s3_meta_sync 0.13.1 → 0.15.2
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +5 -5
- data/lib/s3_meta_sync.rb +2 -0
- data/lib/s3_meta_sync/syncer.rb +90 -25
- data/lib/s3_meta_sync/version.rb +3 -1
- data/lib/s3_meta_sync/zip.rb +3 -1
- metadata +20 -7
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
|
-
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
2
|
+
SHA256:
|
3
|
+
metadata.gz: eb3dcbb8679154cb0863f3c13ba43d5c828220cbc2f0624346c19cf3617dd358
|
4
|
+
data.tar.gz: 6973f43e1a0bad829be2748ec65e447a2f7e96c16b2d95746d3bfad1e8daa0ea
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: e6aca9b4086bf08b5bb18ea74622fd4ad45fa783440246499459e5d8663b331ab3b63cef7ce2c78a8eb66eb7ae6470e5fffd2beb43ab95363d38f96bf60128b5
|
7
|
+
data.tar.gz: d8c6e284debf19ac52094a3ec1e36cb158b318b3c655fbc5a63a9c65ebc236048c24af184327a8085acfe11b082395a53a3e199e654e63cb0ca6195b455e4a8c
|
data/lib/s3_meta_sync.rb
CHANGED
data/lib/s3_meta_sync/syncer.rb
CHANGED
@@ -1,3 +1,5 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
1
3
|
require "net/http"
|
2
4
|
require "open-uri"
|
3
5
|
require "yaml"
|
@@ -5,17 +7,24 @@ require "digest/md5"
|
|
5
7
|
require "fileutils"
|
6
8
|
require "tmpdir"
|
7
9
|
require "openssl"
|
10
|
+
require "mime/types"
|
8
11
|
|
9
|
-
require "aws-sdk-
|
12
|
+
require "aws-sdk-s3"
|
10
13
|
require "s3_meta_sync/zip"
|
11
14
|
|
12
15
|
module S3MetaSync
|
13
16
|
class Syncer
|
14
|
-
DEFAULT_REGION =
|
17
|
+
DEFAULT_REGION = "us-east-1"
|
15
18
|
STAGING_AREA_PREFIX = "s3ms_"
|
16
19
|
|
20
|
+
AWS_PUBLIC_ACCESS = "public-read"
|
21
|
+
AWS_PRIVATE_ACCESS = "private"
|
22
|
+
|
17
23
|
def initialize(config)
|
18
|
-
@config =
|
24
|
+
@config = {
|
25
|
+
acl: AWS_PUBLIC_ACCESS,
|
26
|
+
region: DEFAULT_REGION
|
27
|
+
}.merge(config)
|
19
28
|
end
|
20
29
|
|
21
30
|
def sync(source, destination)
|
@@ -93,7 +102,7 @@ module S3MetaSync
|
|
93
102
|
# Sometimes SIGTERM causes Dir.mktmpdir to not properly delete the temp folder
|
94
103
|
# Remove 1 day old folders
|
95
104
|
def delete_old_temp_folders
|
96
|
-
path = File.join(Dir.tmpdir, STAGING_AREA_PREFIX +
|
105
|
+
path = File.join(Dir.tmpdir, STAGING_AREA_PREFIX + "*")
|
97
106
|
|
98
107
|
day = 24 * 60 * 60
|
99
108
|
dirs = Dir.glob(path)
|
@@ -161,10 +170,12 @@ module S3MetaSync
|
|
161
170
|
content = Zip.zip(content) if @config[:zip] && path != META_FILE
|
162
171
|
|
163
172
|
object = {
|
173
|
+
acl: @config[:acl],
|
164
174
|
bucket: @bucket,
|
165
175
|
body: content,
|
166
|
-
|
167
|
-
|
176
|
+
content_encoding: content.encoding.to_s,
|
177
|
+
content_type: MIME::Types.of(path).first.to_s,
|
178
|
+
key: "#{destination}/#{path}"
|
168
179
|
}
|
169
180
|
|
170
181
|
object[:server_side_encryption] = @config[:server_side_encryption] if @config[:server_side_encryption]
|
@@ -195,11 +206,19 @@ module S3MetaSync
|
|
195
206
|
end
|
196
207
|
|
197
208
|
def s3
|
198
|
-
@s3 ||=
|
199
|
-
|
200
|
-
|
201
|
-
|
202
|
-
|
209
|
+
@s3 ||= begin
|
210
|
+
config = { region: @config[:region] }
|
211
|
+
|
212
|
+
if @config[:credentials_path]
|
213
|
+
config[:credentials] = Aws::SharedCredentials.new(path: @config[:credentials_path], profile_name: "default")
|
214
|
+
else
|
215
|
+
config[:access_key_id] = @config[:key]
|
216
|
+
config[:secret_access_key] = @config[:secret]
|
217
|
+
config[:session_token] = @config[:session_token] if @config[:session_token]
|
218
|
+
end
|
219
|
+
|
220
|
+
Aws::S3::Client.new(config)
|
221
|
+
end
|
203
222
|
end
|
204
223
|
|
205
224
|
def generate_meta(source)
|
@@ -230,21 +249,50 @@ module S3MetaSync
|
|
230
249
|
|
231
250
|
def read_meta(source)
|
232
251
|
file = "#{source}/#{META_FILE}"
|
233
|
-
|
252
|
+
if File.exist?(file)
|
253
|
+
content = File.read(file)
|
254
|
+
parse_yaml_content(content) if content.size > 0
|
255
|
+
end
|
234
256
|
end
|
235
257
|
|
236
258
|
def download_meta(destination)
|
259
|
+
if private?
|
260
|
+
private_access_download_meta(destination)
|
261
|
+
else
|
262
|
+
public_access_download_meta(destination)
|
263
|
+
end
|
264
|
+
end
|
265
|
+
|
266
|
+
def private_access_download_meta(destination)
|
267
|
+
content = private_content_download(destination, META_FILE).string
|
268
|
+
|
269
|
+
raise S3MetaSync::RemoteWithoutMeta if content.empty? # if missing, upload everything
|
270
|
+
|
271
|
+
parse_yaml_content(content)
|
272
|
+
rescue Aws::S3::Errors::NoSuchKey, Aws::S3::Errors::AccessDenied # if requesting a file that doesn't exist AccessDenied is raised
|
273
|
+
retries ||= 0
|
274
|
+
|
275
|
+
raise S3MetaSync::RemoteWithoutMeta if retries >= 1
|
276
|
+
|
277
|
+
retries += 1
|
278
|
+
sleep 1 # maybe the remote meta was just updated ... give aws a second chance ...
|
279
|
+
retry
|
280
|
+
end
|
281
|
+
|
282
|
+
def public_access_download_meta(destination)
|
237
283
|
content = download_content("#{destination}/#{META_FILE}") { |io| io.read }
|
284
|
+
|
285
|
+
raise OpenURI::HTTPError.new("Content is empty", nil) if content.size == 0
|
286
|
+
|
238
287
|
parse_yaml_content(content)
|
239
288
|
rescue OpenURI::HTTPError
|
240
289
|
retries ||= 0
|
290
|
+
|
291
|
+
raise S3MetaSync::RemoteWithoutMeta if retries >= 1
|
292
|
+
|
241
293
|
retries += 1
|
242
|
-
|
243
|
-
|
244
|
-
retry
|
245
|
-
else
|
246
|
-
raise RemoteWithoutMeta
|
247
|
-
end
|
294
|
+
sleep 1 # maybe the remote meta was just updated ... give aws a second chance ...
|
295
|
+
retry
|
248
296
|
end
|
249
297
|
|
250
298
|
def parse_yaml_content(content)
|
@@ -253,16 +301,30 @@ module S3MetaSync
|
|
253
301
|
end
|
254
302
|
|
255
303
|
def download_file(source, path, destination, zip)
|
256
|
-
download =
|
257
|
-
|
258
|
-
|
259
|
-
|
304
|
+
download = if private?
|
305
|
+
private_content_download(source, path)
|
306
|
+
else
|
307
|
+
public_content_download(source, path)
|
308
|
+
end
|
309
|
+
|
310
|
+
download = S3MetaSync::Zip.unzip(download) if zip
|
311
|
+
FileUtils.mkdir_p(File.dirname("#{destination}/#{path}"))
|
260
312
|
|
261
313
|
# consumes less ram then File.write(path, content), possibly also faster
|
262
|
-
File.open(path,
|
314
|
+
File.open("#{destination}/#{path}", "wb") { |f| IO.copy_stream(download, f) }
|
263
315
|
download.close
|
264
316
|
end
|
265
317
|
|
318
|
+
def private_content_download(source, path)
|
319
|
+
log "Downloading #{path}"
|
320
|
+
obj = s3.get_object(bucket: @bucket, key: "#{source}/#{path}")
|
321
|
+
obj.body
|
322
|
+
end
|
323
|
+
|
324
|
+
def public_content_download(source, path)
|
325
|
+
download_content("#{source}/#{path}") # warning: using block form consumes more ram
|
326
|
+
end
|
327
|
+
|
266
328
|
def download_content(path)
|
267
329
|
log "Downloading #{path}"
|
268
330
|
url =
|
@@ -287,8 +349,7 @@ module S3MetaSync
|
|
287
349
|
log "#{e.class} error downloading #{url}, retrying #{http_error_retries}/#{max_retries}"
|
288
350
|
retry
|
289
351
|
else
|
290
|
-
|
291
|
-
raise
|
352
|
+
raise $!, "#{$!.message} -- while trying to download #{url}", $!.backtrace
|
292
353
|
end
|
293
354
|
rescue OpenSSL::SSL::SSLError
|
294
355
|
ssl_error_retries ||= 0
|
@@ -340,5 +401,9 @@ module S3MetaSync
|
|
340
401
|
def log(text, important=false)
|
341
402
|
$stderr.puts text if @config[:verbose] or important
|
342
403
|
end
|
404
|
+
|
405
|
+
def private?
|
406
|
+
@config[:acl] == AWS_PRIVATE_ACCESS
|
407
|
+
end
|
343
408
|
end
|
344
409
|
end
|
data/lib/s3_meta_sync/version.rb
CHANGED
data/lib/s3_meta_sync/zip.rb
CHANGED
@@ -1,3 +1,5 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
1
3
|
require "zlib"
|
2
4
|
require "stringio"
|
3
5
|
|
@@ -5,7 +7,7 @@ module S3MetaSync
|
|
5
7
|
module Zip
|
6
8
|
class << self
|
7
9
|
def zip(string)
|
8
|
-
io = StringIO.new("w")
|
10
|
+
io = StringIO.new("w".dup)
|
9
11
|
w_gz = Zlib::GzipWriter.new(io)
|
10
12
|
w_gz.write(string)
|
11
13
|
w_gz.close
|
metadata
CHANGED
@@ -1,29 +1,43 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: s3_meta_sync
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.
|
4
|
+
version: 0.15.2
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Michael Grosser
|
8
8
|
autorequire:
|
9
9
|
bindir: bin
|
10
10
|
cert_chain: []
|
11
|
-
date:
|
11
|
+
date: 2020-09-09 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
|
-
name: aws-sdk-
|
14
|
+
name: aws-sdk-s3
|
15
15
|
requirement: !ruby/object:Gem::Requirement
|
16
16
|
requirements:
|
17
17
|
- - "~>"
|
18
18
|
- !ruby/object:Gem::Version
|
19
|
-
version: '
|
19
|
+
version: '1.0'
|
20
20
|
type: :runtime
|
21
21
|
prerelease: false
|
22
22
|
version_requirements: !ruby/object:Gem::Requirement
|
23
23
|
requirements:
|
24
24
|
- - "~>"
|
25
25
|
- !ruby/object:Gem::Version
|
26
|
-
version: '
|
26
|
+
version: '1.0'
|
27
|
+
- !ruby/object:Gem::Dependency
|
28
|
+
name: mime-types
|
29
|
+
requirement: !ruby/object:Gem::Requirement
|
30
|
+
requirements:
|
31
|
+
- - ">="
|
32
|
+
- !ruby/object:Gem::Version
|
33
|
+
version: '0'
|
34
|
+
type: :runtime
|
35
|
+
prerelease: false
|
36
|
+
version_requirements: !ruby/object:Gem::Requirement
|
37
|
+
requirements:
|
38
|
+
- - ">="
|
39
|
+
- !ruby/object:Gem::Version
|
40
|
+
version: '0'
|
27
41
|
description:
|
28
42
|
email: michael@grosser.it
|
29
43
|
executables:
|
@@ -55,8 +69,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
|
|
55
69
|
- !ruby/object:Gem::Version
|
56
70
|
version: '0'
|
57
71
|
requirements: []
|
58
|
-
|
59
|
-
rubygems_version: 2.4.5.1
|
72
|
+
rubygems_version: 3.1.4
|
60
73
|
signing_key:
|
61
74
|
specification_version: 4
|
62
75
|
summary: Sync folders with s3 using a metadata file and md5 diffs
|