from datetime import datetime

import hf_data_loader


def group_changes(changes):
    groups = {}
    for change in changes:
        group = datetime.fromisoformat(change.ts)
        if group not in groups:
            groups[group] = []
        groups[group].append(change)

    grouped_changes = []
    for group in sorted(groups.keys()):
        grouped_changes.sort(key=lambda x: x.p)
        grouped_changes.append(groups[group])

    return grouped_changes


def get_annotated_diff(initial_text, changes):
    grouped_changes = group_changes(changes)
    text = [((c, " ") for c in initial_text)]
    for change_group in grouped_changes:
        text_pointer = 0
        change_pointer = 0
        while text_pointer < len(text):
            pass


df = hf_data_loader.load_raw_dataset_as_pandas()