coelacanth 0.1.4 → 0.1.5

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 25c5b81d90d4db4cfe123670105ad4cf91f3ff8a54ffa8c242385a39d1ef9827
-  data.tar.gz: 567ffbebc172d3747f85dd158610bc4e3baabadae8387306912f7bf4dd1dcc7f
+  metadata.gz: 29b00f207c71fb604269ddb59b70064509dce9a5abc189c5ba98a17fc3701815
+  data.tar.gz: fe1487e4f8b15ebbdc6312dce95d2e50b3338871fedb1cb5c49ec8098d97951c
 SHA512:
-  metadata.gz: 3c05022d1bcca228c5c977776f89057087010dfd5da8617a529984d7cdaedcec8418a5167ef59afef2f0e628fbda425d4fffec528ade81413cf36a7fb6e1cbb4
-  data.tar.gz: e0560ede407f54ea6d840c9096bdfd67557832b4fd797c2fa60746b35f2295a55b5f016b94a7957704eb7457eb88f2710a3b76b2b01b5c2277d604140c6d7fba
+  metadata.gz: 462bbd1ee309ca534b42bdac616c4145f7fc3320971d391c025d7bd9080850b61fb773f8a84d5286e73321c90732f897c9b1760c68c6542d5bd3abdfba5b23d9
+  data.tar.gz: ec1071051319687f27e989871cc4cc78a90fbbf2e07c52310e7150fa651efd908a3a2cde8424ed7abfe03755afca3661a6b64477e37be8f7b81d8427caf02f07

data/Gemfile

@@ -9,3 +9,4 @@ gem "ferrum", "~> 0.15"
 gem "rake", "~> 13.0"
 gem "rspec", "~> 3.0"
 gem "rubocop", "~> 1.21"
+gem "oga", "~> 3.4"

data/Gemfile.lock

@@ -1,13 +1,14 @@
 PATH
   remote: .
   specs:
-    coelacanth (0.1.4)
+    coelacanth (0.1.5)
 
 GEM
   remote: https://rubygems.org/
   specs:
     addressable (2.8.7)
       public_suffix (>= 2.0.2, < 7.0)
+    ansi (1.5.0)
     ast (2.4.2)
     concurrent-ruby (1.3.4)
     diff-lcs (1.4.4)
@@ -18,6 +19,9 @@ GEM
       websocket-driver (~> 0.7)
     json (2.7.1)
     language_server-protocol (3.17.0.3)
+    oga (3.4)
+      ast
+      ruby-ll (~> 2.1)
     parallel (1.24.0)
     parser (3.3.0.5)
       ast (~> 2.4.1)
@@ -54,6 +58,9 @@ GEM
       unicode-display_width (>= 2.4.0, < 3.0)
     rubocop-ast (1.31.2)
       parser (>= 3.3.0.4)
+    ruby-ll (2.1.3)
+      ansi
+      ast
     ruby-progressbar (1.13.0)
     unicode-display_width (2.5.0)
     webrick (1.8.1)
@@ -68,6 +75,7 @@ PLATFORMS
 DEPENDENCIES
   coelacanth!
   ferrum (~> 0.15)
+  oga (~> 3.4)
   rake (~> 13.0)
   rspec (~> 3.0)
   rubocop (~> 1.21)

data/lib/coelacanth/client.rb

@@ -1,46 +1,76 @@
 # frozen_string_literal: true
 
 require "ferrum"
+require "oga"
 
 module Coelacanth
   # Coelacanth::Client
   class Client
-    def initialize
-      @config = Configure.new
+    def initialize(url = nil)
+      @config = Coelacanth.config
+      @url = url if url && valid_url?(url)
     end
 
-    def valid_url?(url)
-      uri = URI.parse(url)
+    def valid_url?(url = nil)
+      @url = url if url
+      uri = URI.parse(@url)
       uri.is_a?(URI::HTTP) || uri.is_a?(URI::HTTPS)
     rescue URI::InvalidURIError
       false
     end
 
-    def resolve_redirect(url, limit = 10)
+    def resolve_redirect(url = nil, limit = 10)
+      @url = url if url && valid_url?(url)
       raise Coelacanth::DeepRedirectError, "Too many redirect" if limit.zero?
-      raise Coelacanth::RedirectError, "Url or location is nil" if url.nil?
+      raise Coelacanth::RedirectError, "Url or location is nil" if @url.nil?
 
-      response = get_response(url)
+      get_response(@url)
+      handle_response(@origin_response, limit)
+    end
 
-      case response
-      when Net::HTTPSuccess then url
-      when Net::HTTPRedirection then resolve_redirect(response["location"], limit - 1)
-      else
-        raise Coelacanth::RedirectError
-      end
+    def oga(url = nil)
+      @url = url if url && valid_url?(url)
+      Oga.parse_xml(get_response(@url))
     end
 
-    def get_response(url)
+    def get_response(url = nil)
+      @url = url if url && valid_url?(url)
       if @config.read("use_remote_client")
-        remote_client.goto(url)
-        remote_client.body
+        response_by_remote_client
       else
-        Net::HTTP.get_response(url)
+        response_by_net_http
       end
     end
 
     private
 
+    def handle_response(response, limit)
+      codes = Net::HTTPResponse::CODE_CLASS_TO_OBJ.invert
+      case @status_code.to_s
+      when /^#{codes[Net::HTTPSuccess]}\d\d$/
+        @url
+      when /^#{codes[Net::HTTPRedirection]}\d\d$/
+        @url = response["location"]
+        resolve_redirect(response["location"], limit - 1)
+      else
+        raise Coelacanth::RedirectError
+      end
+    end
+
+    def response_by_remote_client
+      remote_client.goto(@url)
+      @status_code = remote_client.network.status
+      @origin_response = remote_client
+      remote_client.body
+    end
+
+    def response_by_net_http
+      response = Net::HTTP.get_response(URI.parse(@url))
+      @status_code = response.code
+      @origin_response = response
+      response.body
+    end
+
     def remote_client
       if @remote_client.nil?
         headers = @config.read("remote_client.headers")

data/lib/coelacanth/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Coelacanth
-  VERSION = "0.1.4"
+  VERSION = "0.1.5"
 end

data/lib/coelacanth.rb

@@ -12,9 +12,15 @@ module Coelacanth
   class DeepRedirectError < StandardError; end
 
   def self.analyze(url)
-    Client.get_response(url)
+    @client = Client.new(url)
+    @client.resolve_redirect
     {
-      todo: "implement me"
+      remote_client: @config.read("use_remote_client"),
+      oga: @client.oga
     }
   end
+
+  def self.config
+    @config ||= Configure.new
+  end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: coelacanth
 version: !ruby/object:Gem::Version
-  version: 0.1.4
+  version: 0.1.5
 platform: ruby
 authors:
 - Yusuke
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2024-08-28 00:00:00.000000000 Z
+date: 2024-08-29 00:00:00.000000000 Z
 dependencies: []
 description: |
   coelacanth is a gem that allows you to easily parse and analyze web pages,