Refactor PHP scraper with new initial_paths option

lib/docs/filters/php/entries.rb

@@ -106,7 +106,7 @@ module Docs
       end
 
       def include_default_entry?
-        Php::INDEX_PATHS.exclude?(subpath) && doc.at_css('.reference', '.refentry', '.sect1')
+        !initial_page? && doc.at_css('.reference', '.refentry', '.sect1')
       end
     end
   end

lib/docs/scrapers/php.rb

@@ -5,18 +5,7 @@ module Docs
     self.version = 'up to 5.5.6'
     self.base_url = 'http://www.php.net/manual/en/'
     self.root_path = 'index.html'
-
+    self.initial_paths = %w(
-    # Downloaded from php.net/download-docs.php
-    self.dir = '/Users/Thibaut/DevDocs/Docs/PHP'
-
-    html_filters.push 'php/internal_urls', 'php/entries', 'php/clean_html', 'title'
-    text_filters.push 'php/fix_urls'
-
-    options[:title] = false
-    options[:root_title] = 'PHP: Hypertext Preprocessor'
-
-    INDEX_PATHS = %w(
-      index.html
       funcref.html
       refs.database.html
       set.mysqlinfo.html
@@ -25,11 +14,15 @@ module Docs
       reserved.interfaces.html
       reserved.variables.html)
 
-    options[:skip_links] = ->(filter) do
+    # Downloaded from php.net/download-docs.php
-      INDEX_PATHS.exclude?(filter.subpath)
+    self.dir = '/Users/Thibaut/DevDocs/Docs/PHP'
-    end
+
+    html_filters.push 'php/internal_urls', 'php/entries', 'php/clean_html', 'title'
+    text_filters.push 'php/fix_urls'
 
-    options[:only] = INDEX_PATHS.dup
+    options[:title] = false
+    options[:root_title] = 'PHP: Hypertext Preprocessor'
+    options[:skip_links] = ->(filter) { !filter.initial_page? }
 
     options[:only_patterns] = [
       /\Aclass\./,
@@ -49,7 +42,7 @@ module Docs
       sqlite3 sqlsrv ssh2 stats stream strings taint tidy uodbc url var varnish
       xml xmlreader xmlrpc xmlwriter xsl yaf yaml zip zlib)
 
-    options[:only].concat BOOKS.map { |s| "book.#{s}.html" }
+    options[:only] = BOOKS.map { |s| "book.#{s}.html" }
 
     options[:skip] = %w(
       control-structures.intro.html