Find your secondhand Macbook on Bazos.cz easily

Requirements

• running locally:
  • python (developed with 3.9.5)
  • python libraries (requirements.txt) + ipykernel in case of using jupyter notebook
• running in docker:
  • docker

Run

• locally: python bazos.py [optional arguments - see below] or run inside bazos_macbooks.ipynb jupyter notebook

• in docker:

  1. docker build -t bazos .
  2. docker run -v "absolute/path/to/your/directory:/app/data" bazos

[Optional] Inputs

If you won't provide any arguments, script takes default ones:

DEFAULT_ARGS = dict(
    zip_code = 10400,
    dist = 50,
    min_p = 15000,
    max_p = 25000,
    n_ads = 100,
    send_mail = False,
    mb_years = ['2019', '2020', '2021']
)

It is possible to pass arguments from command line:

usage: bazos.py [-h] [--zip_code [ZIP_CODE]] [--dist [DIST]] [--min_p [MIN_P]] [--max_p [MAX_P]] [--n_ads [N_ADS]] [--mb_years [MB_YEARS ...]] [--send_mail]

optional arguments:
    -h, --help                  show this help message and exit
    --zip_code [ZIP_CODE]       Zip code - e.g. 10500
    --dist [DIST]               Distance in km from zip code - e.g. 50
    --min_p [MIN_P]             Minimum price in CZK - e.g. 10000
    --max_p [MAX_P]             Maximum price in CZK - e.g. 50000
    --n_ads [N_ADS]             Number of ads to go through - e.g. 100
    --mb_years [MB_YEARS ...]   Wanted years of Macbooks - e.g. 2019 2020
    --send_email                Whether to send an email with results

or using the same parameters in uppercase with BAZOS_ prefix as environment variables (for more convenient docker usage):

docker run -v "absolute/path/to/your/directory:/app/data" -e BAZOS_ZIP_CODE=10100 -e BAZOS_DIST=100 -e BAZOS_MIN_P=20000 -e BAZOS_MB_YEARS=20000 -e BAZOS_SEND_EMAIL=True bazos

or using --env-file docker parameter:

docker run -v "absolute/path/to/your/directory:/app/data" --env-file docker_env_vars bazos

Script then takes either:

1. Command line arguments
2. OS environments arguments
3. Default arguments

Output

• csv file with all ads (data/all.csv)
• csv file with macbooks air (data/airs.csv)
• csv file with macbooks pro (data/pros.csv)
• main method also returns list of pandas dataframes with all, airs and pros
• possibility to send result tables as html to email via google SMTP server (yagmail)

Example email output:

Improve

• better parsing (years & cpu mainly)
• run automatically for all existing pages (not specifying page ranges manually)
• more dynamic filtering of final airs and pros tables
• return only new entries (difference from previous run)