crawl_kit 0.0.1 → 0.0.2
Sign up to get free protection for your applications and to get access to all the features.
- data/crawl_kit.gemspec +2 -2
- data/lib/crawl_kit/record/helpers.rb +2 -2
- data/lib/crawl_kit/s3.rb +1 -1
- metadata +5 -5
data/crawl_kit.gemspec
CHANGED
|
@@ -1,10 +1,10 @@
|
|
|
1
1
|
Gem::Specification.new do |s|
|
|
2
2
|
s.name = 'crawl_kit'
|
|
3
|
-
s.version = '0.0.
|
|
3
|
+
s.version = '0.0.2'
|
|
4
4
|
s.date = '2011-12-08'
|
|
5
5
|
s.summary = "A collection of serivce interfaces and models to use with the GoTime crawling infrastructure."
|
|
6
6
|
s.author = "Austin Cargol"
|
|
7
|
-
s.email = '
|
|
7
|
+
s.email = 'austin@gotime.com'
|
|
8
8
|
|
|
9
9
|
s.files = `git ls-files`.split("\n")
|
|
10
10
|
s.test_files = `git ls-files -- {test}/*`.split("\n")
|
|
@@ -17,7 +17,7 @@ module CrawlKit
|
|
|
17
17
|
end
|
|
18
18
|
|
|
19
19
|
def hashed_uri
|
|
20
|
-
self.class.
|
|
20
|
+
self.class.hash_uri(uri)
|
|
21
21
|
end
|
|
22
22
|
alias :crawl_id :hashed_uri
|
|
23
23
|
|
|
@@ -26,7 +26,7 @@ module CrawlKit
|
|
|
26
26
|
end
|
|
27
27
|
|
|
28
28
|
def page
|
|
29
|
-
CrawlKit::S3.get_file(crawl_id)
|
|
29
|
+
fetched? ? CrawlKit::S3.get_file(crawl_id) : nil
|
|
30
30
|
end
|
|
31
31
|
|
|
32
32
|
def previous_version
|
data/lib/crawl_kit/s3.rb
CHANGED
|
@@ -33,7 +33,7 @@ module CrawlKit
|
|
|
33
33
|
options.merge!(content_type: 'text/html', content_encoding: 'deflate', storage_class: :reduced_redundancy)
|
|
34
34
|
object = get_object(key)
|
|
35
35
|
object.write(Zlib::Deflate.deflate(body), options)
|
|
36
|
-
end
|
|
36
|
+
end
|
|
37
37
|
|
|
38
38
|
def get_versions(hashed_uri)
|
|
39
39
|
s3_object = get_object(hashed_uri)
|
metadata
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
|
2
2
|
name: crawl_kit
|
|
3
3
|
version: !ruby/object:Gem::Version
|
|
4
|
-
version: 0.0.
|
|
4
|
+
version: 0.0.2
|
|
5
5
|
prerelease:
|
|
6
6
|
platform: ruby
|
|
7
7
|
authors:
|
|
@@ -13,7 +13,7 @@ date: 2011-12-08 00:00:00.000000000Z
|
|
|
13
13
|
dependencies:
|
|
14
14
|
- !ruby/object:Gem::Dependency
|
|
15
15
|
name: aws-sdk
|
|
16
|
-
requirement: &
|
|
16
|
+
requirement: &70324126127720 !ruby/object:Gem::Requirement
|
|
17
17
|
none: false
|
|
18
18
|
requirements:
|
|
19
19
|
- - ! '>='
|
|
@@ -21,9 +21,9 @@ dependencies:
|
|
|
21
21
|
version: 1.2.3
|
|
22
22
|
type: :runtime
|
|
23
23
|
prerelease: false
|
|
24
|
-
version_requirements: *
|
|
24
|
+
version_requirements: *70324126127720
|
|
25
25
|
description:
|
|
26
|
-
email:
|
|
26
|
+
email: austin@gotime.com
|
|
27
27
|
executables: []
|
|
28
28
|
extensions: []
|
|
29
29
|
extra_rdoc_files: []
|
|
@@ -59,7 +59,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
|
|
|
59
59
|
version: '0'
|
|
60
60
|
requirements: []
|
|
61
61
|
rubyforge_project:
|
|
62
|
-
rubygems_version: 1.8.
|
|
62
|
+
rubygems_version: 1.8.15
|
|
63
63
|
signing_key:
|
|
64
64
|
specification_version: 3
|
|
65
65
|
summary: A collection of serivce interfaces and models to use with the GoTime crawling
|