my-lib 0.2.1 → 0.2.2
Sign up to get free protection for your applications and to get access to all the features.
- data/VERSION +1 -1
- data/lib/lib/job.rb +196 -43
- data/lib/lib/machine.rb +65 -1
- data/my-lib.gemspec +2 -2
- metadata +23 -23
data/VERSION
CHANGED
@@ -1 +1 @@
|
|
1
|
-
0.2.
|
1
|
+
0.2.2
|
data/lib/lib/job.rb
CHANGED
@@ -1,4 +1,157 @@
|
|
1
1
|
#-*- coding:utf-8 -*-
|
2
|
+
|
3
|
+
class MyJobAnime44
|
4
|
+
def initialize(args={})
|
5
|
+
require 'kconv'
|
6
|
+
require 'mechanize'
|
7
|
+
require 'net/http'
|
8
|
+
@a = args
|
9
|
+
@debug = args[:debug] ||= false
|
10
|
+
@a[:url] ||= 'http://www.anime44.com/'
|
11
|
+
@a[:status] ||= :top_page
|
12
|
+
return if @a[:url].nil?
|
13
|
+
@agent = Mechanize.new
|
14
|
+
@a[:recursive] ||= 4
|
15
|
+
end
|
16
|
+
#
|
17
|
+
def top_page
|
18
|
+
print :top_page if @debug
|
19
|
+
@agent.get @a[:url]
|
20
|
+
# p @agent.page
|
21
|
+
@agent.page.search('table')[1].css('td li')
|
22
|
+
targs = @agent.page.search('table')[1].css('td li').select {|ne| ne.text =~ /\(Raw\)/ }
|
23
|
+
a_s = []
|
24
|
+
targs.each {|ne| a_s << ne.css('a').map{|e|e[:href]}[0]}
|
25
|
+
a_s.each do |link|
|
26
|
+
job = MyJobAnime44.new(@a.merge({
|
27
|
+
:url => link,
|
28
|
+
:status => :anime44_first
|
29
|
+
}))
|
30
|
+
@a[:machine].retry job
|
31
|
+
end
|
32
|
+
end
|
33
|
+
|
34
|
+
def anime44_first
|
35
|
+
print :anime44_first if @debug
|
36
|
+
begin
|
37
|
+
@agent.get @a[:url]
|
38
|
+
rescue => ex
|
39
|
+
p ex if @debug
|
40
|
+
p @a[:url] if @debug
|
41
|
+
return
|
42
|
+
end
|
43
|
+
@title = @agent.page.title
|
44
|
+
|
45
|
+
flg = false
|
46
|
+
|
47
|
+
daily_embeds_links = @agent.page.search('embed[@src^="http://www.daily"]').map {|ne| ne[:src] }
|
48
|
+
|
49
|
+
unless daily_embeds_links[0].nil?
|
50
|
+
uri = daily_src2hqurl(daily_embeds_links[0])
|
51
|
+
if uri[:status] == :ok
|
52
|
+
flg = true
|
53
|
+
job = MyJobAnime44.new(
|
54
|
+
@a.merge({
|
55
|
+
:url => uri[:url],
|
56
|
+
:title => @title,
|
57
|
+
:status => :anime44_fetch}))
|
58
|
+
@a[:machine].retry job
|
59
|
+
end
|
60
|
+
end
|
61
|
+
|
62
|
+
unless flg == true
|
63
|
+
begin
|
64
|
+
@agent.get @a[:url]
|
65
|
+
rescue => ex
|
66
|
+
p ex if @debug
|
67
|
+
p @a[:url] if @debug
|
68
|
+
return
|
69
|
+
end
|
70
|
+
|
71
|
+
videozer_embeds_links = @agent.page.search('embed[@src^="http://www.videozer"]').map {|ne| ne[:src] }
|
72
|
+
unless videozer_embeds_links[0].nil?
|
73
|
+
flg = true
|
74
|
+
url1 = videozer_embeds_links[0].gsub('http://www.videozer.com/embed/','http://video195.videozer.com/video?v=')
|
75
|
+
url = "#{url1}&r=1&t=#{Time.new.to_i.to_s}&u=&start=0"
|
76
|
+
job = MyJobAnime44.new(
|
77
|
+
@a.merge({
|
78
|
+
:url => url,
|
79
|
+
:title => @title,
|
80
|
+
:status => :anime44_fetch}))
|
81
|
+
@a[:machine].retry job
|
82
|
+
end
|
83
|
+
end
|
84
|
+
|
85
|
+
|
86
|
+
begin
|
87
|
+
if @a[:recursive] < 4
|
88
|
+
alignleft = @agent.page.search('div.alignleft a')[0]['href']
|
89
|
+
job = MyJobAnime44.new(
|
90
|
+
@a.merge({
|
91
|
+
:url => alignleft,
|
92
|
+
:recursive => @a[:recursive] + 1,
|
93
|
+
:status => :anime44_first}))
|
94
|
+
@a[:machine].retry job
|
95
|
+
end
|
96
|
+
rescue => ex
|
97
|
+
end
|
98
|
+
|
99
|
+
begin
|
100
|
+
if @a[:recursive] < 4
|
101
|
+
alignright = @agent.page.search('div.alignright a')[0]['href']
|
102
|
+
job = MyJobAnime44.new(
|
103
|
+
@a.merge({
|
104
|
+
:url => alignright,
|
105
|
+
:recursive => @a[:recursive] + 1,
|
106
|
+
|
107
|
+
:status => :anime44_first}))
|
108
|
+
@a[:machine].retry job
|
109
|
+
end
|
110
|
+
rescue => ex
|
111
|
+
end
|
112
|
+
end
|
113
|
+
|
114
|
+
def anime44_fetch
|
115
|
+
print "video".yellow
|
116
|
+
|
117
|
+
savedir = @a[:machine].savedir
|
118
|
+
Dir.chdir savedir
|
119
|
+
|
120
|
+
filename = "#{@a[:title].gsub(' ','').gsub(' ','')}.mp4"
|
121
|
+
savepath = "#{savedir}/#{filename}"
|
122
|
+
|
123
|
+
if File.exist?(savepath) && File.size(savepath) > 1024 * 1024 * 3
|
124
|
+
puts "File Already Saved ".yellow.bold + savepath
|
125
|
+
return
|
126
|
+
else
|
127
|
+
puts "Fetching ".green.bold + savepath
|
128
|
+
MyLogger.ln "Fetch Attempt Start ".green.bold + savepath
|
129
|
+
end
|
130
|
+
# use curl command
|
131
|
+
# no need UA...
|
132
|
+
command = "curl -# -L -R -o '#{filename}' '#{@a[:url]}' >/dev/null 2>&1 &"
|
133
|
+
puts command
|
134
|
+
system command unless @debug
|
135
|
+
end
|
136
|
+
|
137
|
+
def daily_src2hqurl(url)
|
138
|
+
require 'json'
|
139
|
+
begin
|
140
|
+
js= JSON.parse(@agent.get(url.gsub('swf','sequence/full')).body)
|
141
|
+
uri = js[0]['layerList'][0]['sequenceList'][1]['layerList'][2]['param']['videoPluginParameters']['hqURL']
|
142
|
+
rescue =>ex
|
143
|
+
return {:status => :error}
|
144
|
+
end
|
145
|
+
{:status => :ok,:url => uri}
|
146
|
+
end
|
147
|
+
|
148
|
+
def run
|
149
|
+
send @a[:status]
|
150
|
+
end
|
151
|
+
|
152
|
+
end
|
153
|
+
|
154
|
+
|
2
155
|
# Job Class for Fetch Anisoku
|
3
156
|
# Function:
|
4
157
|
# access "youtubeanisoku1.blog106.fc2.com" and crawl web site.
|
@@ -335,50 +488,50 @@ class MyJobDojinEventMachine < MyJobDojin
|
|
335
488
|
private
|
336
489
|
|
337
490
|
def do_connect
|
338
|
-
|
339
|
-
|
340
|
-
# if @machine.connection_exceed? #コネクション限界を超えていないか?
|
341
|
-
# @machine.retry(self)
|
342
|
-
# print "E".red.bold
|
343
|
-
# return
|
344
|
-
# end
|
345
|
-
|
346
|
-
|
347
|
-
|
348
|
-
|
349
|
-
|
350
|
-
|
351
|
-
|
352
|
-
|
353
|
-
|
354
|
-
|
355
|
-
|
356
|
-
|
357
|
-
|
358
|
-
|
359
|
-
|
360
|
-
|
361
|
-
|
362
|
-
|
363
|
-
|
364
|
-
|
365
|
-
|
366
|
-
|
367
|
-
|
368
|
-
|
369
|
-
|
370
|
-
|
371
|
-
|
372
|
-
|
373
|
-
|
374
|
-
|
375
|
-
|
376
|
-
|
377
|
-
end
|
378
|
-
ensure
|
379
|
-
@machine.connection_end!
|
491
|
+
return if file_already_saved?
|
492
|
+
|
493
|
+
# if @machine.connection_exceed? #コネクション限界を超えていないか?
|
494
|
+
# @machine.retry(self)
|
495
|
+
# print "E".red.bold
|
496
|
+
# return
|
497
|
+
# end
|
498
|
+
|
499
|
+
@http = EventMachine::Protocols::HttpClient.
|
500
|
+
request(
|
501
|
+
:host => @args[:server],
|
502
|
+
:port => @args[:port],
|
503
|
+
:request => @args[:path],
|
504
|
+
:cookie => @args[:cookie]['Cookie']
|
505
|
+
)
|
506
|
+
@machine.connection_count!
|
507
|
+
@http.errback{
|
508
|
+
begin
|
509
|
+
ensure
|
510
|
+
@machine.connection_end!
|
511
|
+
end
|
512
|
+
}
|
513
|
+
|
514
|
+
@http.callback {|response|
|
515
|
+
begin
|
516
|
+
if response[:status] == 200
|
517
|
+
# 200 はレスポンスの中身を保存する
|
518
|
+
save_content(response[:content])
|
519
|
+
elsif response[:status] == 503 ||
|
520
|
+
response[:status] == 500 ||
|
521
|
+
response[:status] == 403
|
522
|
+
# 503/500/403はリトライする
|
523
|
+
@args[:try] += 1
|
524
|
+
@machine.retry(self) if @args[:try] < 6
|
525
|
+
elsif response[:status] == 404
|
526
|
+
# 404は終了する
|
527
|
+
else
|
528
|
+
puts response[:status].to_s.red.bold
|
529
|
+
puts response[:headers].to_s.red.bold
|
380
530
|
end
|
381
|
-
|
531
|
+
ensure
|
532
|
+
@machine.connection_end!
|
533
|
+
end
|
534
|
+
}
|
382
535
|
end
|
383
536
|
end
|
384
537
|
|
data/lib/lib/machine.rb
CHANGED
@@ -129,6 +129,61 @@ class MyEventMachineDojin
|
|
129
129
|
|
130
130
|
end
|
131
131
|
|
132
|
+
class MyMachineAnime44
|
133
|
+
include MyMachine
|
134
|
+
|
135
|
+
attr_reader :savedir
|
136
|
+
|
137
|
+
def initialize(args={ })
|
138
|
+
require 'eventmachine'
|
139
|
+
super(args)
|
140
|
+
args[:savedir] ||= "#{ENV['HOME']}/Desktop/video"
|
141
|
+
@savedir = args[:savedir]
|
142
|
+
begin
|
143
|
+
Dir::mkdir(@savedir, 0777)
|
144
|
+
rescue => ex
|
145
|
+
warn ex
|
146
|
+
end
|
147
|
+
@args = args
|
148
|
+
@args[:recursive] ||= 2
|
149
|
+
@gaman = 0;
|
150
|
+
end
|
151
|
+
|
152
|
+
# machine go to run eventmachine
|
153
|
+
def go
|
154
|
+
EM.run do
|
155
|
+
EM.add_periodic_timer(0.00001) do
|
156
|
+
# print "loop".green
|
157
|
+
if should_stop_machine?
|
158
|
+
EM.stop
|
159
|
+
end
|
160
|
+
@queue.pop.run unless @queue.empty?
|
161
|
+
end
|
162
|
+
end
|
163
|
+
puts "End of fetch".green.bold
|
164
|
+
end
|
165
|
+
|
166
|
+
# setup jobs
|
167
|
+
def setupjobs
|
168
|
+
ajob = MyJobAnime44
|
169
|
+
.new(
|
170
|
+
:machine => self,
|
171
|
+
:recursive => @args[:recursive],
|
172
|
+
:debug => @debug,
|
173
|
+
)
|
174
|
+
@queue.push ajob
|
175
|
+
end
|
176
|
+
|
177
|
+
# Machineは終了すべきか?
|
178
|
+
def should_stop_machine?
|
179
|
+
@gaman += 1 if @queue.size < 3
|
180
|
+
if @gaman > 1000
|
181
|
+
return true
|
182
|
+
end
|
183
|
+
end
|
184
|
+
end
|
185
|
+
|
186
|
+
|
132
187
|
#
|
133
188
|
# Class of Machine by EventMachine.
|
134
189
|
# this class controll jobs for Anisoku
|
@@ -167,7 +222,7 @@ class MyMachineAnisoku
|
|
167
222
|
@args = args
|
168
223
|
@args[:limit] ||= 7
|
169
224
|
@args[:recent] ||= 7
|
170
|
-
|
225
|
+
@args[:mode] ||= :anisoku
|
171
226
|
make_filelist
|
172
227
|
make_dellist
|
173
228
|
end
|
@@ -267,6 +322,15 @@ class MyMachineAnisoku
|
|
267
322
|
:recent => @args[:recent],
|
268
323
|
:debug => @debug,
|
269
324
|
)
|
325
|
+
if @args[:mode] == :anime44
|
326
|
+
ajob = MyJobAnime44
|
327
|
+
.new(
|
328
|
+
:machine => self,
|
329
|
+
:limit => @args[:limit],
|
330
|
+
:recent => @args[:recent],
|
331
|
+
:debug => @debug,
|
332
|
+
)
|
333
|
+
end
|
270
334
|
@queue.push ajob
|
271
335
|
end
|
272
336
|
|
data/my-lib.gemspec
CHANGED
@@ -5,11 +5,11 @@
|
|
5
5
|
|
6
6
|
Gem::Specification.new do |s|
|
7
7
|
s.name = "my-lib"
|
8
|
-
s.version = "0.2.
|
8
|
+
s.version = "0.2.2"
|
9
9
|
|
10
10
|
s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
|
11
11
|
s.authors = ["modeverv"]
|
12
|
-
s.date = "2011-10-
|
12
|
+
s.date = "2011-10-17"
|
13
13
|
s.description = "longer description of your gem"
|
14
14
|
s.email = "modeverv@gmail.com"
|
15
15
|
s.extra_rdoc_files = [
|
metadata
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: my-lib
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.2.
|
4
|
+
version: 0.2.2
|
5
5
|
prerelease:
|
6
6
|
platform: ruby
|
7
7
|
authors:
|
@@ -9,11 +9,11 @@ authors:
|
|
9
9
|
autorequire:
|
10
10
|
bindir: bin
|
11
11
|
cert_chain: []
|
12
|
-
date: 2011-10-
|
12
|
+
date: 2011-10-17 00:00:00.000000000Z
|
13
13
|
dependencies:
|
14
14
|
- !ruby/object:Gem::Dependency
|
15
15
|
name: term-ansicolor
|
16
|
-
requirement: &
|
16
|
+
requirement: &2153870560 !ruby/object:Gem::Requirement
|
17
17
|
none: false
|
18
18
|
requirements:
|
19
19
|
- - ! '>='
|
@@ -21,10 +21,10 @@ dependencies:
|
|
21
21
|
version: '0'
|
22
22
|
type: :runtime
|
23
23
|
prerelease: false
|
24
|
-
version_requirements: *
|
24
|
+
version_requirements: *2153870560
|
25
25
|
- !ruby/object:Gem::Dependency
|
26
26
|
name: eventmachine
|
27
|
-
requirement: &
|
27
|
+
requirement: &2153868700 !ruby/object:Gem::Requirement
|
28
28
|
none: false
|
29
29
|
requirements:
|
30
30
|
- - ! '>='
|
@@ -32,10 +32,10 @@ dependencies:
|
|
32
32
|
version: '0'
|
33
33
|
type: :runtime
|
34
34
|
prerelease: false
|
35
|
-
version_requirements: *
|
35
|
+
version_requirements: *2153868700
|
36
36
|
- !ruby/object:Gem::Dependency
|
37
37
|
name: mechanize
|
38
|
-
requirement: &
|
38
|
+
requirement: &2153867940 !ruby/object:Gem::Requirement
|
39
39
|
none: false
|
40
40
|
requirements:
|
41
41
|
- - ! '>='
|
@@ -43,10 +43,10 @@ dependencies:
|
|
43
43
|
version: '0'
|
44
44
|
type: :runtime
|
45
45
|
prerelease: false
|
46
|
-
version_requirements: *
|
46
|
+
version_requirements: *2153867940
|
47
47
|
- !ruby/object:Gem::Dependency
|
48
48
|
name: rspec
|
49
|
-
requirement: &
|
49
|
+
requirement: &2153867020 !ruby/object:Gem::Requirement
|
50
50
|
none: false
|
51
51
|
requirements:
|
52
52
|
- - ! '>'
|
@@ -54,10 +54,10 @@ dependencies:
|
|
54
54
|
version: 2.3.0
|
55
55
|
type: :development
|
56
56
|
prerelease: false
|
57
|
-
version_requirements: *
|
57
|
+
version_requirements: *2153867020
|
58
58
|
- !ruby/object:Gem::Dependency
|
59
59
|
name: yard
|
60
|
-
requirement: &
|
60
|
+
requirement: &2153866240 !ruby/object:Gem::Requirement
|
61
61
|
none: false
|
62
62
|
requirements:
|
63
63
|
- - ! '>'
|
@@ -65,10 +65,10 @@ dependencies:
|
|
65
65
|
version: 0.6.0
|
66
66
|
type: :development
|
67
67
|
prerelease: false
|
68
|
-
version_requirements: *
|
68
|
+
version_requirements: *2153866240
|
69
69
|
- !ruby/object:Gem::Dependency
|
70
70
|
name: bundler
|
71
|
-
requirement: &
|
71
|
+
requirement: &2153865400 !ruby/object:Gem::Requirement
|
72
72
|
none: false
|
73
73
|
requirements:
|
74
74
|
- - ! '>'
|
@@ -76,10 +76,10 @@ dependencies:
|
|
76
76
|
version: 1.0.0
|
77
77
|
type: :development
|
78
78
|
prerelease: false
|
79
|
-
version_requirements: *
|
79
|
+
version_requirements: *2153865400
|
80
80
|
- !ruby/object:Gem::Dependency
|
81
81
|
name: jeweler
|
82
|
-
requirement: &
|
82
|
+
requirement: &2153848220 !ruby/object:Gem::Requirement
|
83
83
|
none: false
|
84
84
|
requirements:
|
85
85
|
- - ! '>='
|
@@ -87,10 +87,10 @@ dependencies:
|
|
87
87
|
version: '0'
|
88
88
|
type: :development
|
89
89
|
prerelease: false
|
90
|
-
version_requirements: *
|
90
|
+
version_requirements: *2153848220
|
91
91
|
- !ruby/object:Gem::Dependency
|
92
92
|
name: rcov
|
93
|
-
requirement: &
|
93
|
+
requirement: &2153846360 !ruby/object:Gem::Requirement
|
94
94
|
none: false
|
95
95
|
requirements:
|
96
96
|
- - ! '>='
|
@@ -98,10 +98,10 @@ dependencies:
|
|
98
98
|
version: '0'
|
99
99
|
type: :development
|
100
100
|
prerelease: false
|
101
|
-
version_requirements: *
|
101
|
+
version_requirements: *2153846360
|
102
102
|
- !ruby/object:Gem::Dependency
|
103
103
|
name: guard
|
104
|
-
requirement: &
|
104
|
+
requirement: &2153845280 !ruby/object:Gem::Requirement
|
105
105
|
none: false
|
106
106
|
requirements:
|
107
107
|
- - ! '>='
|
@@ -109,10 +109,10 @@ dependencies:
|
|
109
109
|
version: '0'
|
110
110
|
type: :development
|
111
111
|
prerelease: false
|
112
|
-
version_requirements: *
|
112
|
+
version_requirements: *2153845280
|
113
113
|
- !ruby/object:Gem::Dependency
|
114
114
|
name: guard-spork
|
115
|
-
requirement: &
|
115
|
+
requirement: &2153844520 !ruby/object:Gem::Requirement
|
116
116
|
none: false
|
117
117
|
requirements:
|
118
118
|
- - ! '>='
|
@@ -120,7 +120,7 @@ dependencies:
|
|
120
120
|
version: '0'
|
121
121
|
type: :development
|
122
122
|
prerelease: false
|
123
|
-
version_requirements: *
|
123
|
+
version_requirements: *2153844520
|
124
124
|
description: longer description of your gem
|
125
125
|
email: modeverv@gmail.com
|
126
126
|
executables: []
|
@@ -168,7 +168,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
|
|
168
168
|
version: '0'
|
169
169
|
segments:
|
170
170
|
- 0
|
171
|
-
hash: -
|
171
|
+
hash: -347030689861470656
|
172
172
|
required_rubygems_version: !ruby/object:Gem::Requirement
|
173
173
|
none: false
|
174
174
|
requirements:
|