Source code for gslides.utils

# -*- coding: utf-8 -*-
import datetime
import re
from decimal import Decimal
from typing import Any, Dict, List, Tuple, Union

import numpy as np
import pandas as pd

from .config import CHART_PARAMS


def json_val_extract(obj: Dict[str, Any], key: str) -> List[Any]:
    """Recursively find values based on a given key

    :param obj: JSON to search
    :type obj: dict
    :param key: Key that corresponds to the value to find
    :type key: str
    :return: Value to return
    :rtype: any

    """
    arr: List = []

    def extract(obj: Dict[str, Any], arr: List, key: str) -> List:
        """Recursively search for keys in JSON tree."""
        if isinstance(obj, dict):
            for k, v in obj.items():
                if k == key:
                    arr.append(v)
                else:
                    extract(v, arr, key)
        elif isinstance(obj, list):
            for item in obj:
                extract(item, arr, key)
        return arr

    values = extract(obj, arr, key)
    return values


def json_chunk_extract(
    obj: Dict[str, Any], key: str, val: Union[str, int, float]
) -> List:
    """Recursively fetch chunks from nested JSON based on a given key, value pair.

    :param obj: JSON to search
    :type obj: dict
    :param key: Key that corresponds to the dictionary to chunk
    :type key: str
    :param val: Val that corresponds to the dictionary to chunk
    :type val: str, int, float
    :return: List of chunks
    :rtype: list

    """
    arr: List = []

    def extract(obj: Dict[str, Any], arr: List, val: Union[str, int, float]) -> List:
        """Recursively search for keys in JSON tree."""
        if isinstance(obj, dict):
            if (key, val) in obj.items():
                arr.append(obj)
            else:
                for k, v in obj.items():
                    extract(v, arr, val)
        elif isinstance(obj, list):
            for item in obj:
                extract(item, arr, val)
        return arr

    values = extract(obj, arr, val)
    return values


def json_chunk_key_extract(obj: Dict[str, Any], key: str) -> List:
    """Recursively fetch chunks from nested JSON based on a given key.

    :param obj: JSON to search
    :type obj: dict
    :param key: Key that corresponds to the dictionary to chunk
    :type key: str
    :return: List of chunks
    :rtype: list

    """
    arr: List = []

    def extract(obj: Dict[str, Any], arr: List) -> List:
        """Recursively search for keys in JSON tree."""
        if isinstance(obj, dict):
            if key in obj.keys():
                arr.append(obj)
            else:
                for k, v in obj.items():
                    extract(v, arr)
        elif isinstance(obj, list):
            for item in obj:
                extract(item, arr)
        return arr

    values = extract(obj, arr)
    return values


def json_dict_extract(
    obj: Dict[str, Any],
    keys: Tuple,
) -> Dict:
    """Recursively fetch chunks from nested JSON based on a given key, value pair.

    :param obj: JSON to search
    :type obj: dict
    :param keys: Tuple of multiple keys to search for in dictionary
    :type keys: tuple
    :return: Dictionary of values
    :rtype: dict

    """
    arr: Dict = {}

    def extract(obj: Dict[str, Any], arr: Dict, keys: Tuple) -> Dict:
        """Recursively search for keys in JSON tree."""
        if isinstance(obj, dict):
            if set(keys) <= set(obj.keys()):
                arr[obj[keys[0]]] = obj[keys[1]]
            else:
                for k, v in obj.items():
                    extract(v, arr, keys)
        elif isinstance(obj, list):
            for item in obj:
                extract(item, arr, keys)
        return arr

    values = extract(obj, arr, keys)
    return values


def num_to_char(x: int) -> str:
    """Converts a number to a character

    :param x: Number
    :type x: int
    :return: Corresponding character
    :rtype: str

    """
    if x <= 26:
        return chr(x + 64).upper()
    elif x <= 26 * 26:
        return f"{chr(x//26+64).upper()}{chr(x%26+64).upper()}"
    else:
        raise ValueError("Integer too high to be converted")


def char_to_num(x: str) -> int:
    """Converts a character to a number

    :param x: Character
    :type x: str
    :return: Corresponding number
    :rtype: int

    """
    total = 0
    for i in range(len(x)):
        total += (ord(x[::-1][i]) - 64) * (26 ** i)
    return total


def cell_to_num(x: str) -> Tuple[int, int]:
    """Converts a cell to a row, column index

    :param x: Name of the cell (e.g. A2)
    :type x: str
    :return: Row, column index
    :rtype: tuple

    """
    regex = re.compile("([A-Z]*)([0-9]*)")
    output = regex.search(x)
    if output:
        column = output.group(1)
        row = int(output.group(2))
        return (row, char_to_num(column))
    else:
        raise ValueError("Invalid cell format")


def validate_hex_color_code(x: str) -> str:
    """Short summary.

    :param x: Hexadecimal color code
    :type x: str
    :raises ValueError: Input a hexadecimal color code
    :return: Hexadecimal color code
    :rtype: str

    """
    match = re.search("^#(?:[0-9a-fA-F]{3}){1,2}$", x)
    if match:
        return x
    else:
        raise ValueError("Input a valid hexadecimal color code")


def hex_to_rgb(x: str) -> Tuple[float, ...]:
    """Converts a xexadecimal color code to r,g,b

    :param x: Hexadecimal color code
    :type x: str
    :return: R,G,B color code
    :rtype: Tuple

    """
    x = x[1:]
    return tuple(int(x[i : i + 2], 16) / 255 for i in (0, 2, 4))  # noqa


def emu_to_px(x: int) -> int:
    """Converts EMU to pixels

    :param x: EMU
    :type x: int
    :return: Pixels
    :rtype: int

    """
    return int(x * 220 / (914400))


def optimize_size(y_scale: float, area: float = 222600) -> Tuple[float, float]:
    """Optimizes the size of a chart so that it matches the suggested area (222600 pixels)

    :param y_scale: Scale of the x axis to y axis
    :type y_scale: float
    :param float area: Description of parameter `area`.
    :type area: float, optional
    :return: x and y length combination
    :rtype: tuple

    """
    x_length = (area / (y_scale)) ** 0.5
    return (x_length, x_length * y_scale)


def clean_list_of_list(x: List[List]) -> List[List]:
    """In a list of list with different length lists, appends None values to equalize
    the length of each list

    :param x: List of lists
    :type x: list
    :return: List of lists
    :rtype: list

    """
    max_len = max([len(i) for i in x])
    for i in x:
        i.extend([None] * (max_len - len(i)))
    return x


def clean_nan(df: pd.DataFrame) -> pd.DataFrame:
    """Replaces NaN's in a pandas dataframe

    :param df: :class:`pandas.DataFrame`
    :type df: :class:`pandas.DataFrame`
    :return: :class:`pandas.DataFrame`
    :rtype: :class:`pandas.DataFrame`

    """
    return df.replace({np.nan: None})


def clean_dtypes(x: Any) -> Union[str, float, int, np.int64, np.float64, None]:
    """Cleans the datatypes of an obersevation to either int, float or string or None

    :param x: Observation to clean
    :type x: any
    :return: Clean observation
    :rtype: str, float, int, np.int64, np.float64, None

    """
    if type(x) in [pd._libs.tslibs.timestamps.Timestamp, datetime.date]:
        return str(x)
    elif type(x) in [Decimal]:
        return float(str(x))
    elif type(x) in [str, int, float, np.int64, np.float64, type(None)]:
        return x
    elif type(x) in [pd._libs.missing.NAType, pd._libs.tslibs.nattype.NaTType]:
        return None
    else:
        raise TypeError(
            f"{type(x)} is not an accepted datatype. Type must conform to "
            "str, int, float, NoneType, decimal.Decimal, pd.Timestamp, datetime.date"
        )


def validate_params_list(params: dict) -> None:
    """Validates the parameters for the chart based on a list

    :param params: Dictionary of parameters
    :type params: dict
    :raises ValueError:

    """
    for key, val in CHART_PARAMS.items():
        if key in params.keys() and params[key]:
            if params[key] not in val["params"]:
                raise ValueError(
                    f"{ params[key]} is not a valid parameter for {key}. "
                    f"Accepted parameters include {', '.join(val['params'])}. See "
                    f"{val['url']} for further documentation."
                )


def validate_params_int(params: dict) -> None:
    """Validates the parameters for the chart based on the integer value

    :param params: Dictionary of parameters
    :type params: dict
    :raises ValueError:

    """
    variables = ["line_width", "point_size", "bucket_size"]
    for var in variables:
        if var in params.keys() and params[var]:
            if type(params[var]) != int or params[var] < 0:
                raise ValueError(
                    f"{params[var]} is not a valid parameter for {var}. "
                    f"Accepted values are any integer greater than 0"
                )


def validate_params_float(params: dict) -> None:
    """Validates the parameters for the chart based on the float value

    :param params: Dictionary of parameters
    :type params: dict
    :raises ValueError:

    """
    variables = ["outlier_percentage", "x_border", "y_border", "spacing"]
    for var in variables:
        if var in params.keys() and params[var]:
            if params[var] >= 1 or params[var] < 0:
                raise ValueError(
                    f"{params[var]} is not a valid parameter for {var}. "
                    f"Accepted values are any float between 0 and 1"
                )


def validate_series_columns(params: dict) -> None:
    """Validates that the series column is None or a list

    :param params: Dictionary of parameters
    :type params: dict
    :raises ValueError:

    """
    if "series_columns" in params.keys() and params["series_columns"]:
        if type(params["series_columns"]) != list:
            raise ValueError(
                f"{params['series_columns']} is not a valid parameter for series_columns."
                f"Series columns only accepts a list or None."
            )


def validate_cell_name(x: str) -> str:
    """Validates that a cell name is valid

    :param x: Cell name
    :type x: str
    :type params: dict
    :raises ValueError: Invalid cell name.
    :return: Cell name
    :rtype: str
    """
    pattern = re.compile("([A-Z]{1,2})([0-9]+)")
    output = pattern.search(x)
    if output:
        if output[0] == x:
            return x
        else:
            raise ValueError("Invalid cell name.")
    else:
        raise ValueError("Invalid cell name.")


def determine_col_proportion(df: pd.DataFrame) -> np.ndarray:
    """Determines the percent size of a column based on the length of observiations

    :param type df: Dataframe that will become a table
    :type df: pd.DataFrame
    :return: An array of proportions
    :rtype: np.ndarray

    """
    col_size = df.apply(
        lambda x: max(x.astype("str").apply(lambda y: len(y))), axis=0
    ).values
    per_col_size = col_size / sum(col_size)
    return per_col_size


def black_or_white(rgb: Tuple[float, ...]) -> Tuple[float, ...]:
    """Determines based on the luminosity of a color wether the text on top of
    that color should be black or white. See the following:
    https://en.wikipedia.org/wiki/Luminance_%28relative%29

    :param rgb: The rgb values of the color
    :type rgb: tuple
    :return: Black or white rgb
    :rtype: tuple

    """
    if rgb[0] * 0.2126 + rgb[1] * 0.7152 + rgb[2] * 0.0722 > 0.5:
        return (0, 0, 0)
    else:
        return (1, 1, 1)