right_scraper 5.0.1 → 5.1.1
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.rdoc +2 -2
- data/bin/right_scraper-retrieve +76 -0
- data/bin/right_scraper-scan +82 -0
- data/lib/right_scraper.rb +3 -0
- data/lib/right_scraper/main.rb +104 -55
- data/lib/right_scraper/processes.rb +0 -1
- data/lib/right_scraper/processes/shell.rb +2 -2
- data/lib/right_scraper/processes/ssh_agent.rb +1 -1
- data/lib/right_scraper/repositories/base.rb +1 -0
- data/lib/right_scraper/resources/cookbook.rb +18 -0
- data/lib/right_scraper/retrievers/download.rb +1 -1
- data/lib/right_scraper/scanners.rb +2 -0
- data/lib/right_scraper/scanners/cookbook_filename_scanner.rb +83 -0
- data/lib/right_scraper/scanners/cookbook_metadata.rb +134 -148
- data/lib/right_scraper/scanners/{workflow_metadata.rb → cookbook_metadata_readonly.rb} +32 -31
- data/lib/right_scraper/scrapers/base.rb +1 -1
- data/lib/right_scraper/version.rb +2 -2
- data/right_scraper.gemspec +2 -1
- data/{lib/right_scraper/resources/workflow.rb → scripts/knife_metadata.rb} +17 -29
- data/scripts/stub_ssh_askpass +3 -0
- metadata +11 -9
- data/lib/right_scraper/processes/warden.rb +0 -358
- data/lib/right_scraper/scanners/workflow_manifest.rb +0 -87
- data/lib/right_scraper/scanners/workflow_s3_upload.rb +0 -89
- data/lib/right_scraper/scrapers/workflow.rb +0 -94
@@ -1,87 +0,0 @@
|
|
1
|
-
#--
|
2
|
-
# Copyright: Copyright (c) 2010-2013 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
|
-
# ancestor
|
25
|
-
require 'right_scraper/scanners'
|
26
|
-
|
27
|
-
require 'digest/sha1'
|
28
|
-
|
29
|
-
module RightScraper::Scanners
|
30
|
-
|
31
|
-
# Build manifests from a filesystem.
|
32
|
-
class WorkflowManifest < ::RightScraper::Scanners::Base
|
33
|
-
# Create a new manifest scanner. Does not accept any new arguments.
|
34
|
-
def initialize(*args)
|
35
|
-
super
|
36
|
-
@manifest = {}
|
37
|
-
end
|
38
|
-
|
39
|
-
# Retrieve relative workflow files positions
|
40
|
-
#
|
41
|
-
# === Parameters
|
42
|
-
# workflow(Resources::Workflow):: Workflow whose manifest is being built
|
43
|
-
def begin(workflow)
|
44
|
-
@workflow = workflow
|
45
|
-
@metadata_filename = File.basename(@workflow.metadata_path)
|
46
|
-
@definition_filename = File.basename(@workflow.definition_path)
|
47
|
-
end
|
48
|
-
|
49
|
-
# Complete a scan for the given resource.
|
50
|
-
#
|
51
|
-
# === Parameters ===
|
52
|
-
# resource(RightScraper::Resources::Base):: resource to scan
|
53
|
-
def end(resource)
|
54
|
-
resource.manifest = @manifest
|
55
|
-
@manifest = {}
|
56
|
-
end
|
57
|
-
|
58
|
-
# Notice a file during scanning.
|
59
|
-
#
|
60
|
-
# === Block ===
|
61
|
-
# Return the data for this file. We use a block because it may
|
62
|
-
# not always be necessary to read the data.
|
63
|
-
#
|
64
|
-
# === Parameters ===
|
65
|
-
# relative_position(String):: relative pathname for file from root of resource
|
66
|
-
def notice(relative_position)
|
67
|
-
if [ @metadata_filename, @definition_filename ].include?(relative_position)
|
68
|
-
@manifest[relative_position] = Digest::SHA1.hexdigest(yield)
|
69
|
-
end
|
70
|
-
end
|
71
|
-
|
72
|
-
# Notice a directory during scanning. Since the workflow definition and
|
73
|
-
# metadata live in the root directory we don't need to recurse,
|
74
|
-
# but we do need to go into the first directory (identified by
|
75
|
-
# +relative_position+ being +nil+).
|
76
|
-
#
|
77
|
-
# === Parameters
|
78
|
-
# relative_position(String):: relative pathname for the directory from root of workflow
|
79
|
-
#
|
80
|
-
# === Returns
|
81
|
-
# Boolean:: should the scanning recurse into the directory
|
82
|
-
def notice_dir(relative_position)
|
83
|
-
relative_position == nil
|
84
|
-
end
|
85
|
-
|
86
|
-
end
|
87
|
-
end
|
@@ -1,89 +0,0 @@
|
|
1
|
-
#--
|
2
|
-
# Copyright: Copyright (c) 2010-2013 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
|
-
# ancestor
|
25
|
-
require 'right_scraper/scanners'
|
26
|
-
|
27
|
-
require 'right_aws'
|
28
|
-
require 'json'
|
29
|
-
|
30
|
-
module RightScraper::Scanners
|
31
|
-
|
32
|
-
# Upload workflow definition and metadata to an S3 bucket.
|
33
|
-
class WorkflowS3Upload < ::RightScraper::Scanners::Base
|
34
|
-
|
35
|
-
# Create a new S3Upload. In addition to the options recognized
|
36
|
-
# by Scanner, this class recognizes <tt>:s3_key</tt>,
|
37
|
-
# <tt>:s3_secret</tt>, and <tt>:s3_bucket</tt> and requires all
|
38
|
-
# of those.
|
39
|
-
#
|
40
|
-
# === Options
|
41
|
-
# <tt>:s3_key</tt>:: Required. S3 access key.
|
42
|
-
# <tt>:s3_secret</tt>:: Required. S3 secret key.
|
43
|
-
# <tt>:s3_bucket</tt>:: Required. Bucket to upload workflows to.
|
44
|
-
#
|
45
|
-
# === Parameters
|
46
|
-
# options(Hash):: scanner options
|
47
|
-
def initialize(options={})
|
48
|
-
super
|
49
|
-
s3_key = options.fetch(:s3_key)
|
50
|
-
s3_secret = options.fetch(:s3_secret)
|
51
|
-
s3 = RightAws::S3.new(aws_access_key_id=s3_key,
|
52
|
-
aws_secret_access_key=s3_secret,
|
53
|
-
:logger => @logger)
|
54
|
-
@bucket = s3.bucket(options.fetch(:s3_bucket))
|
55
|
-
raise "Need an actual, existing S3 bucket!" if @bucket.nil?
|
56
|
-
end
|
57
|
-
|
58
|
-
# Upon ending a scan for a workflows, upload the workflows
|
59
|
-
# contents to S3.
|
60
|
-
#
|
61
|
-
# === Parameters
|
62
|
-
# workflows(RightScraper::Workflows):: Workflow to scan
|
63
|
-
def end(workflow)
|
64
|
-
@bucket.put(File.join('Workflows', workflow.resource_hash),
|
65
|
-
{
|
66
|
-
:metadata => workflow.metadata,
|
67
|
-
:manifest => workflow.manifest
|
68
|
-
}.to_json)
|
69
|
-
end
|
70
|
-
|
71
|
-
# Upload a file during scanning.
|
72
|
-
#
|
73
|
-
# === Block
|
74
|
-
# Return the data for this file. We use a block because it may
|
75
|
-
# not always be necessary to read the data.
|
76
|
-
#
|
77
|
-
# === Parameters
|
78
|
-
# relative_position(String):: relative pathname for file from root of cookbook
|
79
|
-
def notice(relative_position)
|
80
|
-
# TBD: Only uplad definition and metadata, will there be more files?
|
81
|
-
contents = yield
|
82
|
-
name = Digest::SHA1.hexdigest(contents)
|
83
|
-
path = File.join('Files', name)
|
84
|
-
unless @bucket.key(path).exists?
|
85
|
-
@bucket.put(path, contents)
|
86
|
-
end
|
87
|
-
end
|
88
|
-
end
|
89
|
-
end
|
@@ -1,94 +0,0 @@
|
|
1
|
-
#--
|
2
|
-
# Copyright: Copyright (c) 2010-2013 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 File.expand_path(File.join(File.dirname(__FILE__), 'base'))
|
25
|
-
|
26
|
-
# ancestor
|
27
|
-
require 'right_scraper/scrapers'
|
28
|
-
|
29
|
-
module RightScraper::Scrapers
|
30
|
-
|
31
|
-
# Workflow scraper
|
32
|
-
class Workflow < ::RightScraper::Scrapers::Base
|
33
|
-
|
34
|
-
# Initialize list of known workflows on top of
|
35
|
-
def initialize(options)
|
36
|
-
@known_workflows = []
|
37
|
-
super(options)
|
38
|
-
end
|
39
|
-
|
40
|
-
# Find the next workflows, starting in dir.
|
41
|
-
#
|
42
|
-
# === Parameters
|
43
|
-
# dir(Dir):: directory to begin search in
|
44
|
-
def find_next(dir)
|
45
|
-
@logger.operation(:finding_next_workflow, "in #{dir.path}") do
|
46
|
-
|
47
|
-
# Note: there could be multiple workflow definitions in one directory
|
48
|
-
# so we need to record the current position whether we found a workflow
|
49
|
-
# or not. The next iteration will search again in the current directory
|
50
|
-
# event if we found one. If we don't find one then we call
|
51
|
-
# 'search_dirs' which will recurse in the sub-directories.
|
52
|
-
@stack << dir
|
53
|
-
|
54
|
-
def_ext = RightScraper::Resources::Workflow::DEFINITION_EXT
|
55
|
-
meta_ext = RightScraper::Resources::Workflow::METADATA_EXT
|
56
|
-
potentials = Dir[File.join(dir.path, "*#{def_ext}")]
|
57
|
-
potentials.reject! { |wdef| !File.exists?(wdef.chomp(File.extname(wdef)) + meta_ext) }
|
58
|
-
potentials.reject! { |wdef| @known_workflows.include?(wdef) }
|
59
|
-
unless potentials.empty?
|
60
|
-
wdef = potentials.first
|
61
|
-
relative_def = strip_repo_dir(wdef)
|
62
|
-
@logger.operation(:reading_workflow, "#{relative_def}") do
|
63
|
-
workflow = RightScraper::Resources::Workflow.new(@repository, relative_def)
|
64
|
-
@builder.go(File.dirname(wdef), workflow)
|
65
|
-
@known_workflows << wdef
|
66
|
-
workflow
|
67
|
-
end
|
68
|
-
else
|
69
|
-
search_dirs
|
70
|
-
end
|
71
|
-
end
|
72
|
-
end
|
73
|
-
|
74
|
-
# List of default scanners for this scaper
|
75
|
-
#
|
76
|
-
# === Return
|
77
|
-
# Array<Scanner>:: Default scanners
|
78
|
-
def default_scanners
|
79
|
-
[RightScraper::Scanners::WorkflowMetadata,
|
80
|
-
RightScraper::Scanners::WorkflowManifest]
|
81
|
-
end
|
82
|
-
|
83
|
-
# List of default builders for this scaper
|
84
|
-
#
|
85
|
-
# === Return
|
86
|
-
# Array<Builder>:: Default builders
|
87
|
-
def default_builders
|
88
|
-
[RightScraper::Builders::Filesystem]
|
89
|
-
end
|
90
|
-
|
91
|
-
# self-register
|
92
|
-
register_self
|
93
|
-
end
|
94
|
-
end
|