Class: Webcache::DiskCache

Inherits:
Object
  • Object
show all
Defined in:
lib/webget/webcache.rb

Instance Method Summary collapse

Instance Method Details

#cached?(url) ⇒ Boolean Also known as: exist?

Returns:

  • (Boolean) 


70
71
72
73
 
# File 'lib/webget/webcache.rb', line 70

def cached?( url )
  body_path = "#{Webcache.root}/#{url_to_path( url )}"
  File.exist?( body_path )
end

#read(url) ⇒ Object 



77
78
79
80
 
# File 'lib/webget/webcache.rb', line 77

def read( url )
  body_path = "#{Webcache.root}/#{url_to_path( url )}"
  File.open( body_path, 'r:utf-8' ) {|f| f.read }
end

#read_json(url) ⇒ Object 



82
83
84
85
86
87
 
# File 'lib/webget/webcache.rb', line 82

def read_json( url )
  body_path = "#{Webcache.root}/#{url_to_path( url )}"
  txt = File.open( body_path, 'r:utf-8' ) {|f| f.read }
  data = JSON.parse( txt )
  data
end

#record(url, response, format: 'html') ⇒ Object

add more save / put / etc. aliases - why? why not? rename to record_html - why? why not?



93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
 
# File 'lib/webget/webcache.rb', line 93

def record( url, response, format: 'html' )

  body_path = "#{Webcache.root}/#{url_to_path( url )}"
  meta_path = "#{body_path}.meta.txt"

  ## make sure path exits
  FileUtils.mkdir_p( File.dirname( body_path ) )


  puts "[cache] saving #{body_path}..."

  ## todo/check: verify content-type - why? why not?
  if format == 'json'
    File.open( body_path, 'w:utf-8' ) {|f| f.write( JSON.pretty_generate( response.json )) }
  else
    ## note - for now always assume utf8!!!!!!!!!
    File.open( body_path, 'w:utf-8' ) {|f| f.write( response.text ) }
  end

  File.open( meta_path, 'w:utf-8' ) do |f|
    response.headers.each do |key, value|  # iterate all response headers
      f.write( "#{key}: #{value}" )
      f.write( "\n" )
    end
  end
end

#url_to_id(str) ⇒ Object

note: use file path as id for DiskCache (is different for DbCache/SqlCache?) use file:// instead of disk:// - why? why not?



124
 
# File 'lib/webget/webcache.rb', line 124

def url_to_id( str ) "disk://#{url_to_path( str )}"; end

#url_to_path(str) ⇒ Object

helpers



128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
 
# File 'lib/webget/webcache.rb', line 128

def url_to_path( str )
  ## map url to file path
  uri = URI.parse( str )

  ## note: ignore scheme (e.g. http/https)
  ##         and  post  (e.g. 80, 8080, etc.) for now
  ##    always downcase for now (internet domain is case insensitive)
  host_dir = uri.host.downcase

  ## "/this/is/everything?query=params"
  ##   cut-off leading slash and
  ##    convert query ? =
  req_path = uri.request_uri[1..-1]



  ### special "prettify" rule for weltfussball
  ##   /eng-league-one-2019-2020/  => /eng-league-one-2019-2020.html
  if host_dir.index( 'weltfussball.de' ) ||
     host_dir.index( 'worldfootball.net' )
        if req_path.end_with?( '/' )
           req_path = "#{req_path[0..-2]}.html"
        else
          puts "ERROR: expected request_uri for >#{host_dir}< ending with '/'; got: >#{req_path}<"
          exit 1
        end
  elsif host_dir.index( 'tipp3.at' )
    req_path = req_path.sub( '.jsp', '' )  # shorten - cut off .jsp extension

    ##   change ? to -I-
    ##   change = to ~
    ##   Example:
    ##   sportwetten/classicresults.jsp?oddsetProgramID=888
    ##     =>
    ##   sportwetten/classicresults-I-oddsetProgramID~888
    req_path = req_path.gsub( '?', '-I-' )
                       .gsub( '=', '~')

    req_path = "#{req_path}.html"
  elsif host_dir.index( 'football-data.org' )
    req_path = req_path.sub( 'v2/', '' )  # shorten - cut off v2/

    ## flattern - make a file path - for auto-save
    ##   change ? to -I-
    ##   change / to ~~
    ##   change = to ~
    req_path = req_path.gsub( '?', '-I-' )
                       .gsub( '/', '~~' )
                       .gsub( '=', '~')

    req_path = "#{req_path}.json"
  else
    ## no special rule
  end

  page_path = "#{host_dir}/#{req_path}"
  page_path
end