_limit = 250
+
+[[vm]]
+  memory = '1gb'
+  cpu_kind = 'shared'
+  cpus = 1

elastic/proxy_port.sh

@@ -0,0 +1,2 @@
+#!/bin/bash
+fly proxy 19200:9200 -a llm-pdf-elastic

fly.toml

@@ -0,0 +1,48 @@
+# fly.toml app configuration file generated for llm-pdf on 2024-06-21T08:48:12+02:00
+#
+# See https://fly.io/docs/reference/configuration/ for information about how to use this file.
+#
+
+app = 'llm-pdf'
+primary_region = 'ams'
+kill_signal = 'SIGTERM'
+kill_timeout = '5s'
+
+[experimental]
+  auto_rollback = true
+
+[build]
+
+[deploy]
+  strategy = 'rolling'
+  wait_timeout = '10m0s'
+
+[env]
+  ELASTIC_URL = 'http://llm-pdf-elastic.internal:9200'
+  HOST = 'llm-pdf.fly.dev'
+  STREAMLIT_SERVER_PORT = '8080'
+
+[http_service]
+  internal_port = 8080
+  force_https = true
+  auto_stop_machines = true
+  auto_start_machines = true
+  min_machines_running = 1
+  processes = ['app']
+
+  [http_service.concurrency]
+    type = 'connections'
+    hard_limit = 1000
+    soft_limit = 1000
+
+  [[http_service.checks]]
+    interval = '30s'
+    timeout = '5s'
+    grace_period = '10s'
+    method = 'GET'
+    path = '/'
+
+[[vm]]
+  memory = '1024mb'
+  cpu_kind = 'shared'
+  cpus = 1

logging.conf

@@ -0,0 +1,21 @@
+[loggers]
+keys=root
+
+[handlers]
+keys=consoleHandler
+
+[formatters]
+keys=normalFormatter
+
+[logger_root]
+level=INFO
+handlers=consoleHandler
+
+[handler_consoleHandler]
+class=StreamHandler
+level=DEBUG
+formatter=normalFormatter
+args=(sys.stdout,)
+
+[formatter_normalFormatter]
+format=%(asctime)s loglevel=%(levelname)-6s logger=%(name)s %(message)s