azure_document_intelligence_layout_analysis_from_url.py

1import os
2from azure.core.credentials import AzureKeyCredential
3from azure.ai.documentintelligence import DocumentIntelligenceClient
4from azure.ai.documentintelligence.models import AnalyzeResult, AnalyzeDocumentRequest
5
6# Set your variables
7endpoint = "YOUR_DOCUMENT_INTELLIGENCE_ENDPOINT"
8key = "YOUR_DOCUMENT_INTELLIGENCE_KEY"
9
10def analyze_layout():
11    # Sample document URL
12    formUrl = "https://raw.githubusercontent.com/Azure-Samples/cognitive-services-REST-api-samples/master/curl/form-recognizer/sample-layout.pdf"
13
14    # Create the client
15    document_intelligence_client = DocumentIntelligenceClient(
16        endpoint=endpoint, credential=AzureKeyCredential(key)
17    )
18
19    # Start the analysis
20    poller = document_intelligence_client.begin_analyze_document(
21        "prebuilt-layout", AnalyzeDocumentRequest(url_source=formUrl)
22    )
23    result: AnalyzeResult = poller.result()
24
25    # Output results
26    if result.styles:
27        for style in result.styles:
28            if style.is_handwritten:
29                print(f"Document contains handwritten content: {style.content}")
30
31    for page in result.pages:
32        print(f"----Analyzing layout from page #{page.page_number}----")
33        print(f"Page has width: {page.width} and height: {page.height}, measured with unit: {page.unit}")
34
35        if page.lines:
36            for line_idx, line in enumerate(page.lines):
37                print(f"...Line # {line_idx} has text content '{line.content}'")
38
39        if page.selection_marks:
40            for selection_mark in page.selection_marks:
41                print(
42                    f"Selection mark is '{selection_mark.state}' and has a confidence of {selection_mark.confidence}"
43                )
44
45    if result.tables:
46        for table_idx, table in enumerate(result.tables):
47            print(f"Table # {table_idx} has {table.row_count} rows and {table.column_count} columns")
48            if table.cells:
49                for cell in table.cells:
50                    print(
51                        f"...Cell[{cell.row_index}][{cell.column_index}] has content '{cell.content}'"
52                    )
53
54    print("----------------------------------------")
55
56if __name__ == "__main__":
57    analyze_layout()