Learning how to parse

Gemfile

@@ -43,6 +43,8 @@ gem 'pg_search'
 # Pagination library
 gem 'will_paginate', '~> 3.3'
 
+gem 'httparty'
+
 group :doc do
   gem 'apipie-rails'
   gem 'sdoc'
@@ -53,6 +55,7 @@ group :development, :test do
   gem 'dotenv-rails'
   gem 'factory_bot_rails'
   gem 'faker'
+  gem 'pry'
   gem 'rspec_junit_formatter'
   gem 'rspec-rails'
 end

Gemfile.lock

@@ -78,6 +78,7 @@ GEM
       msgpack (~> 1.0)
     builder (3.2.4)
     byebug (11.1.3)
+    coderay (1.1.3)
     concurrent-ruby (1.1.9)
     crass (1.0.6)
     database_cleaner (2.0.1)
@@ -113,6 +114,9 @@ GEM
       gemoji (>= 2.1.0)
     globalid (0.5.2)
       activesupport (>= 5.0)
+    httparty (0.20.0)
+      mime-types (~> 3.0)
+      multi_xml (>= 0.5.2)
     i18n (1.8.10)
       concurrent-ruby (~> 1.0)
     jaro_winkler (1.5.4)
@@ -130,9 +134,13 @@ GEM
       mini_mime (>= 0.1.1)
     marcel (1.0.2)
     method_source (1.0.0)
+    mime-types (3.4.1)
+      mime-types-data (~> 3.2015)
+    mime-types-data (3.2022.0105)
     mini_mime (1.1.2)
     minitest (5.14.4)
     msgpack (1.4.2)
+    multi_xml (0.6.0)
     nio4r (2.5.8)
     nokogiri (1.12.5-arm64-darwin)
       racc (~> 1.4)
@@ -146,6 +154,9 @@ GEM
     pg_search (2.3.5)
       activerecord (>= 5.2)
       activesupport (>= 5.2)
+    pry (0.14.1)
+      coderay (~> 1.1)
+      method_source (~> 1.0)
     puma (5.5.2)
       nio4r (~> 2.0)
     racc (1.6.0)
@@ -296,10 +307,12 @@ DEPENDENCIES
   faker
   figaro
   gemoji-parser
+  httparty
   listen
   oj
   pg
   pg_search
+  pry
   puma
   rack-cors
   rails

app/helpers/granblue_wiki.rb

@@ -0,0 +1,125 @@
+# frozen_string_literal: true
+
+require 'rubygems'
+require 'httparty'
+require 'strscan'
+require 'pry'
+
+# GranblueWiki fetches and parses data from gbf.wiki
+class GranblueWiki
+  URL = 'https://gbf.wiki/api.php'
+
+  def fetch(page)
+    query_params = params(page).map do |key, value|
+      "#{key}=#{value}"
+    end.join('&')
+
+    destination = "#{URL}?#{query_params}"
+    response = HTTParty.get(destination)
+
+    response['parse']['wikitext']['*']
+  end
+
+  def parse(page)
+    parsed = parse_string(fetch(page))
+
+    abilities = extract_abilities(parsed)
+    ougis = extract_ougis(parsed)
+
+    ap abilities.merge(ougis)
+  end
+
+  def extract_abilities(hash)
+    abilities = []
+    hash.each do |key, value|
+      next unless key =~ /^a(\d+)_/
+
+      ability_number = Regexp.last_match(1).to_i
+      abilities[ability_number] ||= {}
+
+      case key.gsub(/^a\d+_/, '')
+      when 'cd'
+        cooldown = parse_substring(value)
+        abilities[ability_number]['cooldown'] = cooldown
+      when 'dur'
+        duration = parse_substring(value)
+        abilities[ability_number]['duration'] = duration
+      when 'oblevel'
+        obtained = parse_substring(value)
+        abilities[ability_number]['obtained'] = obtained
+      else
+        abilities[ability_number][key.gsub(/^a\d+_/, '')] = value
+      end
+    end
+
+    { 'abilities' => abilities.compact }
+  end
+
+  def parse_substring(string)
+    hash = {}
+
+    string.scan(/\|([^|=]+?)=([^|]+)/) do |key, value|
+      value.gsub!(/\}\}$/, '') if value.include?('}}')
+      hash[key] = value
+    end
+
+    hash
+  end
+
+  def extract_ougis(hash)
+    ougi = []
+    hash.each do |key, value|
+      next unless key =~ /^ougi(\d*)_(.*)/
+
+      ougi_number = Regexp.last_match(1)
+      ougi_key = Regexp.last_match(2)
+      ougi[ougi_number.to_i] ||= {}
+      ougi[ougi_number.to_i][ougi_key] = value
+    end
+
+    { 'ougis' => ougi.compact }
+  end
+
+  def parse_string(string)
+    # Split the string into lines
+    lines = string.split("\n")
+
+    # Initialize an empty hash to store the key/value pairs
+    data = {}
+
+    # Iterate over the lines
+    good_lines = lines.map do |line|
+      line if line[0] == '|' && line.size > 2
+    end
+
+    good_lines.compact.each do |line|
+      trimmed_line = line[1..]
+
+      # Split the line into key and value by the '=' character
+      key, value = trimmed_line.split('=', 2)
+
+      next unless value
+
+      # Strip leading and trailing whitespace from the key and value
+      key = key.strip
+      value = value.strip
+
+      # Store the key/value pair in the data hash
+      data[key] = value
+    end
+
+    # Return the data hash
+    data
+  end
+
+  private
+
+  def params(page)
+    {
+      action: 'parse',
+      format: 'json',
+      page: page,
+      prop: 'wikitext'
+    }
+  end
+end