feat: Use csv.DictReader to parse header fields (msto#19)

dataclass_io/_lib/assertions.py

@@ -100,18 +100,19 @@ def assert_file_header_matches_dataclass(
     dataclass_type: type[DataclassInstance],
     delimiter: str,
     comment_prefix: str,
+    quoting: int,
 ) -> None:
     """
     Check that the specified file has a header and its fields match those of the provided dataclass.
     """
     header: FileHeader | None
     if isinstance(file, Path):
         with file.open("r") as fin:
-            header = get_header(fin, delimiter=delimiter, comment_prefix=comment_prefix)
+            header = get_header(fin, delimiter=delimiter, comment_prefix=comment_prefix, quoting=quoting)
     else:
         pos = file.tell()
         try:
-            header = get_header(file, delimiter=delimiter, comment_prefix=comment_prefix)
+            header = get_header(file, delimiter=delimiter, comment_prefix=comment_prefix, quoting=quoting)
         finally:
             file.seek(pos)
 

dataclass_io/_lib/file.py

@@ -1,3 +1,4 @@
+from csv import DictReader
 from dataclasses import dataclass
 from enum import Enum
 from enum import unique
@@ -68,6 +69,7 @@ def get_header(
     reader: ReadableFileHandle,
     delimiter: str,
     comment_prefix: str,
+    quoting: int,
 ) -> Optional[FileHeader]:
     """
     Read the header from an open file.
@@ -85,6 +87,7 @@ def get_header(
     Args:
         reader: An open, readable file handle.
         comment_char: The character which indicates the start of a comment line.
+        quoting: Quoting style (enum value from Python csv package).
 
     Returns:
         A `FileHeader` containing the field names and any preceding lines.
@@ -103,6 +106,12 @@ def get_header(
     else:
         return None
 
-    fieldnames = line.strip().split(delimiter)
+    '''
+    msto#19 Read header fields
+
+    Use csv.DictReader because RFC4180 is tricky to implement correctly
+    '''
+    header_reader = DictReader([line], delimiter=delimiter, quoting=quoting)
+    fieldnames = header_reader.fieldnames
 
     return FileHeader(preface=preface, fieldnames=fieldnames)

dataclass_io/reader.py

@@ -1,3 +1,4 @@
+import csv
 from contextlib import contextmanager
 from csv import DictReader
 from pathlib import Path
@@ -27,6 +28,7 @@ def __init__(
         dataclass_type: type[DataclassInstance],
         delimiter: str = "\t",
         comment_prefix: str = "#",
+        quoting: int = csv.QUOTE_MINIMAL,
         **kwds: Any,
     ) -> None:
         """
@@ -35,6 +37,7 @@ def __init__(
             dataclass_type: Dataclass type.
             delimiter: The input file delimiter.
             comment_prefix: The prefix for any comment/preface rows preceding the header row.
+            quoting: Quoting style (enum value from Python csv package).
             dataclass_type: Dataclass type.
 
         Raises:
@@ -46,17 +49,19 @@ def __init__(
             dataclass_type=dataclass_type,
             delimiter=delimiter,
             comment_prefix=comment_prefix,
+            quoting=quoting,
         )
 
         self._dataclass_type = dataclass_type
         self._fin = fin
         self._header = get_header(
-            reader=self._fin, delimiter=delimiter, comment_prefix=comment_prefix
+            reader=self._fin, delimiter=delimiter, comment_prefix=comment_prefix, quoting=quoting
         )
         self._reader = DictReader(
             f=self._fin,
             fieldnames=fieldnames(dataclass_type),
             delimiter=delimiter,
+            quoting=quoting,
         )
 
     def __iter__(self) -> "DataclassReader":

dataclass_io/writer.py

@@ -1,3 +1,4 @@
+import csv
 from contextlib import contextmanager
 from csv import DictWriter
 from dataclasses import asdict
@@ -126,6 +127,7 @@ def open(
         overwrite: bool = True,
         delimiter: str = "\t",
         comment_prefix: str = "#",
+        quoting: int = csv.QUOTE_MINIMAL,
         **kwds: Any,
     ) -> Iterator["DataclassWriter"]:
         """
@@ -146,6 +148,7 @@ def open(
             comment_prefix: The prefix for any comment/preface rows preceding the header row.
                 (This argument is ignored when `mode="write"`. It is used when `mode="append"` to
                 validate that the existing file's header matches the specified dataclass.)
+            quoting: Quoting style (enum value from Python csv package).
             **kwds: Additional keyword arguments to be passed to the `DataclassWriter` constructor.
 
         Yields:
@@ -178,6 +181,7 @@ def open(
                 dataclass_type=dataclass_type,
                 delimiter=delimiter,
                 comment_prefix=comment_prefix,
+                quoting=quoting,
             )
 
         fout = filepath.open(write_mode.abbreviation)

tests/test_reader.py

@@ -35,3 +35,30 @@ class FakeDataclass:
         assert isinstance(rows[0], FakeDataclass)
         assert rows[0].foo == "abc"
         assert rows[0].bar == 1
+
+
+def test_read_csv_with_header_quotes(tmp_path: Path) -> None:
+    """
+    Test that having quotes around column names in header row doesn't break anything
+    https://github.com/msto/dataclass_io/issues/19
+    """
+    fpath = tmp_path / "test.txt"
+
+    @dataclass
+    class FakeDataclass:
+        id: str
+        title: str
+
+    test_csv = [
+        '"id"\t"title"\n',
+        '"fake"\t"A fake object"\n',
+        '"also_fake"\t"Another fake object"\n',
+    ]
+
+    with fpath.open("w") as f:
+        f.writelines(test_csv)
+
+    # Parse CSV using DataclassReader
+    with DataclassReader.open(fpath, FakeDataclass) as reader:
+        for fake_object in reader:
+            print(fake_object)