feat(file): add apache tika for file content extraction

docker-compose.yml

@@ -9,16 +9,20 @@ services:
       # - POSTGRES_PASSWORD=
     volumes:
       - dbdata:/var/lib/postgresql/data
+
   alexandria:
     image: ghcr.io/projectcaluma/alexandria:dev
     ports:
       - "8000:8000"
     depends_on:
       - db
       - minio
+      - tika
     environment:
       - DATABASE_HOST=db
       - DATABASE_PORT=5432
+      - TIKA_CLIENT_ONLY=true
+      - TIKA_SERVER_ENDPOINT=http://tika:9998
       # following options are a must to configure on production system:
       # https://docs.djangoproject.com/en/2.1/ref/settings/#std:setting-SECRET_KEY
       # - SECRET_KEY=
@@ -27,6 +31,7 @@ services:
       # https://docs.djangoproject.com/en/2.1/ref/settings/#password
       # same as postgres password above
       # - DATABASE_PASSWORD=
+
   dms:
     image: ghcr.io/adfinis/document-merge-service:6.4.4
     depends_on:
@@ -42,6 +47,7 @@ services:
       - OIDC_BEARER_TOKEN_REVALIDATION_TIME=300
       - SECRET_KEY=aaa
       - DOCXTEMPLATE_JINJA_EXTENSIONS=
+
   minio:
     image: minio/minio:RELEASE.2023-11-06T22-26-08Z
     volumes:
@@ -54,6 +60,7 @@ services:
       - MINIO_ROOT_USER=very
       - MINIO_ROOT_PASSWORD=secret
     command: server data --console-address ":9090"
+
   mc:
     image: minio/mc:RELEASE.2023-11-06T04-19-23Z
     restart: on-failure
@@ -68,6 +75,12 @@ services:
     depends_on:
       - minio
 
+  tika:
+    image: apache/tika:2.9.2.0-full
+    ports:
+      - "9998:9998"
+
+
 volumes:
   dbdata:
   minio_data:

pyproject.toml

@@ -38,6 +38,7 @@ boto3 = "^1.29.7"
 tqdm = "^4.66.1"
 manabi = "^1.4.0"
 python-magic = "^0.4.27"
+tika = "^2.6.0"
 
 [tool.poetry.group.dev.dependencies]
 black = "24.4.2"