RESTful API Example Pipeline - What is the proper way to pass the list of valid_languages in a POST request?

[msarro]

I am working to modify the example RESTful API tutorial to work with opensearch:

• Tutorial: https://docs.haystack.deepset.ai/docs/rest_api
• Pipeline File: https://github.com/deepset-ai/haystack/blob/main/rest_api/rest_api/pipeline/pipelines.haystack-pipeline.yml

Here's my docker-compose.yml:

version: "3"

volumes:
  opensearch-data1:


services:
  opensearch-node:
    image: opensearchproject/opensearch:latest
    container_name: opensearch-node
    environment:
      - discovery.type=single-node
      - "DISABLE_INSTALL_DEMO_CONFIG=true"
      - "DISABLE_SECURITY_PLUGIN=true"
    volumes:
      - type: bind
        source: ./opensearch-data1
        target: /usr/share/opensearch/data # Creates volume called opensearch-data1 and mounts it to the container

    ulimits:
      memlock:
        soft: -1
        hard: -1
    ports:
      - 9200:9200
      - 9600:9600
    expose:
      - "9200"
    healthcheck:
      test:
        [
          "CMD",
          "curl",
          "--fail",
          "http://localhost:9200/_cat/health"
        ]
      interval: 10s
      timeout: 1s
      retries: 10

  opensearch-dashboards:
    image: opensearchproject/opensearch-dashboards:latest
    container_name: opensearch-dashboards
    ports:
      - 5601:5601
    expose:
      - "5601"
    environment:
      - "DISABLE_SECURITY_DASHBOARDS_PLUGIN=true"
      - "OPENSEARCH_HOSTS=http://opensearch-node:9200"
    depends_on:
      opensearch-node:
        condition: service_healthy

  haystack-api:
    image: "deepset/haystack:cpu-main"
    container_name: haystack-api
    volumes:
      - ./rest_api/rest_api/pipeline:/opt/pipelines
    ports:
      - 8000:8000
    expose:
      - "8000"
    restart: on-failure
    environment:
      - "DOCUMENTSTORE_PARAMS_HOST=http://opensearch-node:9200"
      - "PIPELINE_YAML_PATH=/opt/pipelines/pipelines.haystack-pipeline.yml"
      - "TOKENIZERS_PARALLELISM=false"
      # Uncomment the following line to customise how much time (in seconds) a worker can spend serving a request
      # before it times out. This should include the time required to cache the models and setup the pipelines.

      - GUNICORN_CMD_ARGS="--timeout=60"
    depends_on:
      opensearch-node:
        condition: service_healthy

Here's my tweaked pipeline file:

# To allow your IDE to autocomplete and validate your YAML pipelines, name them as <name of your choice>.haystack-pipeline.yml

version: ignore

components:    # define all the building-blocks for Pipeline
  - name: OpenSearchDocumentStore
    type: OpenSearchDocumentStore
    params:
      host: opensearch-node
      port: 9200
      scheme: http
  - name: Retriever
    type: BM25Retriever
    params:
      document_store: OpenSearchDocumentStore   # params can reference other components defined in the YAML
      top_k: 5
  - name: Reader       # custom-name for the component; helpful for visualization & debugging
    type: FARMReader    # Haystack Class name for the component
    params:
      model_name_or_path: deepset/roberta-base-squad2
      context_window_size: 500
      return_no_answer: true
  - name: TextFileConverter
    type: TextConverter
  - name: PDFFileConverter
    type: PDFToTextConverter
  - name: MarkdownFileConverter
    type: MarkdownConverter
  - name: Preprocessor
    type: PreProcessor
    params:
      split_by: word
      split_length: 1000
  - name: FileTypeClassifier
    type: FileTypeClassifier

pipelines:
  - name: query    # a sample extractive-qa Pipeline
    nodes:
      - name: Retriever
        inputs: [Query]
      - name: Reader
        inputs: [Retriever]
  - name: indexing
    nodes:
      - name: FileTypeClassifier
        inputs: [File]
      - name: TextFileConverter
        inputs: ["FileTypeClassifier.output_1"]
      - name: PDFFileConverter
        inputs: ["FileTypeClassifier.output_2"]
      - name: MarkdownFileConverter
        inputs: ["FileTypeClassifier.output_3"]
      - name: Preprocessor
        inputs: [PDFFileConverter, TextFileConverter, MarkdownFileConverter]
      - name: Retriever
        inputs: [Preprocessor]
      - name: OpenSearchDocumentStore
        inputs: [Retriever]

Using that docker compose seems to work.

In fact, everything works except the valid_languages parameter.

From the docs, it says this should just be a python list, populated with values in ISO language format (ex: ["en"]).

However, it seems like no matter what value I pass for the list, I get a pydantic validation error similar to the following:

Traceback (most recent call last):
  File "/opt/venv/lib/python3.10/site-packages/starlette/middleware/errors.py", line 162, in __call__
    await self.app(scope, receive, _send)
  File "/opt/venv/lib/python3.10/site-packages/starlette/middleware/cors.py", line 83, in __call__
    await self.app(scope, receive, send)
  File "/opt/venv/lib/python3.10/site-packages/starlette/middleware/exceptions.py", line 79, in __call__
    raise exc
  File "/opt/venv/lib/python3.10/site-packages/starlette/middleware/exceptions.py", line 68, in __call__
    await self.app(scope, receive, sender)
  File "/opt/venv/lib/python3.10/site-packages/fastapi/middleware/asyncexitstack.py", line 20, in __call__
    raise e
  File "/opt/venv/lib/python3.10/site-packages/fastapi/middleware/asyncexitstack.py", line 17, in __call__
    await self.app(scope, receive, send)
  File "/opt/venv/lib/python3.10/site-packages/starlette/routing.py", line 718, in __call__
    await route.handle(scope, receive, send)
  File "/opt/venv/lib/python3.10/site-packages/starlette/routing.py", line 276, in handle
    await self.app(scope, receive, send)
  File "/opt/venv/lib/python3.10/site-packages/starlette/routing.py", line 66, in app
    response = await func(request)
  File "/opt/venv/lib/python3.10/site-packages/fastapi/routing.py", line 231, in app
    solved_result = await solve_dependencies(
  File "/opt/venv/lib/python3.10/site-packages/fastapi/dependencies/utils.py", line 622, in solve_dependencies
    solved = await call(**sub_values)
  File "/opt/venv/lib/python3.10/site-packages/rest_api/controller/utils.py", line 44, in _as_form
    return cls(**data)
  File "pydantic/main.py", line 341, in pydantic.main.BaseModel.__init__
pydantic.error_wrappers.ValidationError: 1 validation error for FileConverterParams
valid_languages
  value is not a valid list (type=type_error.list)

All of the other fields can be passed into swagger or via the API without an issue. But as soon as I try to put a value for valid_languages I get the validation error.

Attempted values for valid_languages:

• ["en"]
• ["en",]
• ["en","de"]
• ["en","de",]
• %5B%22en%22%5D

Nothing seems to pass validation.

Is there some specific way I would have to encode ["en"] to pass the pydantic validation?

[ZanSara]

Interesting... Can you share the exact way you send this API request? I still hope that might be just a syntax error somewhere. Alternatively you can also pass the list of valid languages in the YAML file, if that's an option.

[msarro]

My pleasure.

Here's a screenshot from postman:

And here is the postman request in curl format:

curl --location 'http://127.0.0.1:8000/file-upload' \
--form 'files=@"/Users/msarro/Downloads/RFC-all/rfc1.txt"' \
--form 'meta="null"' \
--form 'additional_params="null"' \
--form 'remove_numeric_tables="True"' \
--form 'valid_languages="[\"en\"]"' \
--form 'clean_whitespace="True"' \
--form 'clean_empty_lines="True"' \
--form 'clean_header_footer="True"' \
--form 'split_by="word"' \
--form 'split_length="20"' \
--form 'split_overlap="6"' \
--form 'split_respect_sentence_boundary="True"'

Attempting this gives the same 500 error:

Traceback (most recent call last):
  File "/opt/venv/lib/python3.10/site-packages/starlette/middleware/errors.py", line 162, in __call__
    await self.app(scope, receive, _send)
  File "/opt/venv/lib/python3.10/site-packages/starlette/middleware/cors.py", line 83, in __call__
    await self.app(scope, receive, send)
  File "/opt/venv/lib/python3.10/site-packages/starlette/middleware/exceptions.py", line 79, in __call__
    raise exc
  File "/opt/venv/lib/python3.10/site-packages/starlette/middleware/exceptions.py", line 68, in __call__
    await self.app(scope, receive, sender)
  File "/opt/venv/lib/python3.10/site-packages/fastapi/middleware/asyncexitstack.py", line 20, in __call__
    raise e
  File "/opt/venv/lib/python3.10/site-packages/fastapi/middleware/asyncexitstack.py", line 17, in __call__
    await self.app(scope, receive, send)
  File "/opt/venv/lib/python3.10/site-packages/starlette/routing.py", line 718, in __call__
    await route.handle(scope, receive, send)
  File "/opt/venv/lib/python3.10/site-packages/starlette/routing.py", line 276, in handle
    await self.app(scope, receive, send)
  File "/opt/venv/lib/python3.10/site-packages/starlette/routing.py", line 66, in app
    response = await func(request)
  File "/opt/venv/lib/python3.10/site-packages/fastapi/routing.py", line 231, in app
    solved_result = await solve_dependencies(
  File "/opt/venv/lib/python3.10/site-packages/fastapi/dependencies/utils.py", line 622, in solve_dependencies
    solved = await call(**sub_values)
  File "/opt/venv/lib/python3.10/site-packages/rest_api/controller/utils.py", line 44, in _as_form
    return cls(**data)
  File "pydantic/main.py", line 341, in pydantic.main.BaseModel.__init__
pydantic.error_wrappers.ValidationError: 1 validation error for FileConverterParams
valid_languages
  value is not a valid list (type=type_error.list)

Removing the languages list gets a 200 ok.

To add it into the yaml file, would I just add something like this?

  - name: Preprocessor
    type: PreProcessor
    params:
      split_by: word
      split_length: 1000
      valid_languages=["en"]

[msarro]

Following up, trying to include it in the pipeline also doesn't seem to be helping.

I've tried the following formats:

  - name: Preprocessor
    type: PreProcessor
    params:
      split_by: word
      split_length: 20
      split_overlap: 6
      valid_languages:
        - en

  - name: Preprocessor
    type: PreProcessor
    params:
      split_by: word
      split_length: 20
      split_overlap: 6
      valid_languages: ["en"]

  - name: Preprocessor
    type: PreProcessor
    params:
      split_by: word
      split_length: 20
      split_overlap: 6
      valid_languages: en

In each case I get the following error in the haystack-api container at boot-time:

2023-06-14 09:44:01 haystack.errors.PipelineSchemaError: Node of type PreProcessor found, but it failed validation. Possible causes:
2023-06-14 09:44:01  - The node is missing some mandatory parameter
2023-06-14 09:44:01  - Wrong indentation of some parameter in YAML
2023-06-14 09:44:01 See the stacktrace for more information.

Although looking at the docs it looks like valid_languages is a param for BaseConverter rather than preprocessor, so am I just putting it at the wrong spot?