Anexen

class Desc:
    __slots__ = ("value",)

    def __init__(self, value):
        self.value = value

    def __lt__(self, other):
        return self.value > other.value

data = [

Anexen

import numpy as np
from cityhash import CityHash64WithSeed


H = np.vectorize(CityHash64WithSeed, otypes=['uint64'])


class BloomFilter:
    def __init__(self, n: int, fp_prob: float) -> None:
        self.m = np.ceil(-(n * np.log(fp_prob)) / (np.log(2) ** 2)).astype(int)

Anexen

-- usage:
-- CALL update_table_in_batches('table_1', array['a = 1'], batch_size := 10);
-- CALL update_table_in_batches('table_2', array['b = 2'], track_by := 'created', initial_value := '2021-01-01'::timestamp);

CREATE OR REPLACE PROCEDURE update_table_in_batches(
    target_table text,
    set_expressions text[],
    track_by text default 'id',
    initial_value anyelement default 0,
    batch_size integer default 1000,

Anexen

# syntax = docker/dockerfile:experimental

ARG PYTHON_VERSION=3.7.6
ARG APP_ENV=prod

# ---------------------------------------------------------------

FROM python:${PYTHON_VERSION}-stretch AS base-builder
ENV VIRTUAL_ENV=/opt/venv
RUN python -m venv ${VIRTUAL_ENV}

Anexen

import sys
import time
import resource

from clickhouse_driver import Client


def memory_usage():
    return resource.getrusage(resource.RUSAGE_SELF).ru_maxrss