Docling Integration

pip install semantica
# Docling is included as an optional dependency
# or install separately:
pip install docling

from semantica.parse import DoclingParser

parser = DoclingParser(enable_ocr=True)
result = parser.parse("financial_report.pdf")

print(result["full_text"][:200])
print(f"Found {len(result['tables'])} tables")

from semantica.parse import DoclingParser

parser = DoclingParser(
    enable_ocr=True,
    export_format="markdown",
)

result = parser.parse("complex_invoice.pdf")

# Full text content
print(result["full_text"])

# Extracted tables
for i, table in enumerate(result["tables"]):
    print(f"Table {i+1} headers: {table.get('headers', [])}")
    for row in table.get("rows", [])[:3]:
        print(f"  Row: {row}")

# Metadata
metadata = result["metadata"]
print(f"Title: {metadata.get('title')}")
print(f"Pages: {result.get('total_pages')}")

{
    "full_text":    str,         # Clean document text
    "tables":       List[dict],  # Extracted tables (headers + rows)
    "metadata":     dict,        # Title, author, creation date, etc.
    "total_pages":  int,
}