fix parsing of unit strings with 2 letters (fixes #300) (#301)

Author: jameslamb

.pre-commit-config.yaml

@@ -10,13 +10,13 @@ repos:
       - id: end-of-file-fixer
       - id: trailing-whitespace
   - repo: https://github.com/pycqa/isort
-    rev: 5.13.2
+    rev: 6.0.0
     hooks:
       - id: isort
         name: isort (python)
         args: ["--settings-path", "pyproject.toml"]
   - repo: https://github.com/pre-commit/mirrors-mypy
-    rev: v1.14.1
+    rev: v1.15.0
     hooks:
       - id: mypy
         args: ["--config-file", "pyproject.toml"]
@@ -27,7 +27,7 @@ repos:
           - types-requests
   - repo: https://github.com/astral-sh/ruff-pre-commit
     # Ruff version.
-    rev: v0.9.3
+    rev: v0.9.5
     hooks:
       # Run the linter.
       - id: ruff
@@ -61,7 +61,7 @@ repos:
       - id: yamllint
         additional_dependencies: [pyyaml]
   - repo: https://github.com/codespell-project/codespell
-    rev: v2.4.0
+    rev: v2.4.1
     hooks:
       - id: codespell
         additional_dependencies: [tomli]

src/pydistcheck/_utils.py

@@ -3,6 +3,7 @@
 not specific to package distributions
 """
 
+import re
 from typing import Tuple
 
 # references:
@@ -47,7 +48,10 @@ def from_number(cls, num: int) -> "_FileSize":
 
     @classmethod
     def from_string(cls, size_str: str) -> "_FileSize":
-        return cls(num=float(size_str[:-1]), unit_str=size_str[-1])
+        parsed = re.search(r"^([0-9\.]+)([A-Za-z]+)$", size_str.strip())
+        if parsed is None:
+            raise ValueError(f"Could not parse '{size_str}' as a file size.")
+        return cls(num=float(parsed.group(1)), unit_str=parsed.group(2))
 
     @property
     def total_size_bytes(self) -> int:

tests/test_cli.py

@@ -426,6 +426,7 @@ def test_check_respects_max_allowed_files(distro_file):
     [
         "1K",
         "3.999K",
+        "0.005KB",
         "0.0002M",
         "0.00000008G",
         "708B",
@@ -453,11 +454,45 @@ def test_check_respects_max_allowed_size_compressed(size_str, distro_file):
     _assert_log_matches_pattern(result, "errors found while checking\\: 1")
 
 
+@pytest.mark.parametrize(
+    "size_str",
+    [
+        "1",
+        "K",
+        "1-KB",
+        "GB1",
+        "1G-B",
+    ],
+)
+@pytest.mark.parametrize(
+    "cli_arg",
+    [
+        "--max-allowed-size-compressed",
+        "--max-allowed-size-uncompressed",
+    ],
+)
+def test_check_raises_informative_error_for_malformed_file_size_config(
+    cli_arg, size_str
+):
+    with pytest.raises(
+        ValueError, match=rf"Could not parse '{size_str}' as a file size"
+    ):
+        CliRunner().invoke(
+            check,
+            [
+                os.path.join(TEST_DATA_DIR, BASE_PACKAGES[0]),
+                f"{cli_arg}={size_str}",
+            ],
+            catch_exceptions=False,
+        )
+
+
 @pytest.mark.parametrize(
     "size_str",
     [
         "1K",
         "3.999K",
+        "0.005KB",
         "0.0002M",
         "0.00000008G",
         "708B",

tests/test_utils.py

@@ -58,6 +58,29 @@ def test_file_size_from_different_inputs_all_parsed_consistently(file_size):
     assert file_size == _FileSize(num=3.0, unit_str="M")
 
 
+def test_file_size_from_string_works():
+    # 1 letter
+    assert _FileSize.from_string("57B").total_size_bytes == 57
+    assert _FileSize.from_string("57K").total_size_bytes == 57 * 1024
+    assert _FileSize.from_string("57M").total_size_bytes == 57 * 1024**2
+    assert _FileSize.from_string("57G").total_size_bytes == 57 * 1024**3
+
+    # 2 letter
+    assert _FileSize.from_string("57KB").total_size_bytes == 57 * 1e3
+    assert _FileSize.from_string("57MB").total_size_bytes == 57 * 1e6
+    assert _FileSize.from_string("57GB").total_size_bytes == 57 * 1e9
+
+    # decimals
+    assert _FileSize.from_string("0.005GB").total_size_bytes == 0.005 * 1e9
+    assert _FileSize.from_string("0.05GB").total_size_bytes == 0.05 * 1e9
+    assert _FileSize.from_string("0.5GB").total_size_bytes == 0.5 * 1e9
+    assert _FileSize.from_string("5.0GB").total_size_bytes == 5 * 1e9
+    assert _FileSize.from_string("5.000GB").total_size_bytes == 5 * 1e9
+    assert _FileSize.from_string("5.1GB").total_size_bytes == 5.1 * 1e9
+    assert _FileSize.from_string("5.17GB").total_size_bytes == 5.17 * 1e9
+    assert _FileSize.from_string("5.234GB").total_size_bytes == 5.234 * 1e9
+
+
 def test_file_size_from_number_switches_unit_str_based_on_size():
     assert _FileSize.from_number(1.1) == _FileSize(num=1.1, unit_str="B")
     # fractional bytes don't make sense here, so some rounding happens