Method _convert

docling/document_converter.py:243–270

(
        self, conv_input: _DocumentConversionInput, raises_on_error: bool
    )

            )

    def _convert(
        self, conv_input: _DocumentConversionInput, raises_on_error: bool
    ) -> Iterator[ConversionResult]:
        start_time = time.monotonic()

        for input_batch in chunkify(
            conv_input.docs(self.format_to_options),
            settings.perf.doc_batch_size,  # pass format_options
        ):
            _log.info(f"Going to convert document batch...")

            # parallel processing only within input_batch
            # with ThreadPoolExecutor(
            #    max_workers=settings.perf.doc_batch_concurrency
            # ) as pool:
            #   yield from pool.map(self.process_document, input_batch)
            # Note: PDF backends are not thread-safe, thread pool usage was disabled.

            for item in map(
                partial(self._process_document, raises_on_error=raises_on_error),
                input_batch,
            ):
                elapsed = time.monotonic() - start_time
                start_time = time.monotonic()
                _log.info(
                    f"Finished converting document {item.input.file.name} in {elapsed:.2f} sec."
                )
                yield item

    def _get_pipeline(self, doc_format: InputFormat) -> Optional[BasePipeline]:
        fopt = self.format_to_options.get(doc_format)