right_scraper 1.0.9 → 1.0.10
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.
- data/lib/right_scraper/linux/process_monitor.rb +67 -0
- data/lib/right_scraper/scraper_base.rb +1 -1
- data/lib/right_scraper/scrapers/git_scraper.rb +1 -1
- data/lib/right_scraper/watcher.rb +17 -15
- data/lib/right_scraper/win32/process_monitor.rb +91 -0
- data/right_scraper.gemspec +1 -1
- data/spec/download_scraper_spec.rb +1 -0
- data/spec/scraper_spec.rb +11 -9
- data/spec/svn_scraper_spec.rb +1 -0
- data/spec/watcher_spec.rb +1 -1
- metadata +4 -2
@@ -0,0 +1,67 @@
|
|
1
|
+
#--
|
2
|
+
# Copyright: Copyright (c) 2010 RightScale, Inc.
|
3
|
+
#
|
4
|
+
# Permission is hereby granted, free of charge, to any person obtaining
|
5
|
+
# a copy of this software and associated documentation files (the
|
6
|
+
# 'Software'), to deal in the Software without restriction, including
|
7
|
+
# without limitation the rights to use, copy, modify, merge, publish,
|
8
|
+
# distribute, sublicense, and/or sell copies of the Software, and to
|
9
|
+
# permit persons to whom the Software is furnished to do so, subject to
|
10
|
+
# the following conditions:
|
11
|
+
#
|
12
|
+
# The above copyright notice and this permission notice shall be
|
13
|
+
# included in all copies or substantial portions of the Software.
|
14
|
+
#
|
15
|
+
# THE SOFTWARE IS PROVIDED 'AS IS', WITHOUT WARRANTY OF ANY KIND,
|
16
|
+
# EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
|
17
|
+
# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
|
18
|
+
# IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
|
19
|
+
# CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
|
20
|
+
# TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
|
21
|
+
# SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
22
|
+
#++
|
23
|
+
|
24
|
+
module RightScale
|
25
|
+
|
26
|
+
# *nix specific watcher implementation
|
27
|
+
class ProcessMonitor
|
28
|
+
|
29
|
+
# Spawn given process and callback given block with output and exit code
|
30
|
+
#
|
31
|
+
# === Parameters
|
32
|
+
# cmd(String):: Process command line (including arguments)
|
33
|
+
#
|
34
|
+
# === Block
|
35
|
+
# Given block should take one argument which is a hash which may contain
|
36
|
+
# the keys :output and :exit_code. The value associated with :output is a chunk
|
37
|
+
# of output while the value associated with :exit_code is the process exit code
|
38
|
+
# This block won't be called anymore once the :exit_code key has associated value
|
39
|
+
#
|
40
|
+
# === Return
|
41
|
+
# pid(Integer):: Spawned process pid
|
42
|
+
def spawn(cmd)
|
43
|
+
# Run external process and monitor it in a new thread
|
44
|
+
@io = IO.popen(cmd)
|
45
|
+
@reader = Thread.new do
|
46
|
+
o = @io.read
|
47
|
+
until o == ''
|
48
|
+
yield(:output => o)
|
49
|
+
o = @io.read
|
50
|
+
end
|
51
|
+
Process.wait(@io.pid)
|
52
|
+
yield(:exit_code => $?.exitstatus)
|
53
|
+
end
|
54
|
+
@io.pid
|
55
|
+
end
|
56
|
+
|
57
|
+
# Close io and join reader thread
|
58
|
+
#
|
59
|
+
# === Return
|
60
|
+
# true:: Always return true
|
61
|
+
def cleanup
|
62
|
+
@reader.join
|
63
|
+
@io.close
|
64
|
+
end
|
65
|
+
|
66
|
+
end
|
67
|
+
end
|
@@ -56,7 +56,7 @@ module RightScale
|
|
56
56
|
# Each scraper implementation should override scrape_imp which is called
|
57
57
|
# after this method initializes all the scraper attributes properly.
|
58
58
|
# See RightScale::Scraper#scrape
|
59
|
-
def scrape(repo, incremental, &callback)
|
59
|
+
def scrape(repo, incremental=true, &callback)
|
60
60
|
@repo = repo
|
61
61
|
@callback = callback
|
62
62
|
@current_repo_dir = ScraperBase.repo_dir(root_dir, repo)
|
@@ -182,7 +182,7 @@ module RightScale
|
|
182
182
|
# "-o StrictHostKeyChecking=no" in the GIT_SSH executable, but it is
|
183
183
|
# still a mystery why this doesn't work properly in windows.
|
184
184
|
# so make a ssh call which creates the proper "known_hosts" file.
|
185
|
-
system("ssh -o StrictHostKeyChecking=no #{repo.url.split(':').first} exit")
|
185
|
+
system("ssh -o StrictHostKeyChecking=no #{repo.url.split(':').first} exit 2>&1")
|
186
186
|
end
|
187
187
|
return ''
|
188
188
|
end
|
@@ -22,6 +22,11 @@
|
|
22
22
|
#++
|
23
23
|
|
24
24
|
require 'find'
|
25
|
+
if RUBY_PLATFORM =~ /mswin/
|
26
|
+
require File.join(File.dirname(__FILE__), 'win32', 'process_monitor')
|
27
|
+
else
|
28
|
+
require File.join(File.dirname(__FILE__), 'linux', 'process_monitor')
|
29
|
+
end
|
25
30
|
|
26
31
|
module RightScale
|
27
32
|
|
@@ -72,18 +77,13 @@ module RightScale
|
|
72
77
|
# res(RightScale::WatchStatus):: Outcome of watch, see RightScale::WatchStatus
|
73
78
|
def launch_and_watch(cmd, dest_dir)
|
74
79
|
exit_code = nil
|
75
|
-
output
|
76
|
-
|
77
|
-
|
78
|
-
|
79
|
-
|
80
|
-
|
81
|
-
|
82
|
-
output += o
|
83
|
-
o = io.read
|
84
|
-
end
|
85
|
-
Process.wait(io.pid)
|
86
|
-
exit_code = $?.exitstatus
|
80
|
+
output = ''
|
81
|
+
monitor = ProcessMonitor.new
|
82
|
+
|
83
|
+
# Run external process and monitor it in a new thread, platform specific
|
84
|
+
pid = monitor.spawn(cmd) do |data|
|
85
|
+
output << data[:output] if data.include?(:output)
|
86
|
+
exit_code = data[:exit_code] if data.include?(:exit_code)
|
87
87
|
end
|
88
88
|
|
89
89
|
# Loop until process is done or times out or takes too much space
|
@@ -106,10 +106,12 @@ module RightScale
|
|
106
106
|
else
|
107
107
|
exit_status = -1
|
108
108
|
outcome = (timed_out ? :timeout : :size_exceeded)
|
109
|
-
Process.kill('INT',
|
109
|
+
Process.kill('INT', pid)
|
110
110
|
end
|
111
|
-
|
112
|
-
|
111
|
+
|
112
|
+
# Cleanup any open handle etc., platform specific
|
113
|
+
monitor.cleanup
|
114
|
+
|
113
115
|
res = WatchStatus.new(outcome, exit_status, output)
|
114
116
|
end
|
115
117
|
|
@@ -0,0 +1,91 @@
|
|
1
|
+
#--
|
2
|
+
# Copyright: Copyright (c) 2010 RightScale, Inc.
|
3
|
+
#
|
4
|
+
# Permission is hereby granted, free of charge, to any person obtaining
|
5
|
+
# a copy of this software and associated documentation files (the
|
6
|
+
# 'Software'), to deal in the Software without restriction, including
|
7
|
+
# without limitation the rights to use, copy, modify, merge, publish,
|
8
|
+
# distribute, sublicense, and/or sell copies of the Software, and to
|
9
|
+
# permit persons to whom the Software is furnished to do so, subject to
|
10
|
+
# the following conditions:
|
11
|
+
#
|
12
|
+
# The above copyright notice and this permission notice shall be
|
13
|
+
# included in all copies or substantial portions of the Software.
|
14
|
+
#
|
15
|
+
# THE SOFTWARE IS PROVIDED 'AS IS', WITHOUT WARRANTY OF ANY KIND,
|
16
|
+
# EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
|
17
|
+
# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
|
18
|
+
# IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
|
19
|
+
# CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
|
20
|
+
# TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
|
21
|
+
# SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
22
|
+
#++
|
23
|
+
|
24
|
+
require 'win32/process'
|
25
|
+
|
26
|
+
module RightScale
|
27
|
+
|
28
|
+
# Windows specific watcher implementation
|
29
|
+
class ProcessMonitor
|
30
|
+
|
31
|
+
include Windows::Process
|
32
|
+
include Windows::Synchronize
|
33
|
+
include Windows::Handle
|
34
|
+
|
35
|
+
# Spawn given process and callback given block with output and exit code
|
36
|
+
#
|
37
|
+
# === Parameters
|
38
|
+
# cmd(String):: Process command line (including arguments)
|
39
|
+
#
|
40
|
+
# === Block
|
41
|
+
# Given block should take one argument which is a hash which may contain
|
42
|
+
# the keys :output and :exit_code. The value associated with :output is a chunk
|
43
|
+
# of output while the value associated with :exit_code is the process exit code
|
44
|
+
# This block won't be called anymore once the :exit_code key has associated value
|
45
|
+
#
|
46
|
+
# === Return
|
47
|
+
# pid(Integer):: Spawned process pid
|
48
|
+
def spawn(cmd)
|
49
|
+
# Run external process and monitor it in a new thread
|
50
|
+
@io = IO.popen(cmd)
|
51
|
+
@handle = OpenProcess(PROCESS_ALL_ACCESS, 0, @io.pid)
|
52
|
+
case @handle
|
53
|
+
when INVALID_HANDLE_VALUE
|
54
|
+
# Something bad happened
|
55
|
+
yield(:exit_code => 1)
|
56
|
+
when 0
|
57
|
+
# Process already finished
|
58
|
+
yield(:exit_code => 0)
|
59
|
+
else
|
60
|
+
# Start output read
|
61
|
+
@reader = Thread.new do
|
62
|
+
o = @io.read
|
63
|
+
until o == ''
|
64
|
+
yield(:output => o)
|
65
|
+
o = @io.read
|
66
|
+
end
|
67
|
+
status = WaitForSingleObject(@handle, INFINITE)
|
68
|
+
exit_code = [0].pack('L')
|
69
|
+
if GetExitCodeProcess(@handle, exit_code)
|
70
|
+
exit_code = exit_code.unpack('L').first
|
71
|
+
else
|
72
|
+
exit_code = 1
|
73
|
+
end
|
74
|
+
yield(:exit_code => exit_code)
|
75
|
+
end
|
76
|
+
end
|
77
|
+
@io.pid
|
78
|
+
end
|
79
|
+
|
80
|
+
# Cleanup underlying handle
|
81
|
+
#
|
82
|
+
# === Return
|
83
|
+
# true:: Always return true
|
84
|
+
def cleanup
|
85
|
+
@reader.join
|
86
|
+
CloseHandle(@handle) if @handle > 0
|
87
|
+
@io.close
|
88
|
+
end
|
89
|
+
|
90
|
+
end
|
91
|
+
end
|
data/right_scraper.gemspec
CHANGED
@@ -23,7 +23,7 @@ require 'rubygems'
|
|
23
23
|
|
24
24
|
spec = Gem::Specification.new do |spec|
|
25
25
|
spec.name = 'right_scraper'
|
26
|
-
spec.version = '1.0.
|
26
|
+
spec.version = '1.0.10'
|
27
27
|
spec.authors = ['Raphael Simon']
|
28
28
|
spec.email = 'raphael@rightscale.com'
|
29
29
|
spec.homepage = 'https://github.com/rightscale/right_scraper'
|
data/spec/scraper_spec.rb
CHANGED
@@ -22,10 +22,12 @@
|
|
22
22
|
#++
|
23
23
|
|
24
24
|
require File.join(File.dirname(__FILE__), 'spec_helper')
|
25
|
-
require File.join('
|
26
|
-
require File.join('scrapers', '
|
27
|
-
require File.join('scrapers', '
|
28
|
-
require '
|
25
|
+
require File.join(File.dirname(__FILE__), '..', 'lib', 'right_scraper', 'scraper_base')
|
26
|
+
require File.join(File.dirname(__FILE__), '..', 'lib', 'right_scraper', 'scrapers', 'git_scraper')
|
27
|
+
require File.join(File.dirname(__FILE__), '..', 'lib', 'right_scraper', 'scrapers', 'svn_scraper')
|
28
|
+
require File.join(File.dirname(__FILE__), '..', 'lib', 'right_scraper', 'scrapers', 'download_scraper')
|
29
|
+
require File.join(File.dirname(__FILE__), '..', 'lib', 'right_scraper', 'repository')
|
30
|
+
require File.join(File.dirname(__FILE__), '..', 'lib', 'right_scraper', 'scraper')
|
29
31
|
|
30
32
|
describe RightScale::Scraper do
|
31
33
|
|
@@ -33,17 +35,17 @@ describe RightScale::Scraper do
|
|
33
35
|
@scraper = RightScale::Scraper.new('/tmp')
|
34
36
|
@mock_scraper = flexmock('MockScraper')
|
35
37
|
mock_scraper_klass = flexmock('MockScraperClass', :new => @mock_scraper)
|
36
|
-
RightScale::SCRAPERS.merge!(
|
38
|
+
RightScale::SCRAPERS.merge!('mock' => mock_scraper_klass)
|
37
39
|
end
|
38
40
|
|
39
41
|
after(:all) do
|
40
|
-
RightScale::SCRAPERS.delete(
|
42
|
+
RightScale::SCRAPERS.delete('mock')
|
41
43
|
end
|
42
44
|
|
43
45
|
it 'should scrape' do
|
44
46
|
repo = RightScale::Repository.new
|
45
47
|
repo.repo_type = :mock
|
46
|
-
@mock_scraper.should_receive(:scrape).with(repo, Proc).and_return(true)
|
48
|
+
@mock_scraper.should_receive(:scrape).with(repo, true, Proc).and_return(true)
|
47
49
|
@mock_scraper.should_receive(:succeeded?).and_return(true)
|
48
50
|
@mock_scraper.should_receive(:current_repo_dir).and_return('42')
|
49
51
|
@scraper.scrape(repo) { }.should be_true
|
@@ -51,14 +53,14 @@ describe RightScale::Scraper do
|
|
51
53
|
end
|
52
54
|
|
53
55
|
it 'should scrape from a hash' do
|
54
|
-
@mock_scraper.should_receive(:scrape).with(RightScale::Repository, Proc).and_return(true)
|
56
|
+
@mock_scraper.should_receive(:scrape).with(RightScale::Repository, true, Proc).and_return(true)
|
55
57
|
@mock_scraper.should_receive(:succeeded?).and_return(true)
|
56
58
|
@mock_scraper.should_receive(:current_repo_dir).and_return('42')
|
57
59
|
@scraper.scrape({:repo_type => :mock}) { }.should be_true
|
58
60
|
end
|
59
61
|
|
60
62
|
it 'should report failures' do
|
61
|
-
@mock_scraper.should_receive(:scrape).with(RightScale::Repository, Proc).and_return(true)
|
63
|
+
@mock_scraper.should_receive(:scrape).with(RightScale::Repository, true, Proc).and_return(true)
|
62
64
|
@mock_scraper.should_receive(:succeeded?).and_return(false)
|
63
65
|
@mock_scraper.should_receive(:current_repo_dir).and_return('42')
|
64
66
|
@scraper.scrape({:repo_type => :mock}) { }.should be_false
|
data/spec/svn_scraper_spec.rb
CHANGED
data/spec/watcher_spec.rb
CHANGED
@@ -45,7 +45,7 @@ describe RightScale::Watcher do
|
|
45
45
|
|
46
46
|
it 'should report timeouts' do
|
47
47
|
watcher = RightScale::Watcher.new(max_bytes=1, max_seconds=2)
|
48
|
-
status = watcher.launch_and_watch('ruby -e "STDOUT.sync = true; puts 42; sleep 5"', @dest_dir)
|
48
|
+
status = watcher.launch_and_watch('ruby -e "STDOUT.sync = true; puts 42; sleep 5" 2>&1', @dest_dir)
|
49
49
|
status.status.should == :timeout
|
50
50
|
status.exit_code.should == -1
|
51
51
|
status.output.should == "42\n"
|
metadata
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: right_scraper
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 1.0.
|
4
|
+
version: 1.0.10
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Raphael Simon
|
@@ -9,7 +9,7 @@ autorequire:
|
|
9
9
|
bindir: bin
|
10
10
|
cert_chain: []
|
11
11
|
|
12
|
-
date: 2010-02-
|
12
|
+
date: 2010-02-15 00:00:00 -08:00
|
13
13
|
default_executable:
|
14
14
|
dependencies: []
|
15
15
|
|
@@ -26,6 +26,7 @@ files:
|
|
26
26
|
- README.rdoc
|
27
27
|
- Rakefile
|
28
28
|
- lib/right_scraper.rb
|
29
|
+
- lib/right_scraper/linux/process_monitor.rb
|
29
30
|
- lib/right_scraper/repository.rb
|
30
31
|
- lib/right_scraper/scraper.rb
|
31
32
|
- lib/right_scraper/scraper_base.rb
|
@@ -33,6 +34,7 @@ files:
|
|
33
34
|
- lib/right_scraper/scrapers/git_scraper.rb
|
34
35
|
- lib/right_scraper/scrapers/svn_scraper.rb
|
35
36
|
- lib/right_scraper/watcher.rb
|
37
|
+
- lib/right_scraper/win32/process_monitor.rb
|
36
38
|
- right_scraper.gemspec
|
37
39
|
- spec/download_scraper_spec.rb
|
38
40
|
- spec/git_scraper_spec.rb
|