vore 0.2.4-arm64-linux → 0.2.6-arm64-linux
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- 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: 72fc70a334c49c0cd3d49fa18a0754e593024555276c128503f6d2897c3bb25a
|
|
4
|
+
data.tar.gz: 47d4b811f806876617702d1d599848eaa3948f4ebae715a686d05919a37a9f1c
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: 5b6d5a89270fdf1b4630cc09ea1e7e0e9622b3af61a5965ba48718a2b6b2706d5984f59ffa510a8b192910604e7a1ff79f3b373466f3181d3f69ebd714bad3e3
|
|
7
|
+
data.tar.gz: 9906bcb9a83a4ea7c12de3be523fa3b534a6e0faeb2bf4dcc586a3e097c4efc5fd511d091641a902b4b050be6cff65e25809fb453298b587074aeda8ae3f932d
|
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