[A-00092]pythonでDocument AI API(GCP)を使用する

Google Cloudが提供するDocument AI APIをpythonから使用する方法を記載する。

公式ドキュメントはこちら

https://cloud.google.com/document-ai/docs/libraries

・セットアップ

pythonのライブラリをインストールします。

pip install --upgrade google-cloud-documentai

・Document AIを使用する前にProcessor-Typeを調べる

Document AIを使用する前に自分の使用するProcessor Typeを決める必要があります。

Processor-Typeを調べる方法は下記のとおりです。

from google.api_core.client_options import ClientOptions
from google.cloud import documentai  # type: ignore


if __name__ == "__main__":
    client = documentai.DocumentProcessorServiceClient()
    # Initialize request argument(s)
    request = documentai.FetchProcessorTypesRequest(
        parent="projects/<project_id>/locations/us",
    )

    # Make the request
    response = client.fetch_processor_types(request=request)
    print(response)

上記で取得したProcessor-Typeから使用するドキュメントに合わせて選択して下さい。

今回の場合はPDFファイルなので[OCR_PROCESSOR]で読み取りします。

・ドキュメントを読み込む(新規Processorを作成する)

使用するドキュメントを下記のように適当なローカルフォルダに格納してください。

今回テスト用で使用するファイルは下記です。

下記のようにローカルの適当なディレクトリに格納してください。

from google.api_core.client_options import ClientOptions
from google.cloud import documentai  # type: ignore


def quickstart(
    project_id: str,
    location: str,
    processor_display_name: str,
    processor_type: str,
    file_path: str,
    mime_type: str,
):
    # You must set the `api_endpoint`if you use a location other than "us".
    opts = ClientOptions(api_endpoint=f"{location}-documentai.googleapis.com")

    client = documentai.DocumentProcessorServiceClient(client_options=opts)

    # The full resource name of the location, e.g.:
    parent = client.common_location_path(project_id, location)

    # Create a Processor
    processor = client.create_processor(
        parent=parent,
        processor=documentai.Processor(
            display_name=processor_display_name, type_=processor_type
        ),
    )

    # Print the processor information
    print(f"Processor Name: {processor.name}")

    # Read the file into memory
    with open(file_path, "rb") as image:
        image_content = image.read()

    # Load binary data
    raw_document = documentai.RawDocument(content=image_content, mime_type=mime_type)

    # Configure the process request
    request = documentai.ProcessRequest(name=processor.name, raw_document=raw_document)

    result = client.process_document(request=request)
    document = result.document

    # Read the text recognition output from the processor
    print("The document contains the following text:")
    print(document.text)
    # [END documentai_quickstart]
    return processor

def main():
    quickstart("<project_id>"
               , "us"
               , "MyProcessorsOCR"
               , "OCR_PROCESSOR"
               , "<path_to_pdf>/test.pdf"
               , "application/pdf")
    

if __name__ == "__main__":
    main()
    # end main

(.venv) MacBook-Pro:test2_project$ python3 documentaitest.py 
Processor Name: projects/xxxxxxx/locations/us/processors/xxxxxxx
The document contains the following text:
備考欄
下記のとおりご請求申し上げます。
お振込先
お支払い期限
御中
品名
請求書
0
数量 単位
所在地:
TEL:
担当:
請求書発行日:
請求番号:
円 (消費税込)
単価
小 計
税率
消費税
合計
金額
10%
0
0
0
0
0
0
0
0
確定

上記の通り、Document AIでpdfファイルがうまく読み込めました。

・カスタムプロセッサのAPIでドキュメントを分析する。

詳しくは下記の記事に書いてます。

[A-00100]DocumentAIのCustomProcessorを作成する

ソースコードは下記です。

from typing import Optional

from google.api_core.client_options import ClientOptions
from google.cloud import documentai  # type: ignore

# TODO(developer): Uncomment these variables before running the sample.
# project_id = "YOUR_PROJECT_ID"
# location = "YOUR_PROCESSOR_LOCATION" # Format is "us" or "eu"
# processor_id = "YOUR_PROCESSOR_ID" # Create processor before running sample
# file_path = "/path/to/local/pdf"
# mime_type = "application/pdf" # Refer to https://cloud.google.com/document-ai/docs/file-types for supported file types
# field_mask = "text,entities,pages.pageNumber"  # Optional. The fields to return in the Document object.
# processor_version_id = "YOUR_PROCESSOR_VERSION_ID" # Optional. Processor version to use


def process_document_sample(
    project_id: str,
    location: str,
    processor_id: str,
    file_path: str,
    mime_type: str,
    field_mask: Optional[str] = None,
    processor_version_id: Optional[str] = None,
) -> None:
    # You must set the `api_endpoint` if you use a location other than "us".
    opts = ClientOptions(api_endpoint=f"{location}-documentai.googleapis.com")

    client = documentai.DocumentProcessorServiceClient(client_options=opts)

    if processor_version_id:
        # The full resource name of the processor version, e.g.:
        # `projects/{project_id}/locations/{location}/processors/{processor_id}/processorVersions/{processor_version_id}`
        name = client.processor_version_path(
            project_id, location, processor_id, processor_version_id
        )
    else:
        # The full resource name of the processor, e.g.:
        # `projects/{project_id}/locations/{location}/processors/{processor_id}`
        name = client.processor_path(project_id, location, processor_id)

    # Read the file into memory
    with open(file_path, "rb") as image:
        image_content = image.read()

    # Load binary data
    raw_document = documentai.RawDocument(content=image_content, mime_type=mime_type)

    # Configure the process request
    request = documentai.ProcessRequest(
        name=name, raw_document=raw_document, field_mask=field_mask
    )

    result = client.process_document(request=request)

    # For a full list of `Document` object attributes, reference this page:
    # https://cloud.google.com/document-ai/docs/reference/rest/v1/Document
    document = result.document

    # Read the text recognition output from the processor
    with open("out.json", "w") as f:
        print("The document contains the following text:")
        print(document.entities, file=f)


if __name__ == "__main__":
    process_document_sample(
        "<project_id>",
        "us",
        "<processor_id>",
        "<file_path>",
        "image/jpeg",
        None,
        "<processor_version_id>"
    )