mindey

# Author: Mindey
# Basic Baserow.io API Wrapper.

import os, requests, getpass
import urllib.parse
import tqdm

# Auth:
try:
    from envars import API_BASE, API_TOKEN

mindey

# Message to the future, that I wrote to the sattelite KEO ( https://en.wikipedia.org/wiki/KEO )
# I sent it before 2006-06-15). Perhaps it has a few mistakes, but anyway, here is what I wrote:

Hello, the intelligent entities of the future.

This is Inyuki from Japan, from the past.

How were you doing all the time from the year 2006 up to now? How are
you doing now? What problems you are having? It is interesting to
know.

mindey

# inspired by:
# https://www.halfbakery.com/idea/Mirrored_20Mouse

import pyautogui

width, height = pyautogui.size()

def mirror(duration=0.0):
    current = pyautogui.position()
    new_x = width - current.x

mindey

### For finding schema

def schematize(obj):
    '''
    Get schema of nested JSON, assuming first item in lists.
    '''
    if isinstance(obj, dict):
        return {k: schematize(v) for k, v in obj.items()}
    elif isinstance(obj, list):
        return [schematize(elem) for elem in obj][:1]

mindey

### For finding schema

def schematize(obj):
    '''
    Get schema of nested JSON, assuming first item in lists.
    '''
    if isinstance(obj, dict):
        return {k: schematize(v) for k, v in obj.items()}
    elif isinstance(obj, list):
        return [schematize(elem) for elem in obj][:1]

mindey

import os
import collections
import langdetect

LANGUAGE_CODES = os.listdir(langdetect.PROFILES_DIRECTORY)

def detect_language(text, max_length=2):
    """ Make sure we return N-letter keys for languages"""
    shorter = {'zh-cn': 'cn', 'zh-tw': 'zh'}
    code = langdetect.detect(text)

mindey

import pandas
import xarray

ds = xarray.Dataset(
  {'x': ([None], [1,2,3] ),
   'y': ([None], [4,5,6] )},
)

# *is equivalent to*

mindey

import pandas
from dask import dataframe
from dask.diagnostics import ProgressBar

def parallel_apply(df, func, progress=True, chunkrows=100, scheduler_address=None, *args, **kwargs):
    
    if scheduler_address:
        from dask.distributed import Client
        client = Client(scheduler_address)
    

mindey

# Practically, it's useful if we have complex observations (rows) and variables (columns):

df = pandas.DataFrame(
   data=pandas.np.array(
      [[1,2,3,4,5],
       [6,7,8,9,10],
       [11,12,13,14,15]]).T,
   index=pandas.MultiIndex.from_arrays(
       [['x','x','x','y','z'],
        ['a','a','b','b','c'],

mindey

import pandas
import multiprocessing

def apply_parallel(grouped_df, func):
    with multiprocessing.Pool(multiprocessing.cpu_count()) as p:
        ret_list = p.map(func, [group for name, group in grouped_df])
    return pandas.concat(ret_list)