✨ feat(cli): add --extract-media option for image extraction (3d8251c4) · Commits · Jan Reimes / 3gpp-crawler

src/tdoc_crawler/cli/args.py

+7 −0

Original line number	Diff line number	Diff line
		@@ -224,6 +224,13 @@ DocxDirectOption = Annotated[
		help="Feed .docx/.doc directly to backend, skip LibreOffice PDF conversion",
		),
		]
		ExtractMediaOption = Annotated[
		bool,
		typer.Option(
		"--extract-media",
		help="Extract embedded images to a ./media folder next to the markdown",
		),
		]
		WorkspaceNameOption = Annotated[
		str \| None,
		typer.Option("-w", "--workspace", help="Workspace name (default: active workspace)"),

+5 −0

Original line number	Diff line number	Diff line
		@@ -13,6 +13,7 @@ from tdoc_crawler.cli._shared import console, create_progress_bar
		from tdoc_crawler.cli.args import (
		DeviceOption,
		DocxDirectOption,
		ExtractMediaOption,
		FiguresModeOption,
		MdYamlFrontmatterOption,
		ProcessLimitOption,
		@@ -131,6 +132,7 @@ def _process_member(
		md_yaml_frontmatter: bool,
		docling_config: DoclingConfig,
		docx_direct: bool = False,
		extract_media: bool = False,
		) -> tuple[str, bool, bool, int]:
		"""Process a single workspace member.

		@@ -150,6 +152,7 @@ def _process_member(
		force=force,
		docling_config=docling_config,
		docx_direct=docx_direct,
		extract_media=extract_media,
		)
		if result_path:
		suffix = result_path.suffix.lstrip(".")
		@@ -177,6 +180,7 @@ def workspace_process(
		tables: TablesModeOption = "embed",
		device: DeviceOption = "auto",
		docx_direct: DocxDirectOption = False,
		extract_media: ExtractMediaOption = False,
		md_yaml_frontmatter: MdYamlFrontmatterOption = True,
		verbosity: VerbosityOption = str(DEFAULT_VERBOSITY),
		) -> None:
		@@ -258,6 +262,7 @@ def workspace_process(
		md_yaml_frontmatter,
		docling_config,
		docx_direct,
		extract_media,
		)
		if succeeded:
		processed += 1