etlplus 0.10.1__tar.gz → 0.10.4__tar.gz

{etlplus-0.10.1/etlplus.egg-info → etlplus-0.10.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: etlplus
-Version: 0.10.1
+Version: 0.10.4
 Summary: A Swiss Army knife for simple ETL operations
 Home-page: https://github.com/Dagitali/ETLPlus
 Author: ETLPlus Team

{etlplus-0.10.1 → etlplus-0.10.4}/etlplus/enums.py

@@ -300,12 +300,16 @@ class FileFormat(CoercibleStrEnum):
             '.yml': 'yaml',
             # MIME types
             'application/avro': 'avro',
+            'application/csv': 'csv',
             'application/feather': 'feather',
             'application/gzip': 'gz',
             'application/json': 'json',
             'application/jsonlines': 'ndjson',
             'application/ndjson': 'ndjson',
             'application/orc': 'orc',
+            'application/parquet': 'parquet',
+            'application/vnd.apache.avro': 'avro',
+            'application/vnd.apache.parquet': 'parquet',
             'application/vnd.apache.arrow.file': 'feather',
             'application/vnd.apache.orc': 'orc',
             'application/vnd.ms-excel': 'xls',
@@ -314,13 +318,20 @@ class FileFormat(CoercibleStrEnum):
                 'officedocument.spreadsheetml.sheet'
             ): 'xlsx',
             'application/x-avro': 'avro',
+            'application/x-csv': 'csv',
+            'application/x-feather': 'feather',
+            'application/x-orc': 'orc',
             'application/x-ndjson': 'ndjson',
             'application/x-parquet': 'parquet',
+            'application/x-yaml': 'yaml',
             'application/xml': 'xml',
             'application/zip': 'zip',
             'text/csv': 'csv',
             'text/plain': 'txt',
             'text/tab-separated-values': 'tsv',
+            'text/tsv': 'tsv',
+            'text/xml': 'xml',
+            'text/yaml': 'yaml',
         }
 
 
@@ -524,6 +535,7 @@ def coerce_http_method(
 
 def infer_file_format_and_compression(
     value: object,
+    filename: object | None = None,
 ) -> tuple[FileFormat | None, CompressionFormat | None]:
     """
     Infer data format and compression from a filename, extension, or MIME type.
@@ -532,6 +544,9 @@ def infer_file_format_and_compression(
     ----------
     value : object
         A filename, extension, MIME type, or existing enum member.
+    filename : object | None, optional
+        A filename to consult for extension-based inference (e.g. when
+        ``value`` is ``application/octet-stream``).
 
     Returns
     -------
@@ -552,10 +567,29 @@ def infer_file_format_and_compression(
     normalized = text.casefold()
     mime = normalized.split(';', 1)[0].strip()
 
+    is_octet_stream = mime == 'application/octet-stream'
     compression = CompressionFormat.try_coerce(mime)
-    fmt = FileFormat.try_coerce(mime)
-
-    suffixes = PurePath(text).suffixes
+    fmt = None if is_octet_stream else FileFormat.try_coerce(mime)
+
+    is_mime = mime.startswith(
+        (
+            'application/',
+            'text/',
+            'audio/',
+            'image/',
+            'video/',
+            'multipart/',
+        ),
+    )
+    suffix_source: object | None = filename if filename is not None else text
+    if is_mime and filename is None:
+        suffix_source = None
+
+    suffixes = (
+        PurePath(str(suffix_source)).suffixes
+        if suffix_source is not None
+        else []
+    )
     if suffixes:
         normalized_suffixes = [suffix.casefold() for suffix in suffixes]
         compression = (

{etlplus-0.10.1 → etlplus-0.10.4/etlplus.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: etlplus
-Version: 0.10.1
+Version: 0.10.4
 Summary: A Swiss Army knife for simple ETL operations
 Home-page: https://github.com/Dagitali/ETLPlus
 Author: ETLPlus Team

{etlplus-0.10.1 → etlplus-0.10.4}/tests/unit/test_u_enums.py

@@ -119,25 +119,33 @@ class TestInferFileFormatAndCompression:
     """Unit test suite for :func:`infer_file_format_and_compression`."""
 
     @pytest.mark.parametrize(
-        'value,expected_format,expected_compression',
+        'value,filename,expected_format,expected_compression',
         [
-            ('data.csv.gz', FileFormat.CSV, CompressionFormat.GZ),
-            ('data.jsonl.gz', FileFormat.NDJSON, CompressionFormat.GZ),
-            ('data.zip', None, CompressionFormat.ZIP),
-            ('application/json; charset=utf-8', FileFormat.JSON, None),
-            ('application/gzip', None, CompressionFormat.GZ),
-            (FileFormat.GZ, None, CompressionFormat.GZ),
-            (CompressionFormat.ZIP, None, CompressionFormat.ZIP),
+            ('data.csv.gz', None, FileFormat.CSV, CompressionFormat.GZ),
+            ('data.jsonl.gz', None, FileFormat.NDJSON, CompressionFormat.GZ),
+            ('data.zip', None, None, CompressionFormat.ZIP),
+            ('application/json; charset=utf-8', None, FileFormat.JSON, None),
+            ('application/gzip', None, None, CompressionFormat.GZ),
+            (
+                'application/octet-stream',
+                'payload.csv.gz',
+                FileFormat.CSV,
+                CompressionFormat.GZ,
+            ),
+            ('application/octet-stream', None, None, None),
+            (FileFormat.GZ, None, None, CompressionFormat.GZ),
+            (CompressionFormat.ZIP, None, None, CompressionFormat.ZIP),
         ],
     )
     def test_infers_format_and_compression(
         self,
         value: object,
+        filename: object | None,
         expected_format: FileFormat | None,
         expected_compression: CompressionFormat | None,
     ) -> None:
         """Test mixed inputs for format and compression inference."""
-        fmt, compression = infer_file_format_and_compression(value)
+        fmt, compression = infer_file_format_and_compression(value, filename)
         assert fmt is expected_format
         assert compression is expected_compression