athena-udf 0.1.0 → 0.1.1
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/.rubocop.yml +3 -0
- data/Gemfile +3 -4
- data/Gemfile.lock +10 -0
- data/README.md +7 -0
- data/lib/athena-udf/base_udf.rb +9 -0
- data/lib/athena-udf/version.rb +1 -1
- metadata +2 -2
checksums.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA256:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: 1ddf95a3316e3f1c9f3a6b21d067c4412aef09cc34982a5be141428c64974674
|
|
4
|
+
data.tar.gz: a9a189b6a6a1d2651f96a198f75829b22cd1289bfd41f3a2543ec408531ae064
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: 83ce24ec41d03b9cf909c1a1f7b34cf952a79af775eabc00b2769cb6f862fd44c255a74961b9a99207f8581789b05f8202a3b09fb7b03f3785dde385c90f78d1
|
|
7
|
+
data.tar.gz: 16a47b3b3504e204c65bd71590a8c63618ad5fd25f0453504874f91ba1aed92624a77b3e660ece6815eb31d096cfab6fcfc680e2bbd6ff5943e235a2c24c838b
|
data/.rubocop.yml
CHANGED
data/Gemfile
CHANGED
|
@@ -5,11 +5,10 @@ source 'https://rubygems.org'
|
|
|
5
5
|
gemspec
|
|
6
6
|
|
|
7
7
|
group :development, :test do
|
|
8
|
+
gem 'gem-release', '~> 2.2'
|
|
8
9
|
gem 'rake', '~> 13.0'
|
|
9
|
-
|
|
10
10
|
gem 'rspec', '~> 3.0'
|
|
11
|
-
|
|
12
11
|
gem 'rubocop', '~> 1.21'
|
|
13
|
-
|
|
14
|
-
gem '
|
|
12
|
+
gem 'simplecov', '~> 0.22'
|
|
13
|
+
gem 'simplecov-lcov', '~> 0.8.0'
|
|
15
14
|
end
|
data/Gemfile.lock
CHANGED
|
@@ -14,6 +14,7 @@ GEM
|
|
|
14
14
|
bigdecimal (3.1.8)
|
|
15
15
|
csv (3.3.0)
|
|
16
16
|
diff-lcs (1.5.1)
|
|
17
|
+
docile (1.4.1)
|
|
17
18
|
extpp (0.1.1)
|
|
18
19
|
fiddle (1.1.2)
|
|
19
20
|
gem-release (2.2.2)
|
|
@@ -72,6 +73,13 @@ GEM
|
|
|
72
73
|
rubocop-ast (1.32.1)
|
|
73
74
|
parser (>= 3.3.1.0)
|
|
74
75
|
ruby-progressbar (1.13.0)
|
|
76
|
+
simplecov (0.22.0)
|
|
77
|
+
docile (~> 1.1)
|
|
78
|
+
simplecov-html (~> 0.11)
|
|
79
|
+
simplecov_json_formatter (~> 0.1)
|
|
80
|
+
simplecov-html (0.12.3)
|
|
81
|
+
simplecov-lcov (0.8.0)
|
|
82
|
+
simplecov_json_formatter (0.1.4)
|
|
75
83
|
strscan (3.1.0)
|
|
76
84
|
unicode-display_width (2.5.0)
|
|
77
85
|
|
|
@@ -85,6 +93,8 @@ DEPENDENCIES
|
|
|
85
93
|
rake (~> 13.0)
|
|
86
94
|
rspec (~> 3.0)
|
|
87
95
|
rubocop (~> 1.21)
|
|
96
|
+
simplecov (~> 0.22)
|
|
97
|
+
simplecov-lcov (~> 0.8.0)
|
|
88
98
|
|
|
89
99
|
BUNDLED WITH
|
|
90
100
|
2.5.17
|
data/README.md
CHANGED
|
@@ -1,5 +1,8 @@
|
|
|
1
1
|
# AthenaUDF
|
|
2
2
|
|
|
3
|
+
[![Gem Version][gem-image]][gem-link]
|
|
4
|
+
[![Coverage Status][cov-image]][cov-link]
|
|
5
|
+
|
|
3
6
|
Ruby-version Athena User Defined Function (UDF).
|
|
4
7
|
|
|
5
8
|
This gem is highly inspired by [the Python-version Athena UDF](https://github.com/dmarkey/python-athena-udf).
|
|
@@ -88,3 +91,7 @@ Bug reports and pull requests are welcome on GitHub at https://github.com/dtaniw
|
|
|
88
91
|
|
|
89
92
|
The gem is available as open source under the terms of the [MIT License](https://opensource.org/licenses/MIT).
|
|
90
93
|
|
|
94
|
+
[gem-image]: https://badge.fury.io/rb/athena-udf.svg
|
|
95
|
+
[gem-link]: http://badge.fury.io/rb/athena-udf
|
|
96
|
+
[cov-image]: https://coveralls.io/repos/dtaniwaki/ruby-athena-udf/badge.png
|
|
97
|
+
[cov-link]: https://coveralls.io/r/dtaniwaki/ruby-athena-udf
|
data/lib/athena-udf/base_udf.rb
CHANGED
|
@@ -4,12 +4,16 @@ require 'securerandom'
|
|
|
4
4
|
require 'base64'
|
|
5
5
|
require 'tempfile'
|
|
6
6
|
require 'arrow'
|
|
7
|
+
require 'logger'
|
|
7
8
|
require_relative 'utils'
|
|
8
9
|
|
|
9
10
|
module AthenaUDF
|
|
10
11
|
class BaseUDF
|
|
11
12
|
extend AthenaUDF::Utils
|
|
12
13
|
|
|
14
|
+
@@logger = Logger.new($stdout)
|
|
15
|
+
@@logger.level = Logger.const_get(ENV.fetch('LOG_LEVEL', 'WARN').upcase)
|
|
16
|
+
|
|
13
17
|
def self.lambda_handler(event:, context:) # rubocop:disable Lint/UnusedMethodArgument
|
|
14
18
|
incoming_type = event['@type']
|
|
15
19
|
if incoming_type == 'PingRequest'
|
|
@@ -40,6 +44,7 @@ module AthenaUDF
|
|
|
40
44
|
input_schema_data = Base64.decode64(event['inputRecords']['schema'])
|
|
41
45
|
input_records_data = Base64.decode64(event['inputRecords']['records'])
|
|
42
46
|
read_record_batches(input_schema_data, input_records_data) do |input_schema, record_batch|
|
|
47
|
+
logger.info("Processing #{record_batch.size} records")
|
|
43
48
|
output_builder.append_records(
|
|
44
49
|
record_batch.each_record.map do |record|
|
|
45
50
|
handle_athena_record(input_schema, output_schema, record)
|
|
@@ -69,5 +74,9 @@ module AthenaUDF
|
|
|
69
74
|
def self.handle_athena_record(input_schema, output_schema, records)
|
|
70
75
|
raise NotImplementedError
|
|
71
76
|
end
|
|
77
|
+
|
|
78
|
+
def self.logger
|
|
79
|
+
@@logger
|
|
80
|
+
end
|
|
72
81
|
end
|
|
73
82
|
end
|
data/lib/athena-udf/version.rb
CHANGED
metadata
CHANGED
|
@@ -1,14 +1,14 @@
|
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
|
2
2
|
name: athena-udf
|
|
3
3
|
version: !ruby/object:Gem::Version
|
|
4
|
-
version: 0.1.
|
|
4
|
+
version: 0.1.1
|
|
5
5
|
platform: ruby
|
|
6
6
|
authors:
|
|
7
7
|
- Daisuke Taniwaki
|
|
8
8
|
autorequire:
|
|
9
9
|
bindir: exe
|
|
10
10
|
cert_chain: []
|
|
11
|
-
date: 2024-08-
|
|
11
|
+
date: 2024-08-27 00:00:00.000000000 Z
|
|
12
12
|
dependencies:
|
|
13
13
|
- !ruby/object:Gem::Dependency
|
|
14
14
|
name: base64
|