mirrors/docling
1
0
Fork 0
mirror of https://github.com/DS4SD/docling.git synced 2025-08-01 15:02:21 +00:00
Code Issues Packages Projects Releases Wiki Activity

robustify & simplify format option resolution

Browse Source 
Signed-off-by: Panos Vagenas <35837085+vagenas@users.noreply.github.com>
This commit is contained in:
Panos Vagenas 2024-11-27 19:45:39 +01:00
parent 0bb1e203b6
commit 4138110c6b
1 changed files with 39 additions and 43 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

82
docling/document_converter.py
View File

@ -86,32 +86,37 @@ class ImageFormatOption(FormatOption):
backend: Type[AbstractDocumentBackend] = DoclingParseDocumentBackend backend: Type[AbstractDocumentBackend] = DoclingParseDocumentBackend
_format_to_default_options = { def _get_default_option(format: InputFormat) -> FormatOption:
InputFormat.XLSX: FormatOption( format_to_default_options = {
pipeline_cls=SimplePipeline, backend=MsExcelDocumentBackend InputFormat.XLSX: FormatOption(
), pipeline_cls=SimplePipeline, backend=MsExcelDocumentBackend
InputFormat.DOCX: FormatOption( ),
pipeline_cls=SimplePipeline, backend=MsWordDocumentBackend InputFormat.DOCX: FormatOption(
), pipeline_cls=SimplePipeline, backend=MsWordDocumentBackend
InputFormat.PPTX: FormatOption( ),
pipeline_cls=SimplePipeline, backend=MsPowerpointDocumentBackend InputFormat.PPTX: FormatOption(
), pipeline_cls=SimplePipeline, backend=MsPowerpointDocumentBackend
InputFormat.MD: FormatOption( ),
pipeline_cls=SimplePipeline, backend=MarkdownDocumentBackend InputFormat.MD: FormatOption(
), pipeline_cls=SimplePipeline, backend=MarkdownDocumentBackend
InputFormat.ASCIIDOC: FormatOption( ),
pipeline_cls=SimplePipeline, backend=AsciiDocBackend InputFormat.ASCIIDOC: FormatOption(
), pipeline_cls=SimplePipeline, backend=AsciiDocBackend
InputFormat.HTML: FormatOption( ),
pipeline_cls=SimplePipeline, backend=HTMLDocumentBackend InputFormat.HTML: FormatOption(
), pipeline_cls=SimplePipeline, backend=HTMLDocumentBackend
InputFormat.IMAGE: FormatOption( ),
pipeline_cls=StandardPdfPipeline, backend=DoclingParseDocumentBackend InputFormat.IMAGE: FormatOption(
), pipeline_cls=StandardPdfPipeline, backend=DoclingParseDocumentBackend
InputFormat.PDF: FormatOption( ),
pipeline_cls=StandardPdfPipeline, backend=DoclingParseDocumentBackend InputFormat.PDF: FormatOption(
), pipeline_cls=StandardPdfPipeline, backend=DoclingParseDocumentBackend
} ),
}
if (options := format_to_default_options.get(format)) is not None:
return options
else:
raise RuntimeError(f"No default options configured for {format}")
class DocumentConverter: class DocumentConverter:
@ -125,23 +130,14 @@ class DocumentConverter:
self.allowed_formats = ( self.allowed_formats = (
allowed_formats if allowed_formats is not None else [e for e in InputFormat] allowed_formats if allowed_formats is not None else [e for e in InputFormat]
) )
self.format_to_options = ( self.format_to_options = {
format_options if format_options is not None else _format_to_default_options format: (
) _get_default_option(format=format)
if format_options is not None: if (custom_option := (format_options or {}).get(format)) is None
for f in self.allowed_formats: else custom_option
if f not in self.format_to_options.keys(): )
_log.debug(f"Requested format {f} will use default options.") for format in self.allowed_formats
self.format_to_options[f] = _format_to_default_options[f] }
remove_keys = []
for f in self.format_to_options.keys():
if f not in self.allowed_formats:
remove_keys.append(f)
for f in remove_keys:
self.format_to_options.pop(f)
self.initialized_pipelines: Dict[Type[BasePipeline], BasePipeline] = {} self.initialized_pipelines: Dict[Type[BasePipeline], BasePipeline] = {}
def initialize_pipeline(self, format: InputFormat): def initialize_pipeline(self, format: InputFormat):