# This file is a demo of using aiocsv and aiofiles libraries to speed up reading and parsing CSV files.
#
# Start reading this code from the entrypoint function main() below.
#
import asyncio
import aiofiles
from csv import QUOTE_NONNUMERIC
from typing import AsyncGenerator
from aiocsv import AsyncDictWriter, AsyncDictReader


async def read_lines(file: str) -> AsyncGenerator[dict, None]:
    """
    Read lines from CSV file.
    """
    async with aiofiles.open(file, "r") as afp:
        async for row in AsyncDictReader(afp, delimiter=","):
            yield row


async def parse_lines(generator: AsyncGenerator[dict, None]) -> AsyncGenerator[dict, None]:
    """
    Parse lines from generator.
    """
    async for line in generator:
        # do some parsing here, like that:
        line = line
        yield line


async def save_lines(file: str, generator: AsyncGenerator[dict, None]):
    """
    Save lines from generator to CSV file.
    """
    async with aiofiles.open(
            file,
            mode="w",
            encoding="utf-8",
            newline="",
    ) as afp:
        rows = []
        writer = None
        async for item in generator:
            if writer is None:
                header = list(item.keys())
                writer = AsyncDictWriter(
                    afp,
                    header,
                    quoting=QUOTE_NONNUMERIC,
                )
                await writer.writeheader()
            # gather rows into a list
            # keep the list size reasonable according to your memory constraints
            rows.append(item)
            if len(rows) % 10000 == 0:
                await writer.writerows(rows)
                rows = []
            await afp.flush()
        # write the rest of the rows if any
        if len(rows) > 0:
            await writer.writerows(rows)


async def main(in_file, out_file):
    """
    Main function that reads lines from in_file, parses them and saves to out_file.
    """
    raw_line_generator = read_lines(in_file)
    parsed_line_generator = parse_lines(generator=raw_line_generator)
    await save_lines(file=out_file, generator=parsed_line_generator)


in_file = "some_input_file.csv"
out_file = "some_output_file.csv"
asyncio.run(main(in_file, out_file))

Foreword about addiction

I’m an addict and I can’t help myself. At least not in an easy way. I’m addicted to reading idiotic, moronic, hateful, homophobic comments posted by postimees.ee readers to Postimees website. FYI – Postimees is one of the oldest and biggest newspapers in Estonia.

So that’s why I was looking for a solution on how to remove comments or at least these deceptive links to comments by every article on postimees.ee website.

It only works in Firefox because that’s my main browser.

Let’s get hands dirty

First, open about:config (type it into address field). Firefox config opens after a warning.
Find the key

toolkit.legacyUserProfileCustomizations.stylesheets

Make sure the value is “true” (double click on the value). Close config.
Now open config of profiles.

Type about:profiles to address field.
Find your profile (not the development one) and there should be “Root Directory”. At the end of this line is button “Open in Finder” (or open I-dont-know-where in Windows). Whatever, just click it.

Your Firefox profile folder opens.

Inside that folder create a new folder named “chrome” (mind the lowercase name, case matters!).

Inside that “chrome” folder create an empty text file called “userContent.css”. Again – mind the naming.
Into that file add following lines:

@-moz-document domain(postimees.ee) {
    span.list-article__comment {
        display: none;
    }
}

Save and close the file. Restart Firefox. Go to postimees.ee. Welcome to your new life!