refine sample: export (microsoft#126)

docs/cloud/azureai/quick-start.md

@@ -7,7 +7,7 @@ This is an experimental feature, and may change at any time. Learn [more](https:
 Assuming you have learned how to create and run a flow following [Quick start](../../how-to-guides/quick-start.md). This guide will walk you through the main process of how to submit a promptflow run to [Azure AI](https://learn.microsoft.com/en-us/azure/machine-learning/prompt-flow/overview-what-is-prompt-flow?view=azureml-api-2).
 
 Benefits of use Azure AI comparison to just run locally:
-- **Designed for team collaboration**: Portal UI is a better fix for sharing & presentation your flow and runs. And workspace can better orgnize team shared resources like connections.
+- **Designed for team collaboration**: Portal UI is a better fix for sharing & presentation your flow and runs. And workspace can better organize team shared resources like connections.
 - **Enterprise Readiness Solutions**: prompt flow leverages Azure AI's robust enterprise readiness solutions, providing a secure, scalable, and reliable foundation for the development, experimentation, and deployment of flows.
 
 ## Prerequisites
@@ -116,7 +116,6 @@ ml_client = MLClient(
     resource_group_name="<RESOURCE_GROUP>",
     workspace_name="<AML_WORKSPACE_NAME>",
 )
-# configure global setting pointing to workpsace ml_client
 pf = PFClient(ml_client)
 ```
 

docs/how-to-guides/manage-connections.md

@@ -98,7 +98,7 @@ print(result)
 ::::{tab-set}
 :::{tab-item} CLI
 :sync: CLI
-The commands below show how to update exisiting connections with new values:
+The commands below show how to update existing connections with new values:
 ```bash
 # Update an azure open ai connection with a new api base
 pf connection update -n my_azure_open_ai_connection --set api_base='new_value'
@@ -110,7 +110,7 @@ pf connection update -n my_custom_connection --set configs.other_config='new_val
 
 :::{tab-item} SDK
 :sync: SDK
-The code snippet below shows how to update exisiting connections with new values:
+The code snippet below shows how to update existing connections with new values:
 ```python
 # Update an azure open ai connection with a new api base
 connection = pf.connections.get(name="my_azure_open_ai_connection")

docs/reference/tools-reference/embedding_tool.md

@@ -13,7 +13,7 @@ Create OpenAI resources:
 
 - **Azure OpenAI (AOAI)**
 
-    Create Azure OpenAI resources with [insturction](https://learn.microsoft.com/en-us/azure/cognitive-services/openai/how-to/create-resource?pivots=web-portal)
+    Create Azure OpenAI resources with [instruction](https://learn.microsoft.com/en-us/azure/cognitive-services/openai/how-to/create-resource?pivots=web-portal)
 
 ## **Connections**
 
@@ -22,7 +22,7 @@ Setup connections to provide resources in embedding tool.
 | Type        | Name     | API KEY  | API Type | API Version |
 |-------------|----------|----------|----------|-------------|
 | OpenAI      | Required | Required | -        | -           |
-| AzureOpenAI | Required | Requried | Required | Required    |
+| AzureOpenAI | Required | Required | Required | Required    |
 
 
 ## Inputs

docs/reference/tools-reference/vector_db_lookup_tool.md

@@ -1,6 +1,6 @@
 # Vector DB Lookup
 
-Vector DB Lookup is a vector search tool that allows users to search top k similiar vectors from vector database. This tool is a wrapper for multiple third-party vector databases. The list of current supported databases is as follows.
+Vector DB Lookup is a vector search tool that allows users to search top k similar vectors from vector database. This tool is a wrapper for multiple third-party vector databases. The list of current supported databases is as follows.
 
 | Name | Description |
 | --- | --- |

examples/flows/standard/web-classification/README.md

@@ -67,7 +67,7 @@ pf run show-details --name $run_name
 
 create `evaluation` run:
 ```bash
-# (Optional) save previous run name into variable, and create a new random run name for furthur use
+# (Optional) save previous run name into variable, and create a new random run name for further use
 prev_run_name=$run_name
 run_name="classification_accuracy_"$(openssl rand -hex 12)
 # create run using command line args
@@ -93,7 +93,7 @@ az configure --defaults group=<your_resource_group_name> workspace=<your_workspa
 pfazure run create --flow . --data ./data.jsonl --stream --runtime demo-mir --subscription <your_subscription_id> -g <your_resource_group_name> -w <your_workspace_name>
 # pfazure run create --flow . --data ./data.jsonl --stream # serverless compute
 
-# (Optional) create a new random run name for furthur use
+# (Optional) create a new random run name for further use
 run_name="web_classification_"$(openssl rand -hex 12)
 
 # create run using yaml file, --name is optional
@@ -106,7 +106,7 @@ pfazure run show-details --name $run_name
 pfazure run show-metrics --name $run_name
 
 
-# (Optional) save previous run name into variable, and create a new random run name for furthur use
+# (Optional) save previous run name into variable, and create a new random run name for further use
 prev_run_name=$run_name
 run_name="classification_accuracy_"$(openssl rand -hex 12)
 

examples/requirements.txt

@@ -1,6 +1,6 @@
 # remove when we publish to pypi
 --extra-index-url https://azuremlsdktestpypi.azureedge.net/promptflow/
-promptflow[azure]==0.0.101903259
+promptflow[azure]==0.0.102010695
 promptflow-tools==0.1.0.b3
 python-dotenv
 langchain

examples/tutorials/flow-deploy/.gitignore

@@ -1,2 +1,4 @@
 *.sqlite
-linux/flow
+linux/flow
+linux/connections
+linux/settings.json

examples/tutorials/flow-deploy/deploy.md

@@ -33,9 +33,6 @@ The following CLI commands allows you export a flow as a sharable folder with a
 pf flow export --source ../../flows/standard/basic-with-connection --output <your-output-dir> --format docker
 ```
 
-You'll be asked to input a migration secret when running this command, which needs to be provided when you run the built docker image.
-You can also provide the key via `--migration-secret` directly or passing it with a file via `--migration-secret-file`.
-
 More details about how to use the exported docker can be seen in `<your-output-dir>/README.md`. 
 Part of sample output are under [./linux](./linux/) so you can also check [this README](./linux/README.md) directly.
 

examples/tutorials/flow-deploy/linux/Dockerfile

@@ -11,15 +11,14 @@ ENV CONDA_DEFAULT_ENVIRONMENT=$CONDA_ENVIRONMENT_PATH
 ENV PATH $CONDA_DEFAULT_ENVIRONMENT/bin:$PATH
 RUN conda create -n promptflow-serve python=3.9.16 pip=23.0.1 -q -y && \
     conda run -n promptflow-serve \
-    pip install -r /flow/requirements_txt && \
+    pip install -r /flow/requirements.txt && \
+    conda run -n promptflow-serve pip install keyrings.alt && \
     conda run -n promptflow-serve pip cache purge && \
     conda clean -a -y
 
-RUN conda run -n promptflow-serve sh /flow/setup.sh
 
 EXPOSE 8080
 
-COPY ./connections.sqlite /
-COPY ./connections_setup.py /
+COPY ./connections/* /connections/
 COPY ./start.sh /
-CMD ./start.sh
+CMD ["bash", "./start.sh"]

examples/tutorials/flow-deploy/linux/README.md

@@ -5,11 +5,12 @@
 Exported Dockerfile & its dependencies are located in the same folder. The structure is as below:
 - flow: the folder contains all the flow files
   - ...
+- connections: the folder contains yaml files to create all related connections
+  - ...
 - Dockerfile: the dockerfile to build the image
-- connections.sqlite: the sqlite database file to store the connections used in the flow
-- connections_setup.py: the python script to migrate the connections used in the flow
 - start.sh: the script used in `CMD` of `Dockerfile` to start the service
-- docker-compose.yaml: a sample compose file to run the service
+- deploy.sh & deploy.ps1: the script to deploy the docker image to Azure App Service
+- settings.json: a json file to store the settings of the docker image
 - README.md: the readme file to describe how to use the dockerfile
 
 ## Build Docker image
@@ -27,89 +28,58 @@ docker build . -t promptflow-serve
 Run the docker image will start a service to serve the flow inside the container. Service will listen on port 8080.
 You can map the port to any port on the host machine as you want.
 
-If the service involves connections, you need to migrate the connections before the first request to the service.
-Given api_key in connections are secrets, we have provided a way to migrate them with the `migration-secret` you have provided in export command.
-
-### Manage migration-secret with `docker-secret`
+If the service involves connections, all related connections will be exported as yaml files and recreated in containers.
 
-As a pre-requirement of connections migration, you need to create a docker secret named `MIGRATION_SECRET`
-to store the migration secret first. Sample command is like below:
+Secrets in connections won't be exported directly. Instead, we will export them as a reference to environment variables:
 
-```bash
-#### Init host machine as a swarm manager
-docker swarm init
-#### Create a secret to store the migration secret
-# You can create the docker secret directly from shell with bash
-(read -sp "Enter your migration secret: "; echo $REPLY) | docker secret create MIGRATION_SECRET -
-# or you can also create secret from a local file with migration secret as its content, like in Powershell
-docker secret create MIGRATION_SECRET <migration_secret_file>
+```yaml
+configs:
+  AZURE_OPENAI_API_BASE: xxx
+  CHAT_DEPLOYMENT_NAME: xxx
+module: promptflow.connections
+name: custom_connection
+secrets:
+  AZURE_OPENAI_API_KEY: ${env:<connection-name>_<secret-key>}
+type: custom
 ```
 
-You can check below documents for more details:
-- [Swam mode overview](https://docs.docker.com/engine/swarm/)
-- [Secrets management](https://docs.docker.com/engine/swarm/secrets/)
+You'll need to set up the environment variables in the container to make the connections work.
 
-### Run with `docker-service` [WIP]
+### Deploy with Azure App Service
 
-To avoid manually migrate the connections, you can use `docker-secret` to manage the migration secret
-and `docker-service` to run the service:
+Azure App Service is an HTTP-based service for hosting web applications, REST APIs, and mobile back ends.
+Promptflow has provided a [script](./deploy.sh) to help deploy the docker image to Azure App Service.
 
+Example command to use bash script:
 ```bash
-#### Start the service
-docker service create --name promptflow-service -p 8080:8080 --secret MIGRATION_SECRET promptflow-serve
+bash deploy.sh -i <image_tag> -r "promptflow.azurecr.io" -g <resource_group>
 ```
 
-You can check below documents for more details:
-- [Run Docker Engine in swarm mode](https://docs.docker.com/engine/swarm/swarm-mode/)
+Example command to use powershell script:
+```powershell
+.\deploy.ps1 -i <image_tag> -r "promptflow.azurecr.io" -g <resource_group>
+```
+
+See the full parameters by `bash deploy.sh -h` or `.\deploy.ps1 -h`.
 
-### Run with `docker-compose`
+### Run with `docker run`
 
-You can also use `docker-secret` in your compose file and use compose file to start your service:
+You can run the docker image directly set via below commands:
 
 ```bash
-#### Deploy the service
-docker stack deploy --compose-file=./docker-compose.yaml service1
+docker run -p 8080:8080 -e <connection-name>_<secret-key>=<secret-value> promptflow-serve
 ```
 
-Note that you need to deploy the service to a swarm cluster to use `docker-secret`.
-So connections won't be migrated successfully if you run `docker-compose` directly.
-More details can be found in the official document:
-- [Deploy a stack to a swarm](https://docs.docker.com/engine/swarm/stack-deploy/)
-
-In the sample compose file `docker-compose.yaml` in the output directory, we claim secret `MIGRATION_SECRET`
-as external, which means you need to create the secret first before running the compose file.
+As explain in previously, secrets in connections will be passed to container via environment variables.
+You can set up multiple environment variables for multiple connection secrets:
 
-You can also specify the migration secret file and docker image in the compose file:
-
-```yaml
-services:
-  promptflow:
-    image: <your-image>
-...
-secrets:
-  MIGRATION_SECRET:
-    file: <your-migration-secret-file>
+```bash
+docker run -p 8080:8080 -e <connection-name-1>_<secret-key>=<secret-value-1> -e <connection-name-2>_<secret-key>=<secret-value-2> promptflow-serve
 ```
 
-Official document:
-- [Manage secrets in Docker Compose](https://docs.docker.com/compose/compose-file/compose-file-v3/#secrets)
-- [Using secrets in Compose](https://docs.docker.com/compose/use-secrets/)
-
 ## Test the endpoint
 After start the service, you can use curl to test it:
 
 ```bash
 curl http://localhost:8080/score --data '{"text":"Hello world!"}' -X POST  -H "Content-Type: application/json"
 ```
-
-## Advanced: Run with `docker run`
-
-If you want to debug or have provided a wrong migration-secret, you can run the docker image directly and manually migrate the connections via below commands:
-
-```bash
-docker run -p 8080:8080 promptflow-serve
-#### Migrate connections
-docker exec -it <container_id> python connections_setup.py --file /connections.sqlite --migration-secret <migration_secret> --clean
-```
-
-Note that the command to migrate the connections must be run before any requests to the service.