Handle malformed YAML

jacebrowning · Jun 6, 2020 · a0b617c · a0b617c
1 parent 3960089
commit a0b617c
Show file tree

Hide file tree

Showing 13 changed files with 69 additions and 44 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -1,6 +1,7 @@
 # 0.10 (beta)
 
 - Added support for recursively matching arbitrary depth paths of files.
+- Fixed `AttributeError` when attempting to load malformed YAML files.
 
 # 0.9 (2020-04-13)
 

diff --git a/datafiles/formats.py b/datafiles/formats.py
@@ -5,7 +5,7 @@
 from contextlib import suppress
 from io import StringIO
 from pathlib import Path
-from typing import IO, Any, Dict, List
+from typing import IO, Dict, List
 
 import log
 
@@ -30,7 +30,7 @@ def extensions(cls) -> List[str]:
 
     @classmethod
     @abstractmethod
-    def deserialize(cls, file_object: IO[Any]) -> Dict:
+    def deserialize(cls, file_object: IO) -> Dict:
         raise NotImplementedError
 
     @classmethod
@@ -48,7 +48,7 @@ def extensions(cls):
 
     @classmethod
     def deserialize(cls, file_object):
-        return json.load(file_object) or {}
+        return json.load(file_object)
 
     @classmethod
     def serialize(cls, data):
@@ -66,7 +66,7 @@ def extensions(cls):
     def deserialize(cls, file_object):
         import tomlkit
 
-        return tomlkit.loads(file_object.read()) or {}
+        return tomlkit.loads(file_object.read())
 
     @classmethod
     def serialize(cls, data):
@@ -87,12 +87,10 @@ def deserialize(cls, file_object):
         from ruamel import yaml
 
         try:
-            data = yaml.round_trip_load(file_object, preserve_quotes=True)
+            return yaml.round_trip_load(file_object, preserve_quotes=True)
         except NotImplementedError as e:
             log.error(str(e))
             return {}
-        else:
-            return data or {}
 
     @classmethod
     def serialize(cls, data):
@@ -131,9 +129,7 @@ def extensions(cls):
     def deserialize(cls, file_object):
         import yaml
 
-        data = yaml.safe_load(file_object) or {}
-
-        return data
+        return yaml.safe_load(file_object)
 
     @classmethod
     def serialize(cls, data):
@@ -160,7 +156,14 @@ def deserialize(path: Path, extension: str, *, formatter=None) -> Dict:
     if formatter is None:
         formatter = _get_formatter(extension)
     with path.open('r') as file_object:
-        return formatter.deserialize(file_object)
+        data = formatter.deserialize(file_object)
+        if data is None:
+            log.debug(f"No data in {path}")
+            data = {}
+        elif not isinstance(data, dict):
+            log.error(f"Invalid data in {path}: {data!r}")
+            data = {}
+        return data
 
 
 def serialize(data: Dict, extension: str = '.yml', *, formatter=None) -> str:

diff --git a/notebooks/file_inference.ipynb b/notebooks/file_inference.ipynb
@@ -287,7 +287,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.2"
+   "version": "3.8.2"
   }
  },
  "nbformat": 4,

diff --git a/notebooks/format_options.ipynb b/notebooks/format_options.ipynb
@@ -256,7 +256,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.2"
+   "version": "3.8.2"
   }
  },
  "nbformat": 4,

diff --git a/notebooks/manager_api.ipynb b/notebooks/manager_api.ipynb
@@ -331,7 +331,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.2"
+   "version": "3.8.2"
   }
  },
  "nbformat": 4,

diff --git a/notebooks/mapper_api.ipynb b/notebooks/mapper_api.ipynb
@@ -325,7 +325,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.2"
+   "version": "3.8.2"
   }
  },
  "nbformat": 4,

diff --git a/notebooks/nested_dataclass.ipynb b/notebooks/nested_dataclass.ipynb
@@ -184,7 +184,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.2"
+   "version": "3.8.2"
   }
  },
  "nbformat": 4,

diff --git a/notebooks/patched_containers.ipynb b/notebooks/patched_containers.ipynb
@@ -364,7 +364,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.2"
+   "version": "3.8.2"
   }
  },
  "nbformat": 4,

diff --git a/notebooks/readme.ipynb b/notebooks/readme.ipynb
@@ -154,7 +154,14 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Overwriting inventory/items/widget.yml\n"
+      "Overwriting inventory/items/widget.yml"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n"
      ]
     }
    ],
@@ -244,7 +251,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.2"
+   "version": "3.8.2"
   }
  },
  "nbformat": 4,

diff --git a/notebooks/roundtrip_comments.ipynb b/notebooks/roundtrip_comments.ipynb
@@ -239,7 +239,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.2"
+   "version": "3.8.2"
   }
  },
  "nbformat": 4,

diff --git a/poetry.lock b/poetry.lock
diff --git a/pyproject.toml b/pyproject.toml
@@ -1,7 +1,7 @@
 [tool.poetry]
 
 name = "datafiles"
-version = "0.10b2"
+version = "0.10b3"
 description = "File-based ORM for dataclasses."
 
 license = "MIT"

diff --git a/tests/test_loading.py b/tests/test_loading.py
@@ -79,6 +79,19 @@ def with_extra_fields(sample, expect):
 
         expect(hasattr(sample, 'extra')).is_(False)
 
+    def with_invalid_data(sample, expect):
+        write(
+            'tmp/sample.yml',
+            """
+            - foo
+            - bar
+            """,
+        )
+
+        sample.datafile.load()
+
+        expect(sample.int_) == 0
+
 
 def describe_alternate_formats():
     @pytest.fixture