Skip to content

Tool used to scrap websites, convert text to english, and clean works for ENS purposes

Notifications You must be signed in to change notification settings

0xlucyy/translator

Repository files navigation

translator

I should have unit tests for this but I dont. I do test everything manually from my terminal using import pdb; pdb.set_trace().

Definitions of text file states

  • Clean: Lower cased, stripped of white spaces on both sides.
  • Special-Word: Any Spanish word(s) that contain any character(s) in this list ['á', 'é', 'í', 'ó', 'ú', 'ü', 'ñ'].
  • Normalize: Clean + white space replaced with no space + Special-Word words get a non-special character copy.

Instructions

  • Clone
  • Run python3 -m venv venv.
  • Run . venv/bin/activate.
  • Run npm i
  • Run python install -e ..

Commands

  • Run clean spanish-animals.
    • Creates a clean text file at /tests/fixtures/*-cleaned.txt.
    • Creates a normalized text file at /tests/fixtures/*-normalizeded.txt.
    • Alt text
  • Run run-csv.
    • Translates existing English ens.vision csv files into Spanish.
    • CSV files must be in /collections/*.csv.
    • Translated file will be in /tests/fixtures/.*txt.
    • Language values hardcoded but can be easily changed to be automated.
  • Run run-txt english-animals-file1 spanish-animals-file2.
    • Translates file1 and writes output into file2.
    • file1 must exist at /tests/fixtures/*.txt..
    • file2 will exist at /tests/fixtures/*.txt.
  • Run animals cleaned-animals.
    • Scraper for animals list.
  • Run node scripts/normalize.js
    • Update the location of a normalized text file on line 8.
    • Prints to console ens-collection ready csv data.
    • Copy text on console into your local copy of ens-collections, in ens-collections/collections/*.csv.

working on starwars characters

About

Tool used to scrap websites, convert text to english, and clean works for ENS purposes

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published