improve: handle comma formatted kmms by default

Author: FelixAbrahamsson

kmm/positions/positions.py

@@ -18,6 +18,7 @@ class Config:
     def from_path(
         path: Path,
         raise_on_malformed_data: bool = True,
+        replace_commas: bool=True,
     ):
         """
         Loads positions from .kmm or .kmm2 file.
@@ -26,7 +27,7 @@ def from_path(
             dataframe = kmm.positions.read_kmm(path)
         elif path.suffix == ".kmm2":
             dataframe = kmm.positions.read_kmm2(
-                path, raise_on_malformed_data=raise_on_malformed_data
+                path, raise_on_malformed_data=raise_on_malformed_data, replace_commas=replace_commas
             )
         else:
             raise ValueError(f"Unable to parse file type {path.suffix}")
@@ -40,14 +41,15 @@ def read_sync_adjust(
         header_path: Path,
         adjustment: kmm.PositionAdjustment = kmm.PositionAdjustment.WIRE_CAMERA,
         raise_on_malformed_data: bool = True,
+        replace_commas: bool=True,
     ):
         """
         Loads positions from .kmm or .kmm2 file + .hdr file, then performs
         frame index sync, position adjustment and geodetic coordinate transformation.
         """
         header = kmm.Header.from_path(header_path, raise_on_malformed_data)
         return (
-            Positions.from_path(kmm_path, raise_on_malformed_data)
+            Positions.from_path(kmm_path, raise_on_malformed_data=raise_on_malformed_data, replace_commas=replace_commas)
             .sync_frame_index(header, adjustment, raise_on_malformed_data)
             .geodetic()
         )

kmm/positions/read_kmm.py

@@ -1,3 +1,4 @@
+from io import StringIO
 from pathlib import Path
 
 import numpy as np
@@ -6,12 +7,19 @@
 
 
 @validate_arguments
-def read_kmm(path: Path):
+def read_kmm(path: Path, replace_commas: bool=True):
     try:
+        if replace_commas:
+            with open(path, "r", encoding="latin1") as f:
+                content = f.read()
+            content = content.replace(",", ".")
+            file_obj = StringIO(content)
+        else:
+            file_obj = path
         return pd.read_csv(
-            path,
+            file_obj,
             sep="\t",
-            encoding="latin1",
+            encoding="latin1" if not replace_commas else None,
             names=[
                 "centimeter",
                 "track_section",

kmm/positions/read_kmm2.py

@@ -1,4 +1,5 @@
 import re
+from io import StringIO
 from pathlib import Path
 
 import numpy as np
@@ -46,7 +47,7 @@
 
 
 @validate_arguments
-def read_kmm2(path: Path, raise_on_malformed_data: bool = True):
+def read_kmm2(path: Path, raise_on_malformed_data: bool = True, replace_commas: bool=True):
     skiprows = [
         index
         for index, line in enumerate(path.read_text(encoding="latin1").splitlines())
@@ -58,13 +59,22 @@ def read_kmm2(path: Path, raise_on_malformed_data: bool = True):
             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:
+        if replace_commas:
+            with open(path, "r", encoding="latin1") as f:
+                content = f.read()
+            content = content.replace(",", ".")
+            file_obj = StringIO(content)
+        else:
+            file_obj = path
+            
         try:
             df = pd.read_csv(
-                path,
+                file_obj,
                 skiprows=skiprows,
                 delimiter="\t",
-                encoding="latin1",
+                encoding="latin1" if not replace_commas else None,
                 low_memory=False,
                 header=None,
             )