Fix legacy manifest test expectations

Exclude V3-only properties from V1 and V2 fastavro comparisons and clarify the latest read-version constant name.

Author: KaiqiJinWow

pyiceberg/manifest.py

@@ -56,7 +56,7 @@
 UNASSIGNED_SEQ = -1
 DEFAULT_BLOCK_SIZE = 67108864  # 64 * 1024 * 1024
 DEFAULT_READ_VERSION: Literal[2] = 2
-_LATEST_MANIFEST_VERSION: Literal[3] = 3
+_LATEST_MANIFEST_READ_VERSION: Literal[3] = 3
 
 INITIAL_SEQUENCE_NUMBER = 0
 
@@ -894,7 +894,7 @@ def fetch_manifest_entry(self, io: FileIO, discard_deleted: bool = True) -> list
         input_file = io.new_input(self.manifest_path)
         with AvroFile[ManifestEntry](
             input_file,
-            MANIFEST_ENTRY_SCHEMAS[_LATEST_MANIFEST_VERSION],
+            MANIFEST_ENTRY_SCHEMAS[_LATEST_MANIFEST_READ_VERSION],
             read_types={-1: ManifestEntry, 2: DataFile},
             read_enums={0: ManifestEntryStatus, 101: FileFormat, 134: DataFileContent},
         ) as reader:
@@ -1017,7 +1017,7 @@ def read_manifest_list(input_file: InputFile) -> Iterator[ManifestFile]:
     """
     with AvroFile[ManifestFile](
         input_file,
-        MANIFEST_LIST_FILE_SCHEMAS[_LATEST_MANIFEST_VERSION],
+        MANIFEST_LIST_FILE_SCHEMAS[_LATEST_MANIFEST_READ_VERSION],
         read_types={-1: ManifestFile, 508: PartitionFieldSummary},
         read_enums={517: ManifestContent},
     ) as reader:

tests/avro/test_file.py

@@ -164,6 +164,8 @@ def test_write_manifest_entry_with_iceberg_read_with_fastavro_v1() -> None:
         del v2_entry["file_sequence_number"]
         del v2_entry["data_file"]["content"]
         del v2_entry["data_file"]["equality_ids"]
+        for field in ("first_row_id", "referenced_data_file", "content_offset", "content_size_in_bytes"):
+            del v2_entry["data_file"][field]
 
         # Required in V1
         v2_entry["data_file"]["block_size_in_bytes"] = DEFAULT_BLOCK_SIZE
@@ -222,7 +224,11 @@ def test_write_manifest_entry_with_iceberg_read_with_fastavro_v2() -> None:
 
             fa_entry = next(it)
 
-        assert todict(entry) == fa_entry
+        v2_entry = todict(entry)
+        for field in ("first_row_id", "referenced_data_file", "content_offset", "content_size_in_bytes"):
+            del v2_entry["data_file"][field]
+
+        assert v2_entry == fa_entry
 
 
 @pytest.mark.parametrize("format_version", [1, 2])