fix: update partition_pdf import path and add pi-heif dependency

Co-Authored-By: Aaron <AJ> Steers <[email protected]>

Author: devin-ai-integration[bot]

airbyte_cdk/sources/file_based/file_types/unstructured_parser.py

@@ -80,7 +80,7 @@ def _import_unstructured() -> None:
     global unstructured_partition_docx
     global unstructured_partition_pptx
     from unstructured.partition.docx import partition_docx
-    from unstructured.partition.pdf import partition_pdf
+    from unstructured.partition.pdf_image.pdf import partition_pdf
     from unstructured.partition.pptx import partition_pptx
 
     # separate global variables to properly propagate typing

pyproject.toml

@@ -108,7 +108,7 @@ types-cachetools = "^5.5.0.20240820"
 deptry = "^0.23.0"
 
 [tool.poetry.extras]
-file-based = ["avro", "fastavro", "pyarrow", "unstructured", "pdf2image", "pdfminer.six", "unstructured.pytesseract", "pytesseract", "markdown", "python-calamine", "python-snappy"]
+file-based = ["avro", "fastavro", "pyarrow", "unstructured", "pdf2image", "pdfminer.six", "unstructured.pytesseract", "pytesseract", "markdown", "python-calamine", "python-snappy", "pi-heif"]
 vector-db-based = ["langchain", "openai", "cohere", "tiktoken"]
 sql = ["sqlalchemy"]