Class: XMLTV::TvgidsGrabber
Constant Summary
collapse
- Cattrans =
{
'amusement' => 'Talk',
'animatie' => 'Animated',
'comedy' => 'Comedy',
'documentaire' => 'Documentary',
'educatief' => 'Educational',
'erotiek' => 'Adult',
'film' => 'Movies',
'muziek' => 'Art/Music',
'informatief' => 'Educational',
'jeugd' => 'Children',
'kunst/cultuur' => 'Arts/Culture',
'misdaad' => 'Crime/Mystery',
'muziek' => 'Music',
'natuur' => 'Science/Nature',
'nieuws/actualiteiten' => 'News',
'overige' => 'Unknown',
'religieus' => 'Religion',
'serie/soap' => 'Drama',
'sport' => 'Sports',
'theater' => 'Arts/Culture',
'wetenschap' => 'Science/Nature'
}- Roletrans =
{
'regie' => 'director',
'acteurs' => 'actor',
'presentatie' => 'presenter',
'scenario' => 'writer'
}- Titeltrans =
{
'titel aflevering' => 'sub-title',
'jaar van premiere' => 'date',
'aflevering' => 'episode-num'
}
Constants inherited
from Grabber
Grabber::Dag, Grabber::MythTV, Grabber::Vandaag
Instance Attribute Summary
Attributes inherited from Grabber
#all_channels, #base_url, #channel_list, #chnbasedir, #config, #config_file_name, #generator, #lang, #myname, #reject_file_name, #spooldir
Instance Method Summary
collapse
Methods inherited from Grabber
#add_channels_to_config, #cachefile, #channel_display, #channel_name, #check_argv, #check_channel, #clean_cache_dir, #clean_spool_dir, #config_channels, #date_stats, #delete_channels_from_config, #do_list, #do_options, #dump, #fetch, #fix_times, #get_channels, #initialize, #list_all, #list_config, #load_cachefile, #load_channel_file, #load_config_file, #outputfile, #printline, #proghash, #reject, #report, #run, #save, #save_config, #save_object, #version
Constructor Details
This class inherits a constructor from XMLTV::Grabber
Instance Method Details
#channel_url(chan_id) ⇒ Object
74
75
76
|
# File 'lib/xmltv/sites/tvgids.rb', line 74
def channel_url(chan_id)
"#{base_url}//zoeken/?periode=9&station=#{chan_id}"
end
|
#clean_cache(cache) ⇒ Object
90
91
92
93
94
95
96
97
98
99
|
# File 'lib/xmltv/sites/tvgids.rb', line 90
def clean_cache(cache)
count = 0
cache.delete_if do |dt, en|
rsl = (Date.dutch(en['datum']) < Vandaag)
count += 1 if rsl
rsl
end
count
end
|
#fetch_all_channels ⇒ Object
78
79
80
81
82
83
84
85
86
87
88
|
# File 'lib/xmltv/sites/tvgids.rb', line 78
def fetch_all_channels
page = fetch(channel_url(1))
channels = Hash.new
page.search('//optgroup')[0..1].each do |og|
og.search('/option').each do |g|
channels[g['value']] = g.inner_text
end
end
save_object(channels, channel_list)
channels
end
|
#grab_channel(chan_id) ⇒ Object
101
102
103
104
105
106
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
145
146
147
148
149
|
# File 'lib/xmltv/sites/tvgids.rb', line 101
def grab_channel(chan_id)
url = channel_url(chan_id)
page = fetch(url)
@channelhash = load_cachefile(chan_id)
period = datum = nil
fetched = 0
begin
found = remaining = page.at("//table.overzicht//tr//td/strong").inner_text.to_i
rescue NoMethodError
niks = page.at("//div#resultaten").at("//td").inner_text
STDERR.puts url, niks
return
end
page.search("//table.overzicht//tr").each do |pg|
td = pg.at('td')
next if td.nil? || pg['class'] == 'zoekstring'
if td['class'] == 'bloktitel'
period = td.at('h5').inner_text rescue period
datum = td.at('h4').inner_text rescue datum
next
end
if (tijd = pg.at('/th').inner_text) =~ /\d\d:\d\d/
det = pg.at('/td//a')
href = det['href']
id = href[/ID=(\d+)/,1]
remaining -= 1
next if @channelhash[id]
fetched += 1
begin
@channelhash[id] = program = grab_detail(href)
rescue
STDERR.puts href, pg, '====='
raise
end
program['title'] = det.inner_text.strip.to_utf
program['period'] = period
program['datum'] = datum
program['tijd'] = tijd
program['progtip'] = '4/5' if pg['class'] == 'progTip'
end
end
STDERR.puts "Something wrong remaining: #{remaining}" if remaining != 0
save_object(@channelhash, cachefile(chan_id)) if fetched > 0
found
end
|
#grab_detail(href) ⇒ Object
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
|
# File 'lib/xmltv/sites/tvgids.rb', line 47
def grab_detail(href)
if href[0] == ?/
href="#{base_url}#{href}"
end
STDERR.puts "#{Time.now}: #{href} #{@channelhash.size}" if XmltvOptions.verbose
program = Hash.new
details = fetch(href)
desc = []
details.at('//table#progDetail').search('//tr//p').each do |p|
break if p['class'] == 'meerLinks'
line = p.inner_text.strip
desc << line unless line.empty?
end
program['desc'] = desc.join(' ').to_utf
details.search('//div#progPropt//tr/th').each do |pg|
content = pg.at('../td')
if content['class'] == 'personen'
rsl = content.at('div').search('.').find_all { |x| x.text? }.map{|x| x.to_s.strip.to_utf}.find_all{|x| ! x.empty?}
else
rsl = content.inner_text.strip.to_utf
end
program[pg.inner_text.strip.gsub(':','').downcase] = rsl
end
program
end
|
#parse_times(str) ⇒ Object
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
|
# File 'lib/xmltv/sites/tvgids.rb', line 151
def parse_times(str)
rsl = nil
md = /(\d+)\s(\w+)\s(\d+),\s(\d+):(\d+)/.match(str)
if md
rsl = md.captures.map do |x|
x =~ /\d/ ? x.to_i : Date::Maanden.index(x.downcase)
end
mdstop = /(\d+):(\d+)/.match(str[md.offset(0)[1]..-1])
if mdstop
rsl << mdstop.captures.map {|x| x.to_i}
end
rsl.flatten!
end
if rsl
raise DateError.new(str) if rsl.index(nil)
end
rsl
end
|
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
222
223
224
225
226
227
228
229
230
231
232
|
# File 'lib/xmltv/sites/tvgids.rb', line 170
def transform(chan_id)
progdata_array = Array.new
@channelhash.each_pair do |id, entry|
begin
progdata = proghash(entry, chan_id)
a=entry['datum en tijdstip']
shift = entry['period'] == 'Nacht' ? Dag : 0
dag, maand, jaar, startuur, startmin, stopuur, stopmin = parse_times(a)
next if dag.nil?
begin
progdata['start'] = start = Time.local(jaar, maand, dag, startuur, startmin) + shift
if stopuur
stop = Time.local(jaar, maand, dag, stopuur, stopmin) + shift
if start > stop && start.hour >= 21 && stop.hour <= 7
stop += Dag
end
progdata['stop'] = stop
end
rescue ArgumentError
STDERR.puts 'ArgumentError'
STDERR.puts jaar, maand, dag, stopuur, stopmin
next
end
date_stats(chan_id, progdata['start'])
if (b = entry['bijzonderheden'])
b.downcase.split(',').each do |bijz|
case bijz
when /breedbeeld/
progdata['video']['aspect'] = '16:9'
when /zwart/
progdata['video']['colour'] = 'no'
when /teletekst/
progdata['subtitles']['type'] = 'teletext'
when /stereo/
progdata['audio']['stereo'] = 'stereo'
end
end
end
%w{ regie acteurs scenario presentatie }.each do |role|
if entry[role]
progdata['credits'][Roletrans[role]] = entry[role]
end
end
progdata['category'] = Cattrans[entry['genre'].downcase] if entry['genre']
progdata['star-rating']['value'] = entry['progtip'] if entry['progtip']
Titeltrans.each do |key|
progdata[Titeltrans[key]] = entry[key] if entry[key]
end
progdata_array << progdata
rescue DateError => exc
STDERR.puts exc.class, exc.message
PP.pp(entry, STDERR)
next
rescue StandardError => exc
STDERR.puts exc, exc.message, exc.backtrace
PP.pp(entry, STDERR)
raise
end
end
progdata_array
end
|