unitlab 3.1.1__tar.gz → 3.1.3__tar.gz

{unitlab-3.1.1 → unitlab-3.1.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: unitlab
-Version: 3.1.1
+Version: 3.1.3
 Summary: Python SDK for the Unitlab.ai data annotation platform
 Project-URL: Homepage, https://unitlab.ai
 Project-URL: Documentation, https://docs.unitlab.ai
@@ -441,7 +441,7 @@ Supported inputs are COCO JSON, Unitlab exports (UUEF item JSON, Audio JSON and
 GeoJSON, as `.json` or `.geojson`), `unitlab-image/1` and `unitlab-video/1`. They
 cover image, video, medical, document, audio, time series, text, HTML, geospatial
 and pathology items. A UUEF file holds one item: import each file as its own batch.
-Each JSON file is limited to 600 MiB. The backend matches original media paths/IDs,
+Each JSON file is limited to 80 MiB. The backend matches original media paths/IDs,
 with a unique-filename fallback; ambiguous names require an explicit mapping.
 YOLO text/ZIP files and other tools' project files are not accepted by this importer.
 

{unitlab-3.1.1 → unitlab-3.1.3}/README.md

@@ -408,7 +408,7 @@ Supported inputs are COCO JSON, Unitlab exports (UUEF item JSON, Audio JSON and
 GeoJSON, as `.json` or `.geojson`), `unitlab-image/1` and `unitlab-video/1`. They
 cover image, video, medical, document, audio, time series, text, HTML, geospatial
 and pathology items. A UUEF file holds one item: import each file as its own batch.
-Each JSON file is limited to 600 MiB. The backend matches original media paths/IDs,
+Each JSON file is limited to 80 MiB. The backend matches original media paths/IDs,
 with a unique-filename fallback; ambiguous names require an explicit mapping.
 YOLO text/ZIP files and other tools' project files are not accepted by this importer.
 

{unitlab-3.1.1 → unitlab-3.1.3}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "unitlab"
-version = "3.1.1"
+version = "3.1.3"
 description = "Python SDK for the Unitlab.ai data annotation platform"
 readme = "README.md"
 license = "MIT"

{unitlab-3.1.1 → unitlab-3.1.3}/src/unitlab/resources/annotation_imports.py

@@ -13,7 +13,11 @@ from ._base import Namespace, identifier
 if TYPE_CHECKING:
     from unitlab.client import UnitlabClient
 
-MAX_FILE_BYTES = 600 * 1024 * 1024
+# The server's cap (nbackend annotation_import_service.MAX_FILE_BYTES): a batch
+# stores its parsed rows as one jsonb value, which PostgreSQL caps at ~256 MiB,
+# and parsed rows take 2.2-2.8x the file. Checked here so a larger file fails
+# before its upload instead of with the server's 400 after it.
+MAX_FILE_BYTES = 80 * 1024 * 1024
 # Preview and commit re-check every row and item on the server; live runs on
 # large files took over a minute, past the client's 60 s default.
 SERVER_WORK_TIMEOUT = 300
@@ -55,7 +59,7 @@ class AnnotationImports(Namespace):
         """Upload COCO, Unitlab export or native JSON and return a label preview.
 
         Supply a local ``annotation_file`` or a saved ``cloud_storage`` handle/ID
-        and ``annotation_key``. JSON files are limited to 600 MiB each. Media must
+        and ``annotation_key``. JSON files are limited to 80 MiB each. Media must
         already belong to this project; use ``UploadBatch.wait()`` after upload.
 
         Exactly one prediction namespace, ``source_name`` or ``source_id``, is
@@ -118,7 +122,7 @@ class AnnotationImports(Namespace):
                 stream = stack.enter_context(path.open("rb"))
                 stream.seek(0, 2)
                 if stream.tell() > MAX_FILE_BYTES:
-                    raise ValueError(f"{name} must be 600 MiB or smaller.")
+                    raise ValueError(f"{name} must be 80 MiB or smaller.")
                 stream.seek(0)
                 files[name] = (path.name, stream, "application/json")
             if files:

{unitlab-3.1.1 → unitlab-3.1.3}/src/unitlab/resources/releases.py

@@ -35,6 +35,7 @@ class ReleasesNamespace(Namespace):
         data_types=None,
         bundle_formats: dict[str, str] | None = None,
         license_id: str | None = None,
+        completed_only: bool = False,
         wait: bool = True,
         timeout: float = 7200,
     ) -> Release:
@@ -61,6 +62,11 @@ class ReleasesNamespace(Namespace):
                 includes every available concrete type.
             bundle_formats: Per-concrete-data-type formats for multimodal bundles.
             license_id: Optional Dataset license ID.
+            completed_only: Only items whose current workflow status is
+                Complete (a Data Group by its own status), with their latest
+                labels. Raises ``UnitlabError`` (code
+                ``completed_only_unsupported``) if the server does not know the
+                option, instead of returning a Release of every item.
             wait: Block until the Release is ready or has failed.
             timeout: Seconds to wait for preparation when ``wait`` is true.
 
@@ -98,6 +104,9 @@ class ReleasesNamespace(Namespace):
             }
         if license_id is not None:
             payload["license"] = identifier(license_id)
+        if completed_only:
+            # Sent only when asked, so every other request body stays as before.
+            payload["completed_only"] = True
         # The long request timeout stays: a backend that predates background
         # preparation still builds the whole Release inside this POST.
         raw = self._api.post(
@@ -106,6 +115,14 @@ class ReleasesNamespace(Namespace):
             timeout=600.0,
         )
         release = Release._from_raw(self._client, raw)
+        if completed_only and not release.completed_only:
+            # An older server drops the unknown field and releases every item;
+            # its answer then lacks the echo. Say so rather than hand that back.
+            raise UnitlabError(
+                "This Unitlab server does not support completed_only releases; "
+                f"Release {release.id} was created with every labelled item.",
+                code="completed_only_unsupported",
+            )
         return release.wait(timeout=timeout) if wait else release
 
 
@@ -124,6 +141,8 @@ class Release:
     status: str | None = None
     progress: int | None = None
     status_error: str | None = None
+    # True when the Release holds only items that were Complete.
+    completed_only: bool = False
 
     @classmethod
     def _from_raw(cls, client: UnitlabClient, raw: dict[str, Any]) -> Release:
@@ -140,6 +159,7 @@ class Release:
             status=raw.get("status"),
             progress=raw.get("progress"),
             status_error=raw.get("status_error"),
+            completed_only=bool(raw.get("completed_only", False)),
         )
 
     def refresh(self) -> Release:

{unitlab-3.1.1 → unitlab-3.1.3}/tests/test_annotation_imports.py

@@ -267,7 +267,7 @@ def test_file_size_limit_and_closed_streams_on_request_failure(
         project.annotation_imports.create(labels, source_name="model")
     assert opened and all(stream.closed for stream in opened)
     monkeypatch.setattr("unitlab.resources.annotation_imports.MAX_FILE_BYTES", 1)
-    with pytest.raises(ValueError, match="600 MiB"):
+    with pytest.raises(ValueError, match="80 MiB"):
         project.annotation_imports.create(labels, source_name="model")
     assert len(opened) == 1