faster_s3 1.0.0 → 1.1.0
Sign up to get free protection for your applications and to get access to all the features.
- data/faster_s3.gemspec +1 -1
- data/lib/faster_s3/download.rb +68 -0
- data/lib/faster_s3/part.rb +33 -0
- data/lib/faster_s3/version.rb +1 -1
- data/lib/faster_s3.rb +3 -65
- metadata +12 -10
data/faster_s3.gemspec
CHANGED
@@ -10,7 +10,7 @@ Gem::Specification.new do |spec|
|
|
10
10
|
spec.email = ["patrick@redbubble.com"]
|
11
11
|
spec.description = %q{Download files from s3 in parallel}
|
12
12
|
spec.summary = %q{Faster s3 downloads}
|
13
|
-
spec.homepage = ""
|
13
|
+
spec.homepage = "https://github.com/stiak/faster_s3"
|
14
14
|
spec.license = "MIT"
|
15
15
|
|
16
16
|
spec.files = `git ls-files`.split($/)
|
@@ -0,0 +1,68 @@
|
|
1
|
+
require 'aws/s3'
|
2
|
+
require 'parallel'
|
3
|
+
|
4
|
+
module FasterS3
|
5
|
+
class Download
|
6
|
+
DEFAULT_PARTS = 8
|
7
|
+
CHUNK_SIZE = 1024*1024
|
8
|
+
|
9
|
+
attr_reader :target_path, :config, :parts
|
10
|
+
|
11
|
+
def initialize(target_path, config = {})
|
12
|
+
@target_path = target_path
|
13
|
+
@config = config
|
14
|
+
@parts = config[:parallel] || DEFAULT_PARTS
|
15
|
+
end
|
16
|
+
|
17
|
+
def download
|
18
|
+
reconstitute(download_parts)
|
19
|
+
end
|
20
|
+
|
21
|
+
def download_parts
|
22
|
+
part_objects = create_parts
|
23
|
+
Parallel.map(part_objects, in_threads: part_objects.length) do |file_part|
|
24
|
+
file_part.download(s3_object)
|
25
|
+
file_part.part_path
|
26
|
+
end
|
27
|
+
end
|
28
|
+
|
29
|
+
def create_parts
|
30
|
+
length = s3_object.content_length
|
31
|
+
raise "File is too small to download in parallel" if length < parts
|
32
|
+
|
33
|
+
# Split into parts
|
34
|
+
part_length = length / parts
|
35
|
+
remainder = length % parts
|
36
|
+
|
37
|
+
part_objects = []
|
38
|
+
parts.times do |index|
|
39
|
+
# Last part includes any remaining bytes
|
40
|
+
extra_bytes = (index == (parts - 1)) ? remainder : 0
|
41
|
+
part_objects << Part.new(target_path, index, part_length, extra_bytes)
|
42
|
+
end
|
43
|
+
|
44
|
+
part_objects
|
45
|
+
end
|
46
|
+
|
47
|
+
def reconstitute(file_parts)
|
48
|
+
File.open(target_path, 'wb') do |file|
|
49
|
+
file_parts.each do |part_path|
|
50
|
+
File.open(part_path, 'rb') do |part_file|
|
51
|
+
file.write(part_file.read(CHUNK_SIZE)) until part_file.eof?
|
52
|
+
end
|
53
|
+
File.delete(part_path)
|
54
|
+
end
|
55
|
+
end
|
56
|
+
end
|
57
|
+
|
58
|
+
private
|
59
|
+
|
60
|
+
def s3_object
|
61
|
+
@s3_object ||= begin
|
62
|
+
s3 = AWS::S3.new({:access_key_id => config[:access_key_id], :secret_access_key => config[:secret_access_key]})
|
63
|
+
bucket = s3.buckets[config[:bucket_name]]
|
64
|
+
bucket.objects[config[:path]]
|
65
|
+
end
|
66
|
+
end
|
67
|
+
end
|
68
|
+
end
|
@@ -0,0 +1,33 @@
|
|
1
|
+
class FasterS3::Part
|
2
|
+
attr_accessor :file_path, :index, :part_length, :extra_bytes
|
3
|
+
|
4
|
+
def initialize(file_path, index, part_length, extra_bytes)
|
5
|
+
self.file_path = file_path
|
6
|
+
self.index = index
|
7
|
+
self.part_length = part_length
|
8
|
+
self.extra_bytes = extra_bytes
|
9
|
+
end
|
10
|
+
|
11
|
+
def part_path
|
12
|
+
"#{file_path}.part.#{index}"
|
13
|
+
end
|
14
|
+
|
15
|
+
def byte_range
|
16
|
+
start = index * part_length
|
17
|
+
end_pos = start + part_length + extra_bytes
|
18
|
+
|
19
|
+
(start + existing_size)...end_pos
|
20
|
+
end
|
21
|
+
|
22
|
+
def existing_size
|
23
|
+
File.exists?(part_path) ? File.size(part_path) : 0
|
24
|
+
end
|
25
|
+
|
26
|
+
def download(s3_object)
|
27
|
+
File.open(part_path, 'ab') do |file|
|
28
|
+
s3_object.read(range: byte_range) do |chunk|
|
29
|
+
file.write(chunk)
|
30
|
+
end
|
31
|
+
end
|
32
|
+
end
|
33
|
+
end
|
data/lib/faster_s3/version.rb
CHANGED
data/lib/faster_s3.rb
CHANGED
@@ -1,6 +1,6 @@
|
|
1
1
|
require "faster_s3/version"
|
2
|
-
require
|
3
|
-
require
|
2
|
+
require "faster_s3/part"
|
3
|
+
require "faster_s3/download"
|
4
4
|
|
5
5
|
module FasterS3
|
6
6
|
# @param [String] destination File will be written to this destination
|
@@ -17,69 +17,7 @@ module FasterS3
|
|
17
17
|
# @option options [String] :parallel Optional - number of parallel parts to download. Default is 8
|
18
18
|
#
|
19
19
|
def self.download(destination, options)
|
20
|
-
Download.new(options).
|
21
|
-
end
|
22
|
-
|
23
|
-
class Download
|
24
|
-
DEFAULT_PARTS = 8
|
25
|
-
CHUNK_SIZE = 1024*1024
|
26
|
-
|
27
|
-
attr_reader :config, :parts
|
28
|
-
|
29
|
-
def initialize(config = {})
|
30
|
-
@config = config
|
31
|
-
@parts = config[:parallel] || DEFAULT_PARTS
|
32
|
-
end
|
33
|
-
|
34
|
-
def download_to(path)
|
35
|
-
s3_object = remote_object
|
36
|
-
length = s3_object.content_length
|
37
|
-
raise "File is too small to download in parallel" if length < parts
|
38
|
-
|
39
|
-
# Split into parts
|
40
|
-
part_length = length / parts
|
41
|
-
remainder = length % parts
|
42
|
-
|
43
|
-
part_array = Array(1..parts)
|
44
|
-
file_parts = Parallel.map(part_array, in_threads: parts) do |part|
|
45
|
-
start = (part - 1) * part_length
|
46
|
-
end_pos = start + part_length
|
47
|
-
|
48
|
-
if part == part_array.last
|
49
|
-
end_pos += remainder
|
50
|
-
byte_range = start..end_pos
|
51
|
-
else
|
52
|
-
byte_range = start...end_pos
|
53
|
-
end
|
54
|
-
|
55
|
-
part_path = "#{path}.part.#{part}"
|
56
|
-
File.open("#{path}.part.#{part}", 'wb') do |file|
|
57
|
-
s3_object.read(range: byte_range) do |chunk|
|
58
|
-
file.write(chunk)
|
59
|
-
end
|
60
|
-
end
|
61
|
-
part_path
|
62
|
-
end
|
63
|
-
|
64
|
-
reconstitute_parts(path, file_parts)
|
65
|
-
end
|
66
|
-
|
67
|
-
def reconstitute_parts(path, file_parts)
|
68
|
-
File.open(path, 'wb') do |file|
|
69
|
-
file_parts.each do |part_path|
|
70
|
-
File.open(part_path, 'rb') do |part_file|
|
71
|
-
file.write(part_file.read(CHUNK_SIZE)) until part_file.eof?
|
72
|
-
end
|
73
|
-
File.delete(part_path)
|
74
|
-
end
|
75
|
-
end
|
76
|
-
end
|
77
|
-
|
78
|
-
def remote_object
|
79
|
-
s3 = AWS::S3.new({:access_key_id => config[:access_key_id], :secret_access_key => config[:secret_access_key]})
|
80
|
-
bucket = s3.buckets[config[:bucket_name]]
|
81
|
-
bucket.objects[config[:path]]
|
82
|
-
end
|
20
|
+
Download.new(destination, options).download
|
83
21
|
end
|
84
22
|
|
85
23
|
end
|
metadata
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: faster_s3
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 1.
|
4
|
+
version: 1.1.0
|
5
5
|
prerelease:
|
6
6
|
platform: ruby
|
7
7
|
authors:
|
@@ -13,7 +13,7 @@ date: 2013-11-07 00:00:00.000000000 Z
|
|
13
13
|
dependencies:
|
14
14
|
- !ruby/object:Gem::Dependency
|
15
15
|
name: bundler
|
16
|
-
requirement: &
|
16
|
+
requirement: &70342546745840 !ruby/object:Gem::Requirement
|
17
17
|
none: false
|
18
18
|
requirements:
|
19
19
|
- - ~>
|
@@ -21,10 +21,10 @@ dependencies:
|
|
21
21
|
version: '1.3'
|
22
22
|
type: :development
|
23
23
|
prerelease: false
|
24
|
-
version_requirements: *
|
24
|
+
version_requirements: *70342546745840
|
25
25
|
- !ruby/object:Gem::Dependency
|
26
26
|
name: rake
|
27
|
-
requirement: &
|
27
|
+
requirement: &70342546745360 !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: :development
|
34
34
|
prerelease: false
|
35
|
-
version_requirements: *
|
35
|
+
version_requirements: *70342546745360
|
36
36
|
- !ruby/object:Gem::Dependency
|
37
37
|
name: aws-sdk
|
38
|
-
requirement: &
|
38
|
+
requirement: &70342546744860 !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: *70342546744860
|
47
47
|
- !ruby/object:Gem::Dependency
|
48
48
|
name: parallel
|
49
|
-
requirement: &
|
49
|
+
requirement: &70342546744400 !ruby/object:Gem::Requirement
|
50
50
|
none: false
|
51
51
|
requirements:
|
52
52
|
- - ! '>='
|
@@ -54,7 +54,7 @@ dependencies:
|
|
54
54
|
version: '0'
|
55
55
|
type: :runtime
|
56
56
|
prerelease: false
|
57
|
-
version_requirements: *
|
57
|
+
version_requirements: *70342546744400
|
58
58
|
description: Download files from s3 in parallel
|
59
59
|
email:
|
60
60
|
- patrick@redbubble.com
|
@@ -71,8 +71,10 @@ files:
|
|
71
71
|
- Rakefile
|
72
72
|
- faster_s3.gemspec
|
73
73
|
- lib/faster_s3.rb
|
74
|
+
- lib/faster_s3/download.rb
|
75
|
+
- lib/faster_s3/part.rb
|
74
76
|
- lib/faster_s3/version.rb
|
75
|
-
homepage:
|
77
|
+
homepage: https://github.com/stiak/faster_s3
|
76
78
|
licenses:
|
77
79
|
- MIT
|
78
80
|
post_install_message:
|