From cca2ca40c848ee9ffe1ab3d7de391a84bb20e7f6 Mon Sep 17 00:00:00 2001 From: Chris Trahey Date: Wed, 25 Jun 2025 02:06:32 -0700 Subject: [PATCH 1/2] Allow disabling .json extension in fsspec connector uploads # Conflicts: # unstructured_ingest/__version__.py --- unstructured_ingest/__version__.py | 2 +- unstructured_ingest/processes/connectors/fsspec/fsspec.py | 3 ++- 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/unstructured_ingest/__version__.py b/unstructured_ingest/__version__.py index b360c032a..1bccd442a 100644 --- a/unstructured_ingest/__version__.py +++ b/unstructured_ingest/__version__.py @@ -1 +1 @@ -__version__ = "1.0.48" # pragma: no cover +__version__ = "1.1.0" # pragma: no cover diff --git a/unstructured_ingest/processes/connectors/fsspec/fsspec.py b/unstructured_ingest/processes/connectors/fsspec/fsspec.py index 22c6242e2..250fb6016 100644 --- a/unstructured_ingest/processes/connectors/fsspec/fsspec.py +++ b/unstructured_ingest/processes/connectors/fsspec/fsspec.py @@ -346,7 +346,8 @@ def get_upload_path(self, file_data: FileData) -> Path: upload_path = Path( self.upload_config.path_without_protocol ) / file_data.source_identifiers.relative_path.lstrip("/") - updated_upload_path = upload_path.parent / f"{upload_path.name}.json" + suffix = os.environ.get("UNSTRUCTURED_INGEST_UPLOAD_APPEND_SUFFIX", ".json") + updated_upload_path = upload_path.parent / f"{upload_path.name}{suffix}" return updated_upload_path def run(self, path: Path, file_data: FileData, **kwargs: Any) -> None: From e21d2e3e5278424f1806f748538c442baf972924 Mon Sep 17 00:00:00 2001 From: Chris Trahey Date: Wed, 25 Jun 2025 02:12:33 -0700 Subject: [PATCH 2/2] changelog --- CHANGELOG.md | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/CHANGELOG.md b/CHANGELOG.md index 258d8cf75..449903ebd 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -1,3 +1,7 @@ +## 1.1.0 + +* **Allow skipping .json extension for fsspec uploads** + ## 1.0.48 * **Improve Jira attachment path results**