Change column names to match schema casing

Author: glopesdev

harp/io.py

@@ -61,7 +61,7 @@ def read(
     data = np.fromfile(file, dtype=np.uint8)
     if len(data) == 0:
         return pd.DataFrame(
-            columns=columns, index=pd.Index([], dtype=np.float64, name="time")
+            columns=columns, index=pd.Index([], dtype=np.float64, name="Time")
         )
 
     if address is not None and address != data[2]:
@@ -86,7 +86,7 @@ def read(
         if epoch is not None:
             time = epoch + pd.to_timedelta(time, "s")  # type: ignore
         index = pd.Series(time)
-        index.name = "time"
+        index.name = "Time"
 
     payloadsize = stride - payloadoffset - 1
     payloadtype = _payloadtypes[payloadtype]
@@ -112,5 +112,5 @@ def read(
             nrows, dtype=np.uint8, buffer=data, offset=0, strides=stride
         )
         msgtype = pd.Categorical.from_codes(msgtype, categories=_messagetypes)  # type: ignore
-        result["type"] = msgtype
+        result[MessageType.__name__] = msgtype
     return result

harp/reader.py

@@ -1,5 +1,4 @@
 import os
-import re
 from math import log2
 from os import PathLike
 from pathlib import Path
@@ -11,11 +10,9 @@
 from typing import Any, BinaryIO, Callable, Iterable, Optional, Protocol, Union
 from pandas._typing import Axes
 from harp.model import BitMask, GroupMask, Model, PayloadMember, Register
-from harp.io import read
+from harp.io import MessageType, read
 from harp.schema import read_schema
 
-_camel_to_snake_regex = re.compile(r"(?<!^)(?=[A-Z])")
-
 
 @dataclass
 class _ReaderParams:
@@ -75,18 +72,14 @@ def parser(
     ):
         df = g(file, columns, epoch, keep_type)
         result = f(df)
-        type_col = df.get("type")
+        type_col = df.get(MessageType.__name__)
         if type_col is not None:
-            result["type"] = type_col
+            result[MessageType.__name__] = type_col
         return result
 
     return parser
 
 
-def _id_camel_to_snake(id: str):
-    return _camel_to_snake_regex.sub("_", id).lower()
-
-
 def _create_bit_parser(mask: int):
     def parser(xs: Series) -> Series:
         return (xs & mask) != 0
@@ -95,10 +88,7 @@ def parser(xs: Series) -> Series:
 
 
 def _create_bitmask_parser(bitMask: BitMask):
-    lookup = [
-        (_id_camel_to_snake(k), _create_bit_parser(int(v.root)))
-        for k, v in bitMask.bits.items()
-    ]
+    lookup = [(k, _create_bit_parser(int(v.root))) for k, v in bitMask.bits.items()]
 
     def parser(df: DataFrame):
         return DataFrame({n: f(df[0]) for n, f in lookup}, index=df.index)
@@ -203,7 +193,7 @@ def _create_register_parser(device: Model, name: str, params: _ReaderParams):
 
     if register.payloadSpec is not None:
         payload_parsers = [
-            (_id_camel_to_snake(key), _create_payloadmember_parser(device, member))
+            (key, _create_payloadmember_parser(device, member))
             for key, member in register.payloadSpec.items()
         ]
 
@@ -213,8 +203,7 @@ def parser(df: DataFrame):
         reader = _compose_parser(parser, reader, params)
         return RegisterReader(register, reader)
 
-    columns = [_id_camel_to_snake(name)]
-    reader = partial(reader, columns=columns)
+    reader = partial(reader, columns=[name])
     return RegisterReader(register, reader)
 
 

tests/test_io.py

@@ -2,7 +2,7 @@
 import numpy as np
 from contextlib import nullcontext
 from pytest import mark
-from harp.io import read
+from harp.io import read, MessageType
 from tests.params import DataFileParam
 
 testdata = [
@@ -45,7 +45,10 @@ def test_read(dataFile: DataFileParam):
         )
         assert len(data) == dataFile.expected_rows
         if dataFile.keep_type:
-            assert "type" in data.columns and data["type"].dtype == "category"
+            assert (
+                MessageType.__name__ in data.columns
+                and data[MessageType.__name__].dtype == "category"
+            )
 
         if dataFile.expected_cols:
             for col in dataFile.expected_cols: