from collections import deque
from logging import getLogger

import numpy as np
from scipy import signal

_logger = getLogger(__name__)

buffer_dark = deque()
buffer_savgol = deque()
initialized = False


def initialize(params):
    global buffer_dark, buffer_savgol, initialized

    buffer_dark = deque(maxlen=params["buffer_length"])
    buffer_savgol = deque(maxlen=params["buffer_length"])

    initialized = True


def find_edge(data, step_length=50, edge_type="falling", roi=None):
    """
    Find an edge in the given data using cross-correlation with a step waveform.
    If a region-of-interest (roi) is provided as [start, end], the search is limited
    to that slice of the data and the resulting edge position is offset appropriately.
    """
    # If ROI is provided, slice the data accordingly.
    if roi is not None:
        data_roi = data[roi[0]:roi[1]]
    else:
        data_roi = data

    # Prepare the step function.
    step_waveform = np.ones(shape=(step_length,))
    if edge_type == "rising":
        step_waveform[: int(step_length / 2)] = -1
    elif edge_type == "falling":
        step_waveform[int(step_length / 2):] = -1

    # Perform cross-correlation on the (possibly sliced) data.
    xcorr = signal.correlate(data_roi, step_waveform, mode="valid")
    edge_position = np.argmax(xcorr)
    xcorr_amplitude = np.amax(xcorr)

    # Correct edge_position for step_length.
    edge_position += np.floor(step_length / 2)

    # If ROI is provided, add the offset to get the position in the full signal.
    if roi is not None:
        edge_position += roi[0]

    return {
        "edge_pos": edge_position,
        "xcorr": xcorr,
        "xcorr_ampl": xcorr_amplitude,
        "signal": data,
    }


def process(data, pulse_id, timestamp, params):
    device = params["device"]
    step_length = params["step_length"]
    edge_type = params["edge_type"]
    dark_event = params["dark_event"]
    fel_on_event = params["fel_on_event"]
    calib = params["calib"]
    filter_window = params["filter_window"]

    prof_sig = data[params["prof_sig"]]
    events = data[params["events"]]

    if not initialized:
        initialize(params)

    prof_sig_savgol = signal.savgol_filter(prof_sig, filter_window, 3)

    if events[dark_event]:
        buffer_dark.append(prof_sig)
        buffer_savgol.append(prof_sig_savgol)

    if buffer_savgol:
        prof_sig_norm = prof_sig_savgol / np.mean(np.array(buffer_savgol), axis=0)
    else:
        prof_sig_norm = prof_sig_savgol

    if events[fel_on_event] and not events[dark_event]:
        # Limit the edge search to the region-of-interest provided in params.
        edge_results = find_edge(prof_sig_norm, step_length, edge_type, roi=params.get("roi"))
        edge_results["arrival_time"] = np.polyval(calib, edge_results["edge_pos"])
    else:
        edge_results = {
            "edge_pos": None,
            "xcorr": None,
            "xcorr_ampl": None,
            "signal": prof_sig_norm,
        }
        edge_results["arrival_time"] = None

    # Set beam-synchronization outputs.
    output = {}
    for key, value in edge_results.items():
        output[f"{device}:{key}"] = value

    output[f"{device}:raw_wf"] = prof_sig
    output[f"{device}:raw_wf_savgol"] = prof_sig_savgol

    if buffer_dark:
        output[f"{device}:avg_dark_wf"] = np.mean(buffer_dark, axis=0)
    else:
        output[f"{device}:avg_dark_wf"] = None

    return output