get_sub does not work with D107A because it has a letter in its name

[jimzhang629]

def get_sub(inst: Signal | mne.Info | str) -> str:
    """Gets the subject from the instance

    Parameters
    ----------
    inst : Signal
        The instance to get the subject from

    Returns
    -------
    str
        The subject"""
    if isinstance(inst, Signal):
        inst = inst.info
    elif isinstance(inst, str):
        return f"{inst[0]}{int(inst[1:])}"
    out_str = inst['subject_info']['his_id'][4:]
    if len(out_str) == 1:
        return out_str
    return out_str[0] + str(int("".join(s for s in out_str if s.isdigit())))

The current implementation of get_sub (located in ieeg/viz/mri.py) will strip away the letters in a subject name, if they exist. For example, inputting D0107A will return D107, not D107A.

I propose a new implementation of get_sub below that just strips the leading zeroes. It will strip zeroes until it encounters a non-zero value. This should keep the suffix intact. For example, inputting D0107A will return D107A.

import re

def get_sub(inst: Signal | mne.Info | str) -> str:
    """Gets the subject ID from the instance, removes leading zeros from the numeric parts,
    and preserves any letters (e.g., 'D0107A' becomes 'D107A', 'D0059A' becomes 'D59A').

    Parameters
    ----------
    inst : Signal | mne.Info | str
        The instance (Signal, Info, or string) to get the subject ID from.

    Returns
    -------
    str
        The subject ID with leading zeros removed from the numeric parts (e.g., 'D0107A' becomes 'D107A').
    """
    if isinstance(inst, Signal):
        inst = inst.info
    
    # If the instance is a string, return it after processing
    if isinstance(inst, str):
        # Use regex to match and remove leading zeros from the first numeric part
        return re.sub(r'(\D*)(0*)(\d+)', lambda m: m.group(1) + m.group(3), inst)

    # Extract subject ID from 'subject_info'
    out_str = inst['subject_info']['his_id'][4:]

    # Use regex to match and remove leading zeros from the first numeric part
    return re.sub(r'(\D*)(0*)(\d+)', lambda m: m.group(1) + m.group(3), out_str)