ece_prewarmer 0.1.3 → 0.1.5
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/README.md +8 -0
- data/bin/ece-prewarmer +25 -1
- data/lib/ece_prewarmer/version.rb +1 -1
- data/lib/ece_prewarmer.rb +7 -3
- metadata +2 -2
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA1:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 60b9d13d4482a9a9310981a92a63c716807c1c93
|
4
|
+
data.tar.gz: 61bc0497385142dff1d5bb8ba19f7fa487046e86
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: b721be666c2ae0754ce3d2661cac1a96e9d5590fee39fba3c3577b9c4e786298a375bc722364c8d8c2dfb857593d7196eea9ac796a3396e08fcde63ef13c0764
|
7
|
+
data.tar.gz: 74ffb3a1ced07098ede80272174fca72fb11fde435e5920c44659fe15cca00470bd6d5f09ab52d1189aca5c6b8e6322eec482f9c654d8838f287bd29731443e2
|
data/README.md
CHANGED
@@ -8,6 +8,14 @@ Se usa mas o menos así:
|
|
8
8
|
[host ]$ screen
|
9
9
|
[host ]$ unset http_proxy
|
10
10
|
[host ]$ ece-prewarmer -s www.24horas.cl -i 10.132.17.114 -p 8080 -v -t 8
|
11
|
+
http://www.24horas.cl:8080/ Queue: 0
|
12
|
+
http://www.24horas.cl:8080/envivo/ Queue: 104
|
13
|
+
http://www.24horas.cl:8080/internacional/ Queue: 103
|
14
|
+
http://www.24horas.cl:8080/nacional/ Queue: 137
|
15
|
+
http://www.24horas.cl:8080/videos/ Queue: 169
|
16
|
+
http://www.24horas.cl:8080/galerias/ Queue: 168
|
17
|
+
http://www.24horas.cl:8080/envivo/?articleId=221352 Queue: 167
|
18
|
+
http://www.24horas.cl:8080/envivo/?articleId=110846 Queue: 166
|
11
19
|
```
|
12
20
|
|
13
21
|
## Installation
|
data/bin/ece-prewarmer
CHANGED
@@ -28,6 +28,10 @@ optparse = OptionParser.new do |opts|
|
|
28
28
|
options[:depth_limit] = o.nil? ? nil : o.to_i
|
29
29
|
end
|
30
30
|
|
31
|
+
opts.on('-dUSERAGENT', '--depth=USER-AGENT', 'USER-AGENT a usar') do |o|
|
32
|
+
options[:user_agent] = o.nil? ? nil : o.to_i
|
33
|
+
end
|
34
|
+
|
31
35
|
opts.on('-tTHREADS', '--threads=THREADS', 'Procesos a lanzar') do |o|
|
32
36
|
options[:threads] = o.nil? ? nil : o.to_i
|
33
37
|
end
|
@@ -55,12 +59,32 @@ opts[:verbose] = options[:verbose]
|
|
55
59
|
opts[:port] = options[:port]
|
56
60
|
opts[:threads] = options[:threads]
|
57
61
|
opts[:depth_limit] = options[:depth_limit]
|
62
|
+
opts[:user_agent] = options[:user_agent]
|
63
|
+
|
64
|
+
def shutdown
|
65
|
+
Thread.list.each do |thread|
|
66
|
+
thread.exit unless thread == Thread.current
|
67
|
+
end
|
68
|
+
puts "Eliminando entrada de hosts file"
|
69
|
+
FileUtils.cp "/tmp/hosts", '/etc/hosts'
|
70
|
+
end
|
71
|
+
|
72
|
+
# Trap ^C
|
73
|
+
Signal.trap("INT") {
|
74
|
+
shutdown
|
75
|
+
exit
|
76
|
+
}
|
77
|
+
|
78
|
+
# Trap `Kill `
|
79
|
+
Signal.trap("TERM") {
|
80
|
+
shutdown
|
81
|
+
exit
|
82
|
+
}
|
58
83
|
|
59
84
|
results = EcePrewarmer.prewarm(host, website, opts)
|
60
85
|
|
61
86
|
puts "Paginas visitadas: #{results.size}\n"
|
62
87
|
|
63
|
-
|
64
88
|
BEGIN {
|
65
89
|
at_exit {
|
66
90
|
EcePrewarmer.remove_host_alias(website)
|
data/lib/ece_prewarmer.rb
CHANGED
@@ -7,6 +7,7 @@ module EcePrewarmer
|
|
7
7
|
|
8
8
|
def self.add_host_alias(ip, host_alias, file = nil)
|
9
9
|
file ||= '/etc/hosts'
|
10
|
+
FileUtils.cp('/etc/hosts', '/tmp/hosts') if file == '/etc/hosts'
|
10
11
|
hosts = Hosts::File.read(file)
|
11
12
|
hosts.elements << Hosts::Entry.new(ip, host_alias)
|
12
13
|
hosts.invalidate_cache!
|
@@ -22,6 +23,7 @@ module EcePrewarmer
|
|
22
23
|
end
|
23
24
|
to_delete.compact.each { |i| hosts.elements[i] = nil }
|
24
25
|
hosts.elements.compact!
|
26
|
+
hosts.invalidate_cache!
|
25
27
|
hosts.write
|
26
28
|
end
|
27
29
|
|
@@ -29,14 +31,14 @@ module EcePrewarmer
|
|
29
31
|
opts = parse_opts opts
|
30
32
|
url = "http://#{website}:#{opts[:port]}/"
|
31
33
|
add_host_alias(host, website, opts[:hostfile])
|
32
|
-
results = crawl(url, opts[:port], opts[:verbose], opts[:threads], opts[:depth_limit])
|
34
|
+
results = crawl(url, opts[:port], opts[:verbose], opts[:threads], opts[:depth_limit], opts[:user_agent])
|
33
35
|
remove_host_alias(website, opts[:hostfile])
|
34
36
|
results
|
35
37
|
end
|
36
38
|
|
37
|
-
def self.crawl(url, port, verbose, threads, depth)
|
39
|
+
def self.crawl(url, port, verbose, threads, depth, user_agent)
|
38
40
|
result = []
|
39
|
-
Anemone.crawl(url, verbose: verbose, threads: threads, depth_limit: depth) do |anemone|
|
41
|
+
Anemone.crawl(url, verbose: verbose, threads: threads, depth_limit: depth, user_agent: user_agent) do |anemone|
|
40
42
|
links = []
|
41
43
|
anemone.on_every_page do |page|
|
42
44
|
links.push page.url
|
@@ -58,6 +60,8 @@ module EcePrewarmer
|
|
58
60
|
opts[:verbose] ||= false
|
59
61
|
opts[:port] ||= 80
|
60
62
|
opts[:depth_limit] ||= false
|
63
|
+
# Chrome by default
|
64
|
+
opts[:user_agent] ||= "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/44.0.2403.155 Safari/537.36"
|
61
65
|
opts
|
62
66
|
end
|
63
67
|
|
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: ece_prewarmer
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.1.
|
4
|
+
version: 0.1.5
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Patricio Bruna
|
8
8
|
autorequire:
|
9
9
|
bindir: bin
|
10
10
|
cert_chain: []
|
11
|
-
date: 2015-
|
11
|
+
date: 2015-08-26 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: anemone
|