Module: Tweetlr::Processors::PhotoService

Includes:
LogAware
Defined in:
lib/tweetlr/processors/photo_service.rb

Overview

utilities for dealing with photo services

Constant Summary collapse

LOCATION_START_INDICATOR = 
'Location: '
LOCATION_STOP_INDICATOR = 
"\r\n"
PIC_REGEXP = 
/(.*?)\.(jpg|jpeg|png|gif)/i

Class Method Summary collapse

Methods included from LogAware

log=

Class Method Details

.extract_id(link) ⇒ Object

extract the pic id from a given link



104
105
106
 
# File 'lib/tweetlr/processors/photo_service.rb', line 104

def self.extract_id(link)
  link.split('/').last if link.split('/')
end

.find_image_url(link, embedly_key = nil) ⇒ Object 



20
21
22
23
24
25
26
27
28
 
# File 'lib/tweetlr/processors/photo_service.rb', line 20

def self.find_image_url(link, embedly_key=nil)
  url = nil
  if link && !(photo? link)
    url = process_link link, embedly_key
  elsif photo? link
    url = link
  end
  url
end

.image_url_embedly(link_url, key) ⇒ Object

find the image’s url via embed.ly



52
53
54
55
56
57
58
59
60
 
# File 'lib/tweetlr/processors/photo_service.rb', line 52

def self.image_url_embedly(link_url, key)
  link_url = follow_redirect(link_url)
  log.debug "embedly call: http://api.embed.ly/1/oembed?key=#{key}&url=#{link_url}"
  response = Tweetlr::Processors::Http::http_get_json "http://api.embed.ly/1/oembed?key=#{key}&url=#{link_url}"
  if response && (response['type'] == 'photo' || response['type'] == 'image')
    image_url = response['url'] 
  end
  image_url
end

.image_url_eyeem(link_url) ⇒ Object

extract the image of an eyeem.com pic



37
38
39
 
# File 'lib/tweetlr/processors/photo_service.rb', line 37

def self.image_url_eyeem(link_url)
  retrieve_image_url_by_css link_url, '.viewport-pic img'
end

.image_url_foursqaure(link_url) ⇒ Object

extract the image of a foursquare.com pic



41
42
43
44
45
 
# File 'lib/tweetlr/processors/photo_service.rb', line 41

def self.image_url_foursqaure(link_url)
  link_url = follow_redirect(link_url)
  image_url = retrieve_image_url_by_css link_url, 'meta[property="og:image"]', 'content'
  image_url unless image_url.include? "foursquare.com/img/categories"
end

.image_url_imgly(link_url, embedly_key) ⇒ Object

find the image’s url for a img.ly link



81
82
83
 
# File 'lib/tweetlr/processors/photo_service.rb', line 81

def self.image_url_imgly(link_url, embedly_key)
  retrieve_image_url_by_css link_url, '#the-image'
end

.image_url_instagram(link_url) ⇒ Object

find the image’s url for an instagram link



67
68
69
70
71
 
# File 'lib/tweetlr/processors/photo_service.rb', line 67

def self.image_url_instagram(link_url)
  link_url['instagram.com'] = 'instagr.am' if link_url.index 'instagram.com' #instagram's oembed does not work for .com links
  response = Tweetlr::Processors::Http::http_get_json "http://api.instagram.com/oembed?url=#{link_url}"
  response['url'] if response
end

.image_url_path(link_url) ⇒ Object

extract the image of a path.com pic



47
48
49
 
# File 'lib/tweetlr/processors/photo_service.rb', line 47

def self.image_url_path(link_url)
  retrieve_image_url_by_css link_url, 'img.photo-image'
end

.image_url_redirect(link_url, service_endpoint, stop_indicator = LOCATION_STOP_INDICATOR) ⇒ Object

extract image url from services like twitpic & img.ly that do not offer oembed interfaces



86
87
88
 
# File 'lib/tweetlr/processors/photo_service.rb', line 86

def self.image_url_redirect(link_url, service_endpoint, stop_indicator = LOCATION_STOP_INDICATOR)
  link_url_redirect "#{service_endpoint}#{extract_id link_url}", stop_indicator
end

.image_url_tco(link_url, embedly_key = nil) ⇒ Object

find the image’s url for an twitter shortened link



62
63
64
65
 
# File 'lib/tweetlr/processors/photo_service.rb', line 62

def self.image_url_tco(link_url, embedly_key = nil)
  service_url = link_url_redirect link_url
  find_image_url service_url, embedly_key
end

.image_url_twimg(link_url) ⇒ Object 



33
34
35
 
# File 'lib/tweetlr/processors/photo_service.rb', line 33

def self.image_url_twimg(link_url)
  retrieve_image_url_by_css link_url, '.media img'
end

.image_url_twitpic(link_url) ⇒ Object

find the image’s url for a twitpic link



73
74
75
 
# File 'lib/tweetlr/processors/photo_service.rb', line 73

def self.image_url_twitpic(link_url)
  image_url_redirect link_url, "http://twitpic.com/show/full/"
end

.image_url_yfrog(link_url) ⇒ Object

find the image’S url for a yfrog link



77
78
79
 
# File 'lib/tweetlr/processors/photo_service.rb', line 77

def self.image_url_yfrog(link_url)
  retrieve_image_url_by_css link_url, '#input-direct', 'value'
end

.link_url_redirect(short_url, stop_indicator = LOCATION_STOP_INDICATOR) ⇒ Object 



90
91
92
93
94
95
96
97
98
99
100
101
 
# File 'lib/tweetlr/processors/photo_service.rb', line 90

def self.link_url_redirect(short_url, stop_indicator = LOCATION_STOP_INDICATOR)
  tries = 3
  begin
    resp = Curl::Easy.http_get(short_url) { |res| res.follow_location = true }
  rescue Curl::Err::CurlError => err
    log.error "Curl::Easy.http_get failed: #{err}"
    tries -= 1
    sleep 3
    (tries > 0) ? retry : return
  end
  process_reponse_header resp, stop_indicator
end

.log ⇒ Object 



16
17
18
 
# File 'lib/tweetlr/processors/photo_service.rb', line 16

def self.log
  Tweetlr::LogAware.log #TODO why doesn't the include make the log method accessible?
end

.parse_html_for(element_signature, html_doc, identifier = "src") ⇒ Object

parse html doc for element signature



108
109
110
111
112
113
114
115
116
117
 
# File 'lib/tweetlr/processors/photo_service.rb', line 108

def self.parse_html_for(element_signature, html_doc, identifier="src")
  image_url= nil
  if html_doc
    photo_container_div = html_doc.css(element_signature)
    if photo_container_div && photo_container_div.first && photo_container_div.first.attributes[identifier]
      image_url = photo_container_div.first.attributes[identifier].value
    end
  end
  image_url
end

.photo?(link) ⇒ Boolean

Returns:

  • (Boolean) 


30
31
32
 
# File 'lib/tweetlr/processors/photo_service.rb', line 30

def self.photo?(link)
  link =~ PIC_REGEXP
end

.retrieve_image_url_by_css(link_url, css_path, selector = 'src') ⇒ Object 



118
119
120
121
122
123
 
# File 'lib/tweetlr/processors/photo_service.rb', line 118

def self.retrieve_image_url_by_css(link_url, css_path, selector='src')
  link_url = follow_redirect link_url
  response = Tweetlr::Processors::Http::http_get link_url
  image_url = parse_html_for css_path, Nokogiri::HTML.parse(response.body_str), selector
  return image_url
end