Add is_filtered to annotations in binary.json

src/spikeinterface/core/baserecording.py

@@ -567,6 +567,7 @@ def _save(self, format="binary", verbose: bool = False, **save_kwargs):
                 channel_ids=self.get_channel_ids(),
                 time_axis=0,
                 file_offset=0,
+                is_filtered=self.is_filtered(),
                 gain_to_uV=self.get_channel_gains(),
                 offset_to_uV=self.get_channel_offsets(),
             )

src/spikeinterface/core/binaryrecordingextractor.py

@@ -68,10 +68,12 @@ def __init__(
         num_chan=None,
     ):
         # This assigns num_channels if num_channels is not None, otherwise num_chan is assigned
+        # num_chan needs to be be kept for backward compatibility but should not be used by the
+        # end user
         num_channels = num_channels or num_chan
         assert num_channels is not None, "You must provide num_channels or num_chan"
         if num_chan is not None:
-            warnings.warn("`num_chan` is to be deprecated in version 0.100, please use `num_channels` instead")
+            warnings.warn("`num_chan` is to be deprecated as of version 0.100, please use `num_channels` instead")
 
         if channel_ids is None:
             channel_ids = list(range(num_channels))

src/spikeinterface/core/tests/test_binaryrecordingextractor.py

@@ -93,6 +93,8 @@ def test_sequential_reading_of_small_traces(folder_with_binary_files):
     dtype = "float32"
 
     file_paths = [folder / "traces_cached_seg0.raw"]
+    # `num_chan` is kept for backward compatibility so including it at least one test
+    # run is good to ensure that it is appropriately accepted as an argument
     recording = BinaryRecordingExtractor(
         num_chan=num_channels,
         file_paths=file_paths,