mirrors/docling
1
0
Fork 0
mirror of https://github.com/DS4SD/docling.git synced 2025-07-26 20:14:47 +00:00
Code Issues Packages Projects Releases Wiki Activity

add export options

Browse Source 
Signed-off-by: Michele Dolfi <dol@zurich.ibm.com>
This commit is contained in:
Michele Dolfi 2024-09-12 16:59:03 +02:00
parent fca08ca355
commit 18e019f566
1 changed files with 61 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

81
docling/cli/main.py
View File

@ -55,6 +55,10 @@ class Backend(str, Enum):
def export_documents( def export_documents(
conv_results: Iterable[ConversionResult], conv_results: Iterable[ConversionResult],
output_dir: Path, output_dir: Path,
export_json: bool,
export_md: bool,
export_txt: bool,
export_doctags: bool,
): ):
success_count = 0 success_count = 0
@ -66,28 +70,32 @@ def export_documents(
doc_filename = conv_res.input.file.stem doc_filename = conv_res.input.file.stem
# Export Deep Search document JSON format: # Export Deep Search document JSON format:
fname = output_dir / f"{doc_filename}.json" if export_json:
with fname.open("w") as fp: fname = output_dir / f"{doc_filename}.json"
_log.info(f"writing JSON output to {fname}") with fname.open("w") as fp:
fp.write(json.dumps(conv_res.render_as_dict())) _log.info(f"writing JSON output to {fname}")
fp.write(json.dumps(conv_res.render_as_dict()))
# Export Text format: # Export Text format:
fname = output_dir / f"{doc_filename}.txt" if export_txt:
with fname.open("w") as fp: fname = output_dir / f"{doc_filename}.txt"
_log.info(f"writing Text output to {fname}") with fname.open("w") as fp:
fp.write(conv_res.render_as_text()) _log.info(f"writing Text output to {fname}")
fp.write(conv_res.render_as_text())
# Export Markdown format: # Export Markdown format:
fname = output_dir / f"{doc_filename}.md" if export_md:
with fname.open("w") as fp: fname = output_dir / f"{doc_filename}.md"
_log.info(f"writing Markdown output to {fname}") with fname.open("w") as fp:
fp.write(conv_res.render_as_markdown()) _log.info(f"writing Markdown output to {fname}")
fp.write(conv_res.render_as_markdown())
# Export Document Tags format: # Export Document Tags format:
fname = output_dir / f"{doc_filename}.doctags" if export_doctags:
with fname.open("w") as fp: fname = output_dir / f"{doc_filename}.doctags"
_log.info(f"writing Doc Tags output to {fname}") with fname.open("w") as fp:
fp.write(conv_res.render_as_doctags()) _log.info(f"writing Doc Tags output to {fname}")
fp.write(conv_res.render_as_doctags())
else: else:
_log.warning(f"Document {conv_res.input.file} failed to convert.") _log.warning(f"Document {conv_res.input.file} failed to convert.")
@ -100,14 +108,40 @@ def export_documents(
@app.command(no_args_is_help=True) @app.command(no_args_is_help=True)
def convert( def convert(
input_files: Annotated[ input_sources: Annotated[
List[Path], List[Path],
typer.Argument( typer.Argument(
..., ...,
metavar="file", metavar="source",
help="PDF files to convert. Directories are also accepted.", help="PDF files to convert. Directories are also accepted.",
), ),
], ],
export_json: Annotated[
bool,
typer.Option(
..., "--json/--no-json", help="If enabled the document is exported as JSON."
),
] = False,
export_md: Annotated[
bool,
typer.Option(
..., "--md/--no-md", help="If enabled the document is exported as Markdown."
),
] = True,
export_txt: Annotated[
bool,
typer.Option(
..., "--txt/--no-txt", help="If enabled the document is exported as Text."
),
] = False,
export_doctags: Annotated[
bool,
typer.Option(
...,
"--doctags/--no-doctags",
help="If enabled the document is exported as Doc Tags.",
),
] = False,
ocr: Annotated[ ocr: Annotated[
bool, bool,
typer.Option( typer.Option(
@ -133,7 +167,7 @@ def convert(
logging.basicConfig(level=logging.INFO) logging.basicConfig(level=logging.INFO)
input_doc_paths: List[Path] = [] input_doc_paths: List[Path] = []
for source in input_files: for source in input_sources:
if not source.exists(): if not source.exists():
err_console.print( err_console.print(
f"[red]Error: The input file {source} does not exist.[/red]" f"[red]Error: The input file {source} does not exist.[/red]"
@ -205,7 +239,14 @@ def convert(
conv_results = doc_converter.convert(input) conv_results = doc_converter.convert(input)
output.mkdir(parents=True, exist_ok=True) output.mkdir(parents=True, exist_ok=True)
export_documents(conv_results, output_dir=output) export_documents(
conv_results,
output_dir=output,
export_json=export_json,
export_md=export_md,
export_txt=export_txt,
export_doctags=export_doctags,
)
end_time = time.time() - start_time end_time = time.time() - start_time