vore 0.2.4-x86_64-darwin → 0.2.5-x86_64-darwin
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 +6 -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: 677dfdb96b772683aaed9ef988f34783d0383c4293930dee4720fa366c78bf49
|
|
4
|
+
data.tar.gz: 9a96d48d28172191687eaba5bab0790989249b8c410e0b69e7ccebb286fb7723
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: 951140392a08c8599842ef135229a97bc87b7c47adb27793b07e8d48cc45b59a2cb0a80825e4722aee33d2ebc4206070ac862390ba4e7cc18719bd05992b1b03
|
|
7
|
+
data.tar.gz: 327a8aa155f4d39c3b32dd136105539d12997ae18814bf5f2251d8e948a6ec1d16e52e420409c5860c406a27c4327f418d35a95069b09aad7613c253474e8c2d
|
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 +48,18 @@ 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 = @selma.rewrite(html_file)
|
|
53
54
|
|
|
54
|
-
|
|
55
|
-
if rewritten_html_file.empty?
|
|
55
|
+
if html_file.empty?
|
|
56
56
|
results[:pages_unprocessed] += 1
|
|
57
57
|
results[:unprocessed_pages] << path
|
|
58
58
|
next
|
|
59
59
|
end
|
|
60
60
|
|
|
61
|
+
rewritten_html_file = @selma.rewrite(html_file)
|
|
62
|
+
|
|
61
63
|
# drops the first 3 parts of the path, which are "tmp", "vore", and the site name
|
|
62
64
|
url_path = path.split(FILE_SEPERATOR)[3..].join("/")
|
|
63
65
|
|
data/lib/vore/version.rb
CHANGED