fix: add pi-heif as direct dependency and improve markdown file handling

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

Author: devin-ai-integration[bot]

airbyte_cdk/sources/file_based/file_types/unstructured_parser.py

@@ -218,9 +218,13 @@ def _read_file(
                 self._get_file_type_error_message(filetype),
             )
         if filetype in {FileType.MD, FileType.TXT}:
-            file_content: bytes = file_handle.read()
-            decoded_content: str = optional_decode(file_content)
-            return decoded_content
+            try:
+                file_content: bytes = file_handle.read()
+                decoded_content: str = optional_decode(file_content)
+                return decoded_content
+            except Exception as e:
+                logger.error(f"Error reading {filetype} file: {str(e)}")
+                raise self._create_parse_error(remote_file, str(e))
         if format.processing.mode == "local":
             return self._read_file_locally(
                 file_handle,

poetry.lock

@@ -84,6 +84,7 @@ sqlalchemy = {version = "^2.0,!=2.0.36", optional = true }
 xmltodict = ">=0.13,<0.15"
 anyascii = "^0.3.2"
 whenever = "^0.6.16"
+pi-heif = "^0.22.0"
 
 [tool.poetry.group.dev.dependencies]
 freezegun = "*"
@@ -106,6 +107,7 @@ types-python-dateutil = "^2.9.0.20241003"
 types-pyyaml = "^6.0.12.20240917"
 types-cachetools = "^5.5.0.20240820"
 deptry = "^0.23.0"
+pi-heif = "^0.22.0"
 
 [tool.poetry.extras]
 file-based = ["avro", "fastavro", "pyarrow", "unstructured", "pdf2image", "pdfminer.six", "unstructured.pytesseract", "pytesseract", "markdown", "python-calamine", "python-snappy", "pi-heif"]
@@ -154,7 +156,7 @@ lint-fix = { cmd = "poetry run ruff check --fix .", help = "Auto-fix any lint is
 lint-fix-unsafe = { cmd = "poetry run ruff check --fix --unsafe-fixes .", help = "Lint-fix modified files, including 'unsafe' fixes. It is recommended to first commit any pending changes and then always manually review any unsafe changes applied." }
 
 # ruff fix everything (ignoring non-Python fixes)
-ruff-fix = { sequence = ["lint-fix", "_format-fix-ruff"] , help = "Lint-fix and format-fix all code." }
+ruff-fix = { sequence = ["lint-fix", "_format-fix-ruff"],  help = "Lint-fix and format-fix all code." }
 
 # Combined Check and Fix tasks