Skip to content

feat: smart save_to with full path support and bug fix #85

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
mikesoennichsen wants to merge 6 commits into
base: main
Choose a base branch
from
Open

feat: smart save_to with full path support and bug fix #85

Show file tree
Hide file tree
Changes from 5 commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
8ca7c0f
fix: handle raw markdown strings in _get_input_filename
mikesoennichsen Apr 9, 2026
878e6f6
feat: support full .json file paths in save_to parameter
mikesoennichsen Apr 9, 2026
565718d
feat: add save_to to async client methods
mikesoennichsen Apr 9, 2026
bf45581
fix: treat all string inputs as content in _get_input_filename
mikesoennichsen Apr 9, 2026
cd93721
fix: simplify default filename to avoid redundant "output" prefix
mikesoennichsen Apr 10, 2026
bf91332
fix: update docstrings and handle os.PathLike inputs
mikesoennichsen Apr 10, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
90 changes: 71 additions & 19 deletions src/landingai_ade/_client.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -85,8 +85,12 @@ def _get_input_filename(
) -> str:
"""Extract base filename (without extension) from file or URL input."""
if file_input is not None and not isinstance(file_input, Omit):
if isinstance(file_input, (Path, str)):
return Path(file_input).stem
if isinstance(file_input, Path):
return file_input.stem
elif isinstance(file_input, str):
# Strings are always treated as raw content, not file paths.
# File inputs should use Path objects, tuples, or IO objects.
pass
elif isinstance(file_input, tuple) and len(file_input) > 0:
# Tuple format: (filename, content, mime_type)
return Path(str(file_input[0])).stem
Expand All @@ -111,12 +115,24 @@ def _save_response(
method_name: str,
result: Any,
) -> None:
"""Save API response to a JSON file in the specified folder."""
"""Save API response to a JSON file.

If save_to ends with '.json', it is treated as a full file path and the
response is written there directly. Otherwise it is treated as a directory
and the file is auto-named '{filename}_{method_name}_output.json'.
"""
try:
folder = Path(save_to)
folder.mkdir(parents=True, exist_ok=True)
output_path = folder / f"{filename}_{method_name}_output.json"
output_path.write_text(result.to_json())
save_path = Path(save_to)
if str(save_to).endswith(".json"):
save_path.parent.mkdir(parents=True, exist_ok=True)
save_path.write_text(result.to_json())
else:
save_path.mkdir(parents=True, exist_ok=True)
if filename == "output":
output_path = save_path / f"{method_name}_output.json"
else:
output_path = save_path / f"{filename}_{method_name}_output.json"
output_path.write_text(result.to_json())
except OSError as exc:
raise LandingAiadeError(f"Failed to save {method_name} response to {save_to}: {exc}") from exc

Expand Down Expand Up @@ -328,9 +344,9 @@ def extract(
strict: If True, reject schemas with unsupported fields (HTTP 422). If False, prune
unsupported fields and continue. Only applies to extract versions that support
schema validation.
save_to: Optional output folder path. If provided, the response will be saved as
JSON to this folder with the filename format: {input_file}_extract_output.json.
The folder will be created if it doesn't exist.
save_to: Optional output path. Accepts either a directory path (auto-generates
filename as {input_file}_extract_output.json) or a full file path ending
in .json (saves to that exact path). Parent directories are created automatically.

extra_headers: Send extra headers

Expand Down Expand Up @@ -429,9 +445,9 @@ def parse(
parameter. Set the parameter to page to split documents at the page level. The
splits object in the API output will contain a set of data for each page.

save_to: Optional output folder path. If provided, the response will be saved as
JSON to this folder with the filename format: {input_file}_parse_output.json.
The folder will be created if it doesn't exist.
save_to: Optional output path. Accepts either a directory path (auto-generates
filename as {input_file}_parse_output.json) or a full file path ending
in .json (saves to that exact path). Parent directories are created automatically.

extra_headers: Send extra headers

Expand Down Expand Up @@ -518,9 +534,9 @@ def split(

model: Model version to use for split classification. Defaults to the latest version.

save_to: Optional output folder path. If provided, the response will be saved as
JSON to this folder with the filename format: {input_file}_split_output.json.
The folder will be created if it doesn't exist.
save_to: Optional output path. Accepts either a directory path (auto-generates
filename as {input_file}_split_output.json) or a full file path ending
in .json (saves to that exact path). Parent directories are created automatically.

extra_headers: Send extra headers

Expand Down Expand Up @@ -768,6 +784,7 @@ async def extract(
markdown_url: Optional[str] | Omit = omit,
model: Optional[str] | Omit = omit,
strict: bool | Omit = omit,
save_to: str | Path | None = None,
# Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
# The extra values given here take precedence over values defined on the client or passed to this method.
extra_headers: Headers | None = None,
Expand Down Expand Up @@ -801,6 +818,10 @@ async def extract(
unsupported fields and continue. Only applies to extract versions that support
schema validation.

save_to: Optional output path. Accepts either a directory path (auto-generates
filename as {input_file}_extract_output.json) or a full file path ending
in .json (saves to that exact path). Parent directories are created automatically.

extra_headers: Send extra headers

extra_query: Add additional query parameters to the request
Expand All @@ -809,6 +830,9 @@ async def extract(

timeout: Override the client-level default timeout for this request, in seconds
"""
# Store original inputs for filename extraction before conversion
original_markdown = markdown
original_markdown_url = markdown_url
# Convert local file paths to file parameters
markdown, markdown_url = convert_url_to_file_if_local(markdown, markdown_url)

Expand All @@ -830,7 +854,7 @@ async def extract(
"runtime_tag": f"ade-python-v{_LIB_VERSION}",
**(extra_headers or {}),
}
return await self.post(
result = await self.post(
"/v1/ade/extract",
body=await async_maybe_transform(body, client_extract_params.ClientExtractParams),
files=files,
Expand All @@ -842,6 +866,10 @@ async def extract(
),
cast_to=ExtractResponse,
)
if save_to:
filename = _get_input_filename(original_markdown, original_markdown_url)
_save_response(save_to, filename, "extract", result)
return result

async def parse(
self,
Expand All @@ -852,6 +880,7 @@ async def parse(
model: Optional[str] | Omit = omit,
password: Optional[str] | Omit = omit,
split: Optional[Literal["page"]] | Omit = omit,
save_to: str | Path | None = None,
# Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
# The extra values given here take precedence over values defined on the client or passed to this method.
extra_headers: Headers | None = None,
Expand Down Expand Up @@ -890,6 +919,10 @@ async def parse(
parameter. Set the parameter to page to split documents at the page level. The
splits object in the API output will contain a set of data for each page.

save_to: Optional output path. Accepts either a directory path (auto-generates
filename as {input_file}_parse_output.json) or a full file path ending
in .json (saves to that exact path). Parent directories are created automatically.

extra_headers: Send extra headers

extra_query: Add additional query parameters to the request
Expand All @@ -898,6 +931,9 @@ async def parse(

timeout: Override the client-level default timeout for this request, in seconds
"""
# Store original inputs for filename extraction before conversion
original_document = document
original_document_url = document_url
# Convert local file paths to file parameters
document, document_url = convert_url_to_file_if_local(document, document_url)

Expand All @@ -920,7 +956,7 @@ async def parse(
"runtime_tag": f"ade-python-v{_LIB_VERSION}",
**(extra_headers or {}),
}
return await self.post(
result = await self.post(
"/v1/ade/parse",
body=await async_maybe_transform(body, client_parse_params.ClientParseParams),
files=files,
Expand All @@ -932,6 +968,10 @@ async def parse(
),
cast_to=ParseResponse,
)
if save_to:
filename = _get_input_filename(original_document, original_document_url)
_save_response(save_to, filename, "parse", result)
return result

async def split(
self,
Expand All @@ -940,6 +980,7 @@ async def split(
markdown: Union[FileTypes, str, None] | Omit = omit,
markdown_url: Optional[str] | Omit = omit,
model: Optional[str] | Omit = omit,
save_to: str | Path | None = None,
# Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
# The extra values given here take precedence over values defined on the client or passed to this method.
extra_headers: Headers | None = None,
Expand Down Expand Up @@ -967,6 +1008,10 @@ async def split(

model: Model version to use for split classification. Defaults to the latest version.

save_to: Optional output path. Accepts either a directory path (auto-generates
filename as {input_file}_split_output.json) or a full file path ending
in .json (saves to that exact path). Parent directories are created automatically.

extra_headers: Send extra headers

extra_query: Add additional query parameters to the request
Expand All @@ -975,6 +1020,9 @@ async def split(

timeout: Override the client-level default timeout for this request, in seconds
"""
# Store original inputs for filename extraction
original_markdown = markdown
original_markdown_url = markdown_url
body = deepcopy_minimal(
{
"split_class": split_class,
Expand All @@ -988,7 +1036,7 @@ async def split(
# sent to the server will contain a `boundary` parameter, e.g.
# multipart/form-data; boundary=---abc--
extra_headers = {"Content-Type": "multipart/form-data", **(extra_headers or {})}
return await self.post(
result = await self.post(
"/v1/ade/split",
body=await async_maybe_transform(body, client_split_params.ClientSplitParams),
files=files,
Expand All @@ -997,6 +1045,10 @@ async def split(
),
cast_to=SplitResponse,
)
if save_to:
filename = _get_input_filename(original_markdown, original_markdown_url)
_save_response(save_to, filename, "split", result)
return result

@override
def _make_status_error(
Expand Down
Loading