add control env variables, update README

.env_EXAMPLE

@@ -6,6 +6,12 @@ DB_PASS=password
 DB_HOST=localhost
 DB_DATABASE=academy_ruins
 
+# Tika configuration (required for parsing MTRs)
+USE_TIKA=1
+# By default, the parser downloads and locally runs a Tika server as needed (this requires Java)
+# Uncomment (and adjust) the following line only if you want to connect to an existing Tika server
+# TIKA_URL=http://localhost:9998
+
 # Pushover configuration for notifications
 USE_PUSHOVER=0
 PUSHOVER_APP_TOKEN=

README.md

@@ -11,6 +11,7 @@ It is used to power the [Academy Ruins](https://github.com/lunakv/academyruins)
 - Python 3.10 or later
 - PostgreSQL 14
 - (optional) [Pushover](https://pushover.net/) account
+- (recommended) Working Java installation
 
 ### Installation
 1. Install the [Poetry](https://python-poetry.org/docs/#installation) package manager

app/parsing/mtr/extract_mtr.py

@@ -1,8 +1,10 @@
 import json
+import os
 import re
 from dataclasses import asdict
 from pathlib import Path
 
+from dotenv import load_dotenv
 from tika import parser
 
 from app.utils.models import MtrSubsection, MtrAuxiliarySection, MtrNumberedSection
@@ -139,7 +141,15 @@ def build_structure(chunks: [MtrSubsection]):
 
 
 def extract(filepath: Path | str) -> [dict]:
-    content = parser.from_file(str(filepath))["content"]
+    if os.environ.get("USE_TIKA") != "1":
+        return None
+
+    args = [str(filepath)]
+    tika_server = os.environ.get("TIKA_URL")
+    if tika_server:
+        args.append(tika_server)
+
+    content = parser.from_file(*args)["content"]
     content = remove_page_nums(trim_content(content))
     chunks = split_into_chunks(content)
 
@@ -155,6 +165,7 @@ def extract(filepath: Path | str) -> [dict]:
 
 
 if __name__ == "__main__":
+    load_dotenv("/home/vaasa/rules-api/.env")
     parsed = extract("./in.pdf")
     with open("./out.json", "w") as file:
         json.dump(parsed, file)

docker-compose.yml

@@ -4,24 +4,25 @@ services:
     ports:
       - "8000:80"
     env_file:
-      # dotenv isn't run in production
+      # load_dotenv isn't run in production
       - .env
-#    environment:
-#      # overrides for values set in .env, if required
-#      ADMIN_KEY: "super-secret-value"
-#      DB_USER: server
-#      DB_PASS: password
-#      DB_HOST: db
-#      DB_DATABASE: academy_ruins
+    environment:
+      # overrides for values set in .env, if required
+      USE_TIKA: 1
+      TIKA_URL: http://tika:9998
     volumes:
       - ./app:/code/app:rw
       - ./generated:/code/app/resources/generated:rw
 
-
   db:
     image: postgres
     restart: always
     ports:
       - 5432:5432
     environment:
-      POSTGRES_PASSWORD: password
+      POSTGRES_PASSWORD: password
+
+  tika:
+    image: apache/tika
+    ports:
+      - 9998:9998