feed_parser 0.2.3 → 0.2.4
Sign up to get free protection for your applications and to get access to all the features.
- data/lib/feed_parser.rb +1 -1
- data/lib/feed_parser/feed.rb +25 -11
- data/spec/feed_parser_spec.rb +23 -3
- metadata +3 -3
data/lib/feed_parser.rb
CHANGED
data/lib/feed_parser/feed.rb
CHANGED
@@ -3,17 +3,7 @@ class FeedParser
|
|
3
3
|
attr_reader :type
|
4
4
|
|
5
5
|
def initialize(feed_url)
|
6
|
-
|
7
|
-
|
8
|
-
connection_options = {"User-Agent" => FeedParser::USER_AGENT}
|
9
|
-
connection_options[:http_basic_authentication] = parsed_url[:basic_auth] if parsed_url[:basic_auth]
|
10
|
-
|
11
|
-
raw_feed = if parsed_url[:protocol]
|
12
|
-
open(parsed_url[:url], connection_options)
|
13
|
-
else
|
14
|
-
open(parsed_url[:url])
|
15
|
-
end
|
16
|
-
|
6
|
+
raw_feed = open_or_follow_redirect(feed_url)
|
17
7
|
@feed = Nokogiri::XML(raw_feed)
|
18
8
|
@feed.remove_namespaces!
|
19
9
|
@type = (@feed.search('rss')[0] && :rss || :atom)
|
@@ -53,6 +43,30 @@ class FeedParser
|
|
53
43
|
end
|
54
44
|
|
55
45
|
private
|
46
|
+
|
47
|
+
# Some feeds
|
48
|
+
def open_or_follow_redirect(feed_url)
|
49
|
+
parsed_url = parse_url(feed_url)
|
50
|
+
|
51
|
+
connection_options = {"User-Agent" => FeedParser::USER_AGENT}
|
52
|
+
connection_options[:http_basic_authentication] = parsed_url[:basic_auth] if parsed_url[:basic_auth]
|
53
|
+
|
54
|
+
connection_options[:redirect] = true if RUBY_VERSION >= '1.9'
|
55
|
+
|
56
|
+
if parsed_url[:protocol]
|
57
|
+
open(parsed_url[:url], connection_options)
|
58
|
+
else
|
59
|
+
open(parsed_url[:url])
|
60
|
+
end
|
61
|
+
rescue RuntimeError => ex
|
62
|
+
redirect_url = ex.to_s.split(" ").last
|
63
|
+
if URI.split(feed_url).first == "http" && URI.split(redirect_url).first == "https"
|
64
|
+
open_or_follow_redirect(redirect_url)
|
65
|
+
else
|
66
|
+
raise ex
|
67
|
+
end
|
68
|
+
end
|
69
|
+
|
56
70
|
def parse_url(feed_url)
|
57
71
|
protocol, auth, *the_rest = URI.split(feed_url)
|
58
72
|
# insert a question mark in the beginning of query part of the uri
|
data/spec/feed_parser_spec.rb
CHANGED
@@ -12,20 +12,40 @@ describe FeedParser do
|
|
12
12
|
File.read(File.join(File.dirname(__FILE__), 'fixtures', 'nodeta.rss.xml'))
|
13
13
|
end
|
14
14
|
|
15
|
+
def http_connection_options
|
16
|
+
opts = {"User-Agent" => FeedParser::USER_AGENT}
|
17
|
+
opts[:redirect] = true if RUBY_VERSION >= '1.9'
|
18
|
+
opts
|
19
|
+
end
|
20
|
+
|
15
21
|
it "should fetch a feed by url" do
|
16
|
-
FeedParser::Feed.any_instance.should_receive(:open).with("http://blog.example.com/feed/",
|
22
|
+
FeedParser::Feed.any_instance.should_receive(:open).with("http://blog.example.com/feed/", http_connection_options).and_return(feed_xml)
|
17
23
|
FeedParser::Feed.new("http://blog.example.com/feed/")
|
18
24
|
end
|
19
25
|
|
20
26
|
it "should fetch a feed using basic auth if auth embedded to the url" do
|
21
|
-
FeedParser::Feed.any_instance.should_receive(:open).with("http://blog.example.com/feed/",
|
27
|
+
FeedParser::Feed.any_instance.should_receive(:open).with("http://blog.example.com/feed/", http_connection_options.merge(:http_basic_authentication => ["user", "pass"])).and_return(feed_xml)
|
22
28
|
FeedParser::Feed.new("http://user:pass@blog.example.com/feed/")
|
23
29
|
end
|
24
30
|
|
25
31
|
it "should fetch a feed with only a user name embedded to the url" do
|
26
|
-
FeedParser::Feed.any_instance.should_receive(:open).with("http://blog.example.com/feed/",
|
32
|
+
FeedParser::Feed.any_instance.should_receive(:open).with("http://blog.example.com/feed/", http_connection_options.merge(:http_basic_authentication => ["user"])).and_return(feed_xml)
|
27
33
|
FeedParser::Feed.new("http://user@blog.example.com/feed/")
|
28
34
|
end
|
35
|
+
|
36
|
+
it "should follow redirect based on the exception message" do
|
37
|
+
FeedParser::Feed.any_instance.should_receive(:open).with("http://example.com/feed", http_connection_options).and_raise(RuntimeError.new("redirection forbidden: http://example.com/feed -> https://example.com/feed"))
|
38
|
+
FeedParser::Feed.any_instance.should_receive(:open).with("https://example.com/feed", http_connection_options).and_return(feed_xml)
|
39
|
+
FeedParser::Feed.new("http://example.com/feed")
|
40
|
+
end
|
41
|
+
|
42
|
+
it "should not follow redirect from secure connection to non-secure one" do
|
43
|
+
FeedParser::Feed.any_instance.should_receive(:open).with("https://example.com/feed", http_connection_options).and_raise(RuntimeError.new("redirection forbidden: https://example.com/feed -> http://example.com/feed"))
|
44
|
+
FeedParser::Feed.any_instance.should_not_receive(:open).with("http://example.com/feed", http_connection_options)
|
45
|
+
lambda {
|
46
|
+
FeedParser::Feed.new("https://example.com/feed")
|
47
|
+
}.should raise_error(RuntimeError, "redirection forbidden: https://example.com/feed -> http://example.com/feed")
|
48
|
+
end
|
29
49
|
end
|
30
50
|
|
31
51
|
describe "#parse" do
|
metadata
CHANGED
@@ -1,13 +1,13 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: feed_parser
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
hash:
|
4
|
+
hash: 31
|
5
5
|
prerelease:
|
6
6
|
segments:
|
7
7
|
- 0
|
8
8
|
- 2
|
9
|
-
-
|
10
|
-
version: 0.2.
|
9
|
+
- 4
|
10
|
+
version: 0.2.4
|
11
11
|
platform: ruby
|
12
12
|
authors:
|
13
13
|
- Arttu Tervo
|