coolCrawler 0.4.3 → 0.4.4
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/lib/coolCrawler/version.rb +1 -1
- data/lib/cool_crawler.rb +6 -6
- 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: 79059c4a0ef1c026082b3ef2b25ee13c2310984a2049be4345f0345b226d2e99
|
|
4
|
+
data.tar.gz: 32d02d444ef3553df02c2764073b1d9f5cbdabea272ee1d2b3eb9e80a5a29434
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: 56d89149672219eb082eb188d2d79fde2d97931efe8295983719a5693c5b96c07e607a309d03690e8a739a2a3018f73c7d0fbd18da6003cbafa7969633890b2e
|
|
7
|
+
data.tar.gz: f37a6061b4b318cb19423ab4de961a98b625d1e17e6f85eb78524794a967a2f5dda4bceaa5c5414b2df33a359c9c0fd5a8e8a7bfe22c4f872e41a9e9c5f46d48
|
data/lib/coolCrawler/version.rb
CHANGED
data/lib/cool_crawler.rb
CHANGED
|
@@ -31,7 +31,7 @@ module CoolCrawler
|
|
|
31
31
|
end
|
|
32
32
|
|
|
33
33
|
def run
|
|
34
|
-
until queue.empty?
|
|
34
|
+
until queue.empty? || @visited_pages >= @max_pages
|
|
35
35
|
send_crawlers
|
|
36
36
|
sleep(delay)
|
|
37
37
|
end
|
|
@@ -43,7 +43,10 @@ module CoolCrawler
|
|
|
43
43
|
|
|
44
44
|
def send_crawlers
|
|
45
45
|
pages = []
|
|
46
|
-
|
|
46
|
+
until queue.empty? || pages.size >= max_connections || @visited_pages >= @max_pages
|
|
47
|
+
pages << queue.pop
|
|
48
|
+
@visited_pages += 1
|
|
49
|
+
end
|
|
47
50
|
Async do
|
|
48
51
|
internet = Async::HTTP::Internet.new
|
|
49
52
|
barrier = Async::Barrier.new
|
|
@@ -76,9 +79,7 @@ module CoolCrawler
|
|
|
76
79
|
next if a["href"].nil?
|
|
77
80
|
uri_a = URI(a["href"].strip.split('#')[0].sub(/\\|(\s+$)/, ""))
|
|
78
81
|
begin
|
|
79
|
-
|
|
80
|
-
links << URI.join(page, uri_a).path if (uri_a.host == uri.host || uri_a.host.nil?) && uri_a.path
|
|
81
|
-
end
|
|
82
|
+
links << URI.join(page, uri_a).path if (uri_a.host == uri.host || uri_a.host.nil?) && uri_a.path
|
|
82
83
|
rescue
|
|
83
84
|
# do nothing
|
|
84
85
|
end
|
|
@@ -104,7 +105,6 @@ module CoolCrawler
|
|
|
104
105
|
else
|
|
105
106
|
visited[path] = 1
|
|
106
107
|
end
|
|
107
|
-
@visited_pages += 1
|
|
108
108
|
end
|
|
109
109
|
|
|
110
110
|
def sorted_visited
|