vore 0.2.4-x86_64-linux → 0.2.6-x86_64-linux
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/exe/vore-spider +0 -0
- data/lib/vore/crawler.rb +14 -4
- data/lib/vore/version.rb +1 -1
- metadata +1 -1
checksums.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA256:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: 53cb93bee6f3736369589c1eaaf7b739ca6f898e814d33d4ff747f4daa2077fc
|
|
4
|
+
data.tar.gz: 100cf434502ce236ce6f31aa24c110ed43274043f1ea7a678f4cad9905079e7a
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: b66ad8d522524716bc1f5a904b428934ef5006916cad623893301569afcfd7b255c021d1bc3f1205cd757365a165b6b88dfe278e9979aba4d900fc913f074d73
|
|
7
|
+
data.tar.gz: 9e176c95395e40503fc8a5af9f014a5befc37a63d04b71a0d64549be100700b92abbb439f57be4ab4699a3543ad2ca32203095b83e402774b35e70ae58d6cde2
|
data/exe/vore-spider
CHANGED
|
Binary file
|
data/lib/vore/crawler.rb
CHANGED
|
@@ -31,7 +31,7 @@ module Vore
|
|
|
31
31
|
|
|
32
32
|
output = %x(#{@executable} \
|
|
33
33
|
--user-agent #{user_agent} \
|
|
34
|
-
--delay
|
|
34
|
+
--delay 3500 \
|
|
35
35
|
--url #{website} \
|
|
36
36
|
download \
|
|
37
37
|
-t \
|
|
@@ -48,16 +48,26 @@ module Vore
|
|
|
48
48
|
Dir.glob(File.join(output_dir, "**", "*")).each do |path|
|
|
49
49
|
next unless File.file?(path)
|
|
50
50
|
|
|
51
|
+
results[:pages_visited] += 1
|
|
52
|
+
|
|
51
53
|
html_file = File.read(path).force_encoding("UTF-8")
|
|
52
|
-
rewritten_html_file =
|
|
54
|
+
rewritten_html_file = ""
|
|
53
55
|
|
|
54
|
-
|
|
55
|
-
|
|
56
|
+
if html_file.empty?
|
|
57
|
+
Vore.logger.warn("HTML file empty: #{path}")
|
|
56
58
|
results[:pages_unprocessed] += 1
|
|
57
59
|
results[:unprocessed_pages] << path
|
|
58
60
|
next
|
|
59
61
|
end
|
|
60
62
|
|
|
63
|
+
begin
|
|
64
|
+
rewritten_html_file = @selma.rewrite(html_file)
|
|
65
|
+
rescue StandardError => e
|
|
66
|
+
Vore.logger.warn("Error rewriting #{path}: #{e}")
|
|
67
|
+
results[:pages_unprocessed] += 1
|
|
68
|
+
next
|
|
69
|
+
end
|
|
70
|
+
|
|
61
71
|
# drops the first 3 parts of the path, which are "tmp", "vore", and the site name
|
|
62
72
|
url_path = path.split(FILE_SEPERATOR)[3..].join("/")
|
|
63
73
|
|
data/lib/vore/version.rb
CHANGED