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) 


78
79
80
81
 
# File 'lib/webget/webcache.rb', line 78

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

#read(url) ⇒ Object 



85
86
87
88
 
# File 'lib/webget/webcache.rb', line 85

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

#read_csv(url) ⇒ Object 



97
98
99
100
101
102
 
# File 'lib/webget/webcache.rb', line 97

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

#read_json(url) ⇒ Object 



90
91
92
93
94
95
 
# File 'lib/webget/webcache.rb', line 90

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, path: nil, encoding: 'UTF-8', format: 'html') ⇒ Object

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

rename to record_html - why? why not?



107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
 
# File 'lib/webget/webcache.rb', line 107

def record( url, response,
            path: nil,
            encoding: 'UTF-8',
            format: 'html' )

  body_path = "#{Webcache.root}/#{url_to_path( url, path: path )}"
  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?

  ## note - for now respone.text always assume (converted) to utf8!!!!!!!!!

  if format == 'json'
    File.open( body_path, 'w:utf-8' ) {|f| f.write( JSON.pretty_generate( response.json )) }
  elsif format == 'csv'
    ## fix: newlines - always use "unix" style" - why? why not?

    ## fix:  use :newline => :universal option? translates to univeral "\n"

    text = response.text( encoding: encoding ).gsub( "\r\n", "\n" )
    File.open( body_path, 'w:utf-8' ) {|f| f.write( text ) }
  else   ## html or txt

    text = response.text( encoding: encoding )
    File.open( body_path, 'w:utf-8' ) {|f| f.write( text ) }
  end


  File.open( meta_path, 'w:utf-8' ) do |f|
    ## todo/check:

    ##  do headers also need to converted (like text) if encoding is NOT utf-8 ???

    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?



150
 
# File 'lib/webget/webcache.rb', line 150

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

#url_to_path(str, path: nil) ⇒ Object

helpers



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
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
 
# File 'lib/webget/webcache.rb', line 154

def url_to_path( str, path: nil )
  ## 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

  req_path = if path   ## use "custom" (file)path for cache storage if passed in

               path
             else
              ## "/this/is/everything?query=params"

              ##   cut-off leading slash and

              ##    convert query ? =

               uri.request_uri[1..-1]
             end



  ### 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( 'fbref.com' )
    req_path = req_path.sub( 'en/', '' )      # shorten - cut off en/

    req_path = "#{req_path}.html"             # auto-add html extension

  elsif host_dir.index( 'football-data.co.uk' )
    req_path = req_path.sub( 'mmz4281/', '' )  # shorten - cut off mmz4281/

    req_path = req_path.sub( 'new/', '' )      # shorten - cut off new/

  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