Add compression argument to publish()

audmodel/core/api.py

@@ -434,6 +434,7 @@ def publish(
     repository: Repository,
     alias: str | None = None,
     author: str | None = None,
+    compression: int = 1,
     date: datetime.date | None = None,
     meta: dict[str, object] | None = None,
     subgroup: str | None = None,
@@ -509,6 +510,21 @@ def publish(
             If provided, the model can be accessed using this alias
             in addition to its UID
         author: author name(s), defaults to user name
+        compression: compression level
+            of the model archive.
+            ``0`` stores the model files
+            without compression,
+            ``1``-``9`` selects a deflate level.
+            Deflate is single threaded,
+            and dominates the publication time
+            of large models.
+            Higher levels than ``1``
+            hardly compress better
+            on model weights,
+            but take at least twice as long.
+            Select ``0``,
+            if you want to publish
+            as fast as possible
         date: date, defaults to current timestamp
         meta: dictionary with meta information
         subgroup: subgroup under which
@@ -540,6 +556,7 @@ def publish(
         FileNotFoundError: if ``root`` folder cannot be found
         ValueError: if ``alias`` can be confused with an UID,
             or it does contain chars other than ``[A-Za-z0-9._-]+``
+        ValueError: if ``compression`` is not between 0 and 9
 
     Examples:
         >>> # Assuming your model files are stored under `model_root`
@@ -616,6 +633,9 @@ def publish(
             "and are not allowed to be confused with a model ID."
         )
 
+    if not 0 <= compression <= 9:
+        raise ValueError(f"'compression' has to be between 0 and 9, not {compression}.")
+
     if not os.path.isdir(root):
         raise FileNotFoundError(
             errno.ENOENT,
@@ -662,6 +682,7 @@ def publish(
             subgroup,
             root,
             backend_interface,
+            compression,
             verbose,
             tmp_root=tmp_root,
         )

audmodel/core/backend.py

@@ -389,6 +389,7 @@ def put_archive(
     subgroup: str,
     root: str,
     backend_interface: audbackend.interface.Maven,
+    compression: int,
     verbose: bool,
     tmp_root: str | None = None,
 ) -> str:
@@ -401,6 +402,11 @@ def put_archive(
         subgroup: model subgroup
         root: path to model root folder
         backend_interface: backend interface instance
+        compression: compression level
+            of the model archive.
+            ``0`` stores the model files
+            without compression,
+            ``1``-``9`` selects a deflate level
         verbose: if ``True`` show message
             when uploading file
         tmp_root: folder under which the temporary archive
@@ -433,6 +439,7 @@ def put_archive(
             root,
             files,
             src_path,
+            compression=compression,
             verbose=verbose,
         )
         with backend_interface.backend:

pyproject.toml

@@ -31,6 +31,7 @@ classifiers = [
 requires-python = '>=3.10'
 dependencies = [
     'audbackend[all] >=3.0.0',
+    'audeer >=2.6.0',
     'filelock >=3.10',
     'oyaml',
 ]
@@ -46,7 +47,6 @@ documentation = 'http://tools.pp.audeering.com/audmodel/'
 # ===== Dependency groups =================================================
 [dependency-groups]
 dev = [
-    'audeer >=1.3.0',
     'parse',
     'pytest',
     'pytest-cov',

tests/test_publish.py

@@ -1,4 +1,5 @@
 import os
+import zipfile
 
 import pytest
 
@@ -297,3 +298,68 @@ def test_publish_missing_repository_raises():
             "1.0.0",
             repository=None,
         )
+
+
+@pytest.mark.parametrize(
+    "compression, expected_compress_type",
+    [
+        # deflate level 1 by default
+        (None, zipfile.ZIP_DEFLATED),
+        (0, zipfile.ZIP_STORED),
+        (1, zipfile.ZIP_DEFLATED),
+        (9, zipfile.ZIP_DEFLATED),
+    ],
+)
+def test_publish_compression(tmp_path, compression, expected_compress_type):
+    r"""Test compression of the published model archive.
+
+    Args:
+        tmp_path: tmp_path fixture
+        compression: compression level,
+            ``None`` publishes without the argument
+        expected_compress_type: expected compression method
+            of the entries in the model archive
+
+    """
+    root = audeer.mkdir(tmp_path, "model")
+    with open(os.path.join(root, "model.bin"), "w") as file:
+        file.write("a" * 10000)
+
+    kwargs = {} if compression is None else {"compression": compression}
+    effective_compression = 1 if compression is None else compression
+    uid = audmodel.publish(
+        root,
+        pytest.NAME,
+        {"compression": effective_compression, "compression_arg": compression},
+        "1.0.0",
+        author=pytest.AUTHOR,
+        date=pytest.DATE,
+        subgroup=f"{SUBGROUP}.compression",
+        repository=pytest.REPOSITORIES[0],
+        **kwargs,
+    )
+
+    with zipfile.ZipFile(audmodel.url(uid)) as archive:
+        infos = archive.infolist()
+    assert [info.filename for info in infos] == ["model.bin"]
+    assert infos[0].compress_type == expected_compress_type
+
+
+@pytest.mark.parametrize("compression", [-1, 10])
+def test_publish_compression_error(compression):
+    r"""Test error for a compression level outside 0-9.
+
+    Args:
+        compression: invalid compression level
+
+    """
+    error_msg = f"'compression' has to be between 0 and 9, not {compression}."
+    with pytest.raises(ValueError, match=error_msg):
+        audmodel.publish(
+            pytest.MODEL_ROOT,
+            pytest.NAME,
+            {},
+            "1.0.0",
+            compression=compression,
+            repository=pytest.REPOSITORIES[0],
+        )