dexverse-cli 0.1.0__tar.gz → 0.1.0.post1__tar.gz

{dexverse_cli-0.1.0 → dexverse_cli-0.1.0.post1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: dexverse-cli
-Version: 0.1.0
+Version: 0.1.0.post1
 Summary: Download, pack, and unpack DexVerse datasets
 Requires-Python: >=3.10
 Requires-Dist: av<17,>=14
@@ -41,7 +41,8 @@ type manifest.json | dexverse-dl download --manifest - --out ./dataset
 
 A manifest is either a JSON file list or an object containing `files`. Each
 file requires `download_url` and `file_name`. Optional fields are
-`storage_path`, `file_size`, `sha256`, and `meta_data`. Downloaded metadata is
+`storage_path`, `file_size`, `sha256`, `meta_data`, and `data_protocol`.
+`data_protocol` is retained in the downloaded metadata sidecar. Downloaded metadata is
 stored in a versioned sidecar and becomes `task_instruction.json` when the
 corresponding HDF5 file is unpacked; only top-level `data_protocol` is removed
 from that algorithm-facing projection.

{dexverse_cli-0.1.0 → dexverse_cli-0.1.0.post1}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "dexverse-cli"
-version = "0.1.0"
+version = "0.1.0.post1"
 description = "Download, pack, and unpack DexVerse datasets"
 readme = "src/dexverse_cli/README.md"
 requires-python = ">=3.10"

{dexverse_cli-0.1.0 → dexverse_cli-0.1.0.post1}/src/dexverse_cli/README.md

@@ -26,7 +26,8 @@ type manifest.json | dexverse-dl download --manifest - --out ./dataset
 
 A manifest is either a JSON file list or an object containing `files`. Each
 file requires `download_url` and `file_name`. Optional fields are
-`storage_path`, `file_size`, `sha256`, and `meta_data`. Downloaded metadata is
+`storage_path`, `file_size`, `sha256`, `meta_data`, and `data_protocol`.
+`data_protocol` is retained in the downloaded metadata sidecar. Downloaded metadata is
 stored in a versioned sidecar and becomes `task_instruction.json` when the
 corresponding HDF5 file is unpacked; only top-level `data_protocol` is removed
 from that algorithm-facing projection.

dexverse_cli-0.1.0.post1/src/dexverse_cli/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.1.0.post1"

{dexverse_cli-0.1.0 → dexverse_cli-0.1.0.post1}/src/dexverse_cli/manifest.py

@@ -17,6 +17,7 @@ _ALLOWED_FIELDS = frozenset(
         "file_size",
         "sha256",
         "meta_data",
+        "data_protocol",
     }
 )
 _DRIVE_PREFIX = re.compile(r"^[A-Za-z]:")
@@ -116,12 +117,19 @@ def optional_sha256(value: object, *, index: int) -> str | None:
 
 
 def metadata(item: dict[str, Any], *, index: int) -> dict[str, Any] | None:
-    if "meta_data" not in item:
-        return None
-    value = item["meta_data"]
-    if not isinstance(value, dict):
+    value = item.get("meta_data") if "meta_data" in item else None
+    if "meta_data" in item and not isinstance(value, dict):
         raise ManifestError(f"files[{index}].meta_data must be a JSON object")
-    return value
+    merged = dict(value) if value is not None else {}
+    if "data_protocol" not in item:
+        return merged or None
+    data_protocol = item["data_protocol"]
+    if not isinstance(data_protocol, str) or not data_protocol or data_protocol != data_protocol.strip():
+        raise ManifestError(f"files[{index}].data_protocol must be a non-empty string")
+    if "data_protocol" in merged and merged["data_protocol"] != data_protocol:
+        raise ManifestError(f"files[{index}] has conflicting data_protocol values")
+    merged["data_protocol"] = data_protocol
+    return merged
 
 
 def claim_parts(file_name: str) -> tuple[str, ...]:

{dexverse_cli-0.1.0 → dexverse_cli-0.1.0.post1}/uv.lock

@@ -108,7 +108,7 @@ wheels = [
 
 [[package]]
 name = "dexverse-cli"
-version = "0.1.0"
+version = "0.1.0.post1"
 source = { editable = "." }
 dependencies = [
     { name = "av" },

dexverse_cli-0.1.0/src/dexverse_cli/__init__.py

@@ -1 +0,0 @@
-__version__ = "0.1.0"