Convert URLS to names

[bagelbits]

Overview

A redo of #125
This is a second stab at #121. I still need to fully review the output and could use some help. Also there are few files I had to skip because the either didn't have a name (Levels, Spellcasting, StartingEquipment) or didn't make sense (Test). I've also done some manual fixing as well.

Here is the silly Ruby script I wrote to generate this change. Thanks to @ogregoire for the naming convention.

# frozen_string_literal: true

require 'json'

FILENAMES = %w[
  5e-SRD-Ability-Scores.json
  5e-SRD-Classes.json
  5e-SRD-Conditions.json
  5e-SRD-Damage-Types.json
  5e-SRD-Equipment-Categories.json
  5e-SRD-Equipment.json
  5e-SRD-Features.json
  5e-SRD-Languages.json
  5e-SRD-Levels.json
  5e-SRD-Magic-Schools.json
  5e-SRD-Monsters.json
  5e-SRD-Proficiencies.json
  5e-SRD-Races.json
  5e-SRD-Skills.json
  5e-SRD-Spellcasting.json
  5e-SRD-Spells.json
  5e-SRD-StartingEquipment.json
  5e-SRD-Subclasses.json
  5e-SRD-Subraces.json
  5e-SRD-Test.json
  5e-SRD-Traits.json
  5e-SRD-Weapon-Properties.json
].freeze

FILENAMES_TO_ITERATE = %w[
  5e-SRD-Ability-Scores.json
  5e-SRD-Classes.json
  5e-SRD-Conditions.json
  5e-SRD-Damage-Types.json
  5e-SRD-Equipment-Categories.json
  5e-SRD-Equipment.json
  5e-SRD-Features.json
  5e-SRD-Languages.json
  5e-SRD-Magic-Schools.json
  5e-SRD-Monsters.json
  5e-SRD-Proficiencies.json
  5e-SRD-Races.json
  5e-SRD-Skills.json
  5e-SRD-Spells.json
  5e-SRD-Subclasses.json
  5e-SRD-Subraces.json
  5e-SRD-Weapon-Properties.json
].freeze

def convert_to_id(name)
  id = name.downcase.strip
  id = id.gsub(%r{\/}, ' ')
  id = id.gsub(/[:,\(\)]+/, '')
  id = id.gsub(/(?:(?<=\bo)'|'(?=s\b)|^\W+|\W+$)/, '')
  id.gsub(/\s+/, '-')
end

def replace_in_all(original_url, new_url)
  FILENAMES.each do |file_name|
    text = File.read(file_name)
    new_contents = text.gsub(/#{original_url}(?=")/, new_url)

    # To write changes to the file, use:
    File.open(file_name, 'w') { |file| file.puts new_contents }
  end
end

# iterate files
FILENAMES_TO_ITERATE.each do |file_name|
  puts "Current file: #{file_name}"
  file_contents = IO.readlines(file_name).join
  json = JSON.parse(file_contents)
  json.length.times do |index|
    object_index = json[index]['index']
    # puts json[index]['name']
    id = convert_to_id(json[index]['name'])
    file_contents = file_contents.gsub(/"index": #{object_index},/, "\"id\": \"#{id}\",")

    original_url = json[index]['url']
    new_url = original_url.gsub(%r{(?<=\/)#{object_index}}, id)
    file_contents = file_contents.gsub(/#{original_url}(?=")/, new_url)

    # regenerate file
    File.open(file_name, 'w') do |line|
      line.puts file_contents
    end

    # change all files
    replace_in_all(original_url, new_url)

    file_contents = IO.readlines(file_name).join
    json = JSON.parse(file_contents)
  end
end

Here's a fun image for your troubles

[ogregoire]

This is an impressive and excellent work, again :-)

[bagelbits]

@ogregoire Since the changes we are making are now live with the API, I'm a little worried about changing the index key to id, as it will break everything. I'll probably switch the key back to index but keep the string ids for now.

[bagelbits]

I want to test this locally with a running API to make sure I don't just BREAK EVERYTHING.

[bagelbits]

This needs to be pushed in tandem with 5e-bits/5e-srd-api#26 so it doesn't break everything immediately.

[bagelbits]

@ogregoire I think everything is ready to go for this. Both DB and API side. But I'm terrified of pulling those levers.

[bagelbits]

https://lever.holdings