Tweak run script

.github/pull_request_template.md

@@ -1,14 +1,22 @@
 ## What
 
-...
+-
 
 ## Why
 
-...
+-
 
 ## How
 
-...
+-
+
+## Database Migrations
+
+- 
+
+## Env Config
+
+- 
 
 ## Relevant Docs
 
@@ -18,18 +26,16 @@
 
 -
 
-## Dependencies Versions / Env Variables
+## Dependencies Versions
 
 -
 
 ## Notes on Testing
 
-...
+-
 
 ## Screenshots
 
-...
-
 ## Checklist
 
 I have read and understood the [Contribution Guidelines]().

README.md

@@ -15,7 +15,7 @@ Use LLMs to eliminate manual processes involving unstructured data.
 
 Just run the `run-platform.sh` launch script to get started in few minutes.
 
-The launch script does env setup with default values, pulls public Docker images or builds them locally and finally runs them in containers.
+The launch script does env setup with default values, pulls public docker images or builds them locally and finally runs them in containers.
 
 ```bash
 # Pull and run entire Unstract platform with default env config.
@@ -24,19 +24,22 @@ The launch script does env setup with default values, pulls public Docker images
 # Pull and run docker containers with a specific version tag.
 ./run-platform.sh -v v0.1.0
 
-# Build docker images and run as containers with a specific version tag.
-./run-platform.sh -B -v v0.1.0
+# Build docker images locally and run with a specific version tag.
+./run-platform.sh -b -v v0.1.0
 
 # Display the help information.
 ./run-platform.sh -h
 
 # Only do setup of environment files.
 ./run-platform.sh -e
 
-# Only do docker images pull/build with a specific version tag.
-./run-platform.sh -b v0.1.0
+# Only do docker images pull with a specific version tag.
+./run-platform.sh -p -v v0.1.0
 
-# Pull and run Docker containers in detached mode.
+# Only do docker images pull by building locally with a specific version tag.
+./run-platform.sh -p -b -v v0.1.0
+
+# Pull and run docker containers in detached mode.
 ./run-platform.sh -d -v v0.1.0
 ```
 

docker/docker-compose.yaml

@@ -17,6 +17,9 @@ services:
       - redis
       - reverse-proxy
       - minio
+      - platform-service
+      - prompt-service
+      - x2text-service
     volumes:
       - ./workflow_data:/data
     environment:
@@ -140,6 +143,7 @@ services:
       - ../worker/.env
     volumes:
       - ./workflow_data:/data
+      # Docker socket bind mount to spawn tool containers
       - /var/run/docker.sock:/var/run/docker.sock
     depends_on:
       - redis

docker/dockerfiles/document.Dockerfile

@@ -24,13 +24,17 @@ RUN DEBIAN_FRONTEND=noninteractive apt-get update; \
         unoconv; \
     apt-get clean && rm -rf /var/lib/apt/lists/* /var/cache/apt/archives/*; \
     \
-    pip install --no-cache-dir -U pip pdm~=${PDM_VERSION};
+    pip install --no-cache-dir -U pip pdm~=${PDM_VERSION}; \
+    \
+    # Creates a non-root user with an explicit UID and adds permission to access the /app folder
+    # For more info, please refer to https://aka.ms/vscode-docker-python-configure-containers
+    adduser -u 5678 --disabled-password --gecos "" unstract;
 
-EXPOSE 3002
+USER unstract
 
 WORKDIR /app
 
-COPY ${BUILD_CONTEXT_PATH}/ .
+COPY --chown=unstract ${BUILD_CONTEXT_PATH}/ .
 
 RUN set -e; \
     \
@@ -45,13 +49,10 @@ RUN set -e; \
     # REF: https://docs.gunicorn.org/en/stable/deploy.html#using-virtualenv
     pip install --no-cache-dir gunicorn; \
     \
+    # Storage for document uploads and processing
     mkdir /app/uploads /app/processed;
 
-# Creates a non-root user with an explicit UID and adds permission to access the /app folder
-# For more info, please refer to https://aka.ms/vscode-docker-python-configure-containers
-RUN adduser -u 5678 --disabled-password --gecos "" unstract;
-
-USER unstract
+EXPOSE 3002
 
 # Wrapper to run both python server and libreoffice.
 CMD [ "/app/wrapper.sh" ]

docker/dockerfiles/platform.Dockerfile

@@ -12,13 +12,19 @@ ENV BUILD_CONTEXT_PATH platform-service
 ENV BUILD_PACKAGES_PATH unstract
 ENV PDM_VERSION 2.12.3
 
-RUN pip install --no-cache-dir -U pip pdm~=${PDM_VERSION};
+RUN pip install --no-cache-dir -U pip pdm~=${PDM_VERSION}; \
+    \
+    # Creates a non-root user with an explicit UID and adds permission to access the /app folder
+    # For more info, please refer to https://aka.ms/vscode-docker-python-configure-containers
+    adduser -u 5678 --disabled-password --gecos "" unstract;
+
+USER unstract
 
 WORKDIR /app
 
-COPY ${BUILD_CONTEXT_PATH} .
+COPY --chown=unstract ${BUILD_CONTEXT_PATH} .
 # Copy local dependency packages
-COPY ${BUILD_PACKAGES_PATH} /unstract
+COPY --chown=unstract ${BUILD_PACKAGES_PATH} /unstract
 
 RUN set -e; \
     \
@@ -35,11 +41,5 @@ RUN set -e; \
 
 EXPOSE 3001
 
-# Creates a non-root user with an explicit UID and adds permission to access the /app folder
-# For more info, please refer to https://aka.ms/vscode-docker-python-configure-containers
-RUN adduser -u 5678 --disabled-password --gecos "" unstract;
-
-USER unstract
-
 # During debugging, this entry point will be overridden. For more information, please refer to https://aka.ms/vscode-docker-python-debug
 CMD [".venv/bin/gunicorn", "--bind", "0.0.0.0:3001", "--timeout", "300", "unstract.platform_service.main:app"]

docker/dockerfiles/prompt.Dockerfile

@@ -19,11 +19,17 @@ RUN apt-get update; \
         git; \
     apt-get clean && rm -rf /var/lib/apt/lists/* /var/cache/apt/archives/*; \
     \
-    pip install --no-cache-dir -U pip pdm~=${PDM_VERSION};
+    pip install --no-cache-dir -U pip pdm~=${PDM_VERSION}; \
+    \
+    # Creates a non-root user with an explicit UID and adds permission to access the /app folder
+    # For more info, please refer to https://aka.ms/vscode-docker-python-configure-containers
+    adduser -u 5678 --disabled-password --gecos "" unstract;
+
+USER unstract
 
 WORKDIR /app
 
-COPY ${BUILD_CONTEXT_PATH} .
+COPY --chown=unstract ${BUILD_CONTEXT_PATH} .
 
 RUN set -e; \
     \
@@ -61,10 +67,10 @@ RUN set -e; \
     #
     \
     # REF: https://docs.gunicorn.org/en/stable/deploy.html#using-virtualenv
-    pip install --no-cache-dir gunicorn;
-
-# Storage for prompt studio uploads
-RUN mkdir prompt-studio-data
+    pip install --no-cache-dir gunicorn; \
+    \
+    # Storage for prompt studio uploads
+    mkdir prompt-studio-data;
 
 EXPOSE 3003
 

docker/dockerfiles/worker.Dockerfile

@@ -33,13 +33,8 @@ RUN set -e; \
     # REF: https://docs.gunicorn.org/en/stable/deploy.html#using-virtualenv
     pip install --no-cache-dir gunicorn gevent;
 
-
 EXPOSE 5002
 
-# Creates a non-root user with an explicit UID and adds permission to access the /app folder
-# For more info, please refer to https://aka.ms/vscode-docker-python-configure-containers
-RUN adduser -u 5678 --disabled-password --gecos "" unstract;
-
 # During debugging, this entry point will be overridden. For more information, please refer to https://aka.ms/vscode-docker-python-debug
 # The suggested maximum concurrent requests when using workers and threads is (2*CPU)+1
 CMD [ "./entrypoint.sh" ]

docker/dockerfiles/x2text.Dockerfile

@@ -10,11 +10,17 @@ ENV PYTHONUNBUFFERED 1
 ENV BUILD_CONTEXT_PATH x2text-service
 ENV PDM_VERSION 2.12.3
 
-RUN pip install --no-cache-dir -U pip pdm~=${PDM_VERSION}
+RUN pip install --no-cache-dir -U pip pdm~=${PDM_VERSION}; \
+    \
+    # Creates a non-root user with an explicit UID and adds permission to access the /app folder
+    # For more info, please refer to https://aka.ms/vscode-docker-python-configure-containers
+    adduser -u 5678 --disabled-password --gecos "" unstract;
+
+USER unstract
 
 WORKDIR /app
 
-COPY ${BUILD_CONTEXT_PATH} .
+COPY --chown=unstract ${BUILD_CONTEXT_PATH} .
 
 RUN set -e; \
     \
@@ -31,12 +37,6 @@ RUN set -e; \
 
 EXPOSE 3004
 
-# Creates a non-root user with an explicit UID and adds permission to access the /app folder
-# For more info, please refer to https://aka.ms/vscode-docker-python-configure-containers
-RUN adduser -u 5678 --disabled-password --gecos "" unstract;
-
-USER unstract
-
 # During debugging, this entry point will be overridden.
 # For more information, please refer to https://aka.ms/vscode-docker-python-debug
 CMD [".venv/bin/gunicorn", "--bind", "0.0.0.0:3004", "--timeout", "300", "run:app"]

run-platform.sh

@@ -42,12 +42,12 @@ display_help() {
   echo -e "Options:"
   echo -e "   -h, --help          Displays the help information"
   echo -e "   -e, --only-env      Only do env files setup"
-  echo -e "   -b, --only-build    Only do docker images build"
-  echo -e "   -B, --build-local   Build docker images locally"
+  echo -e "   -p, --only-pull     Only do docker images pull"
+  echo -e "   -b, --build-local   Build docker images locally"
   echo -e "   -d, --detach        Run docker containers in detached mode"
   echo -e "   -x, --trace         Enables trace mode"
   echo -e "   -V, --verbose       Print verbose logs"
-  echo -e "   -v, --version       Docker images version tag (default \"dev\")"
+  echo -e "   -v, --version       Docker images version tag (default \"latest\")"
   echo -e ""
 }
 
@@ -62,10 +62,10 @@ parse_args() {
       -e | --only-env)
         opt_only_env=true
         ;;
-      -b | --only-build)
-        opt_only_build=true
+      -p | --only-pull)
+        opt_only_pull=true
         ;;
-      -B | --build-local)
+      -b | --build-local)
         opt_build_local=true
         ;;
       -d | --detach)
@@ -92,7 +92,7 @@ parse_args() {
   done
 
   debug "OPTION only_env: $opt_only_env"
-  debug "OPTION only_build: $opt_only_build"
+  debug "OPTION only_pull: $opt_only_pull"
   debug "OPTION build_local: $opt_build_local"
   debug "OPTION detach: $opt_detach"
   debug "OPTION verbose: $opt_verbose"
@@ -163,7 +163,7 @@ build_services() {
 
   popd 1>/dev/null
 
-  if [ "$opt_only_build" = true ]; then
+  if [ "$opt_only_pull" = true ]; then
     echo "Done." && exit 0
   fi
 }
@@ -192,11 +192,11 @@ if ! command -v docker compose &> /dev/null; then
 fi
 
 opt_only_env=false
-opt_only_build=false
+opt_only_pull=false
 opt_build_local=false
 opt_detach=""
 opt_verbose=false
-opt_version="dev"
+opt_version="latest"
 
 script_dir=$(dirname "$(readlink -f "$BASH_SOURCE")")
 # Extract service names from docker compose file.