damog-feedbag 0.5.12 → 0.5.13.1
Sign up to get free protection for your applications and to get access to all the features.
- data/lib/feedbag.rb +32 -28
- metadata +1 -1
data/lib/feedbag.rb
CHANGED
@@ -50,7 +50,7 @@ module Feedbag
|
|
50
50
|
end
|
51
51
|
end
|
52
52
|
|
53
|
-
def self.find(url)
|
53
|
+
def self.find(url, args = {})
|
54
54
|
$feeds = []
|
55
55
|
|
56
56
|
url_uri = URI.parse(url)
|
@@ -65,20 +65,22 @@ module Feedbag
|
|
65
65
|
#url = "#{url_uri.scheme or 'http'}://#{url_uri.host}#{url_uri.path}"
|
66
66
|
|
67
67
|
# check if feed_valid is avail
|
68
|
-
|
69
|
-
|
70
|
-
|
71
|
-
|
72
|
-
|
73
|
-
|
74
|
-
|
75
|
-
|
76
|
-
|
77
|
-
|
78
|
-
|
79
|
-
|
80
|
-
|
81
|
-
|
68
|
+
unless args[:narrow]
|
69
|
+
begin
|
70
|
+
require "feed_validator"
|
71
|
+
v = W3C::FeedValidator.new
|
72
|
+
v.validate_url(url)
|
73
|
+
return self.add_feed(url, nil) if v.valid?
|
74
|
+
rescue LoadError
|
75
|
+
# scoo
|
76
|
+
rescue REXML::ParseException
|
77
|
+
# usually indicates timeout
|
78
|
+
# TODO: actually find out timeout. use Terminator?
|
79
|
+
$stderr.puts "Feed looked like feed but might not have passed validation or timed out"
|
80
|
+
rescue => ex
|
81
|
+
$stderr.puts "#{ex.class} error ocurred with: `#{url}': #{ex.message}"
|
82
|
+
end
|
83
|
+
end
|
82
84
|
|
83
85
|
begin
|
84
86
|
html = open(url) do |f|
|
@@ -102,19 +104,21 @@ module Feedbag
|
|
102
104
|
end
|
103
105
|
end
|
104
106
|
|
105
|
-
|
106
|
-
|
107
|
-
|
108
|
-
|
109
|
-
|
110
|
-
|
111
|
-
|
112
|
-
|
113
|
-
|
114
|
-
|
115
|
-
|
116
|
-
|
117
|
-
|
107
|
+
unless args[:narrow]
|
108
|
+
(doc/"a").each do |a|
|
109
|
+
next unless a["href"]
|
110
|
+
if self.looks_like_feed?(a["href"]) and (a["href"] =~ /\// or a["href"] =~ /#{url_uri.host}/)
|
111
|
+
self.add_feed(a["href"], url, $base_uri)
|
112
|
+
end
|
113
|
+
end
|
114
|
+
|
115
|
+
(doc/"a").each do |a|
|
116
|
+
next unless a["href"]
|
117
|
+
if self.looks_like_feed?(a["href"])
|
118
|
+
self.add_feed(a["href"], url, $base_uri)
|
119
|
+
end
|
120
|
+
end
|
121
|
+
end
|
118
122
|
|
119
123
|
end
|
120
124
|
rescue Timeout::Error => err
|