fix: unnecessary skipping of lines in kmm

Author: FelixAbrahamsson

kmm/positions/positions.py

@@ -15,14 +15,19 @@ class Config:
 
     @staticmethod
     @validate_arguments
-    def from_path(path: Path):
+    def from_path(
+        path: Path,
+        raise_on_malformed_data: bool = True,
+    ):
         """
         Loads positions from .kmm or .kmm2 file.
         """
         if path.suffix == ".kmm":
             dataframe = kmm.positions.read_kmm(path)
         elif path.suffix == ".kmm2":
-            dataframe = kmm.positions.read_kmm2(path)
+            dataframe = kmm.positions.read_kmm2(
+                path, raise_on_malformed_data=raise_on_malformed_data
+            )
         else:
             raise ValueError(f"Unable to parse file type {path.suffix}")
 
@@ -42,7 +47,7 @@ def read_sync_adjust(
         """
         header = kmm.Header.from_path(header_path, raise_on_malformed_data)
         return (
-            Positions.from_path(kmm_path)
+            Positions.from_path(kmm_path, raise_on_malformed_data)
             .sync_frame_index(header, adjustment, raise_on_malformed_data)
             .geodetic()
         )

kmm/positions/read_kmm2.py

@@ -46,21 +46,27 @@
 
 
 @validate_arguments
-def read_kmm2(path: Path):
+def read_kmm2(path: Path, raise_on_malformed_data: bool = True):
     skiprows = [
         index
         for index, line in enumerate(path.read_text(encoding="latin1").splitlines())
         if pattern.match(line) or pattern2.match(line)
     ]
-
+    with open(path, "r", encoding="latin1") as f:
+        line = f.readline()
+        if line.startswith("VER"):
+            skiprows = [0] + skiprows
+        elif raise_on_malformed_data and not line.startswith("POS"):
+            raise ValueError("Malformed data, first line is not POS or VER")
     try:
         try:
             df = pd.read_csv(
                 path,
-                skiprows=[0] + skiprows,
+                skiprows=skiprows,
                 delimiter="\t",
                 encoding="latin1",
                 low_memory=False,
+                header=None,
             )
         except pd.errors.EmptyDataError:
             return pd.DataFrame(columns=expected_columns)