embulk-parser-query_string 0.0.2
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 +7 -0
- data/.gitignore +5 -0
- data/.travis.yml +12 -0
- data/CHANGELOG.md +13 -0
- data/Gemfile +2 -0
- data/LICENSE +13 -0
- data/LICENSE.txt +21 -0
- data/README.md +84 -0
- data/Rakefile +60 -0
- data/embulk-parser-query_string.gemspec +23 -0
- data/lib/embulk/guess/query_string.rb +39 -0
- data/lib/embulk/parser/query_string.rb +79 -0
- data/partial-config.yml +8 -0
- data/test/embulk/guess/test_query_string.rb +122 -0
- data/test/embulk/parser/test_query_string_plugin.rb +120 -0
- data/test/prepare_embulk.rb +16 -0
- data/test/run-test.rb +18 -0
- metadata +170 -0
checksums.yaml
ADDED
@@ -0,0 +1,7 @@
|
|
1
|
+
---
|
2
|
+
SHA1:
|
3
|
+
metadata.gz: 97fb67cc2b3cf2233e6e3fd248033c59edd58b70
|
4
|
+
data.tar.gz: 11879dcdad1b06b3ce746ce33c02816a8267f4c8
|
5
|
+
SHA512:
|
6
|
+
metadata.gz: f89539141e02dcb74fbf93956d77bb94446aa4b7dcf67c14b7bc1c6ad3862a162c2fde1a5965e49ceea278da8a60e59babcd8d7f7eeb7482b76bc38e8608b096
|
7
|
+
data.tar.gz: 8e1de420daf02fb16f477d9aaa2f3d97fa9dbb7fca7eb86692337b76989a2867e28c603e2c03dfd5feff0c42e89c9e9cfe1c5cc0ad8fc232d2b0b890151f670e
|
data/.travis.yml
ADDED
@@ -0,0 +1,12 @@
|
|
1
|
+
language: ruby
|
2
|
+
rvm:
|
3
|
+
- jruby-19mode
|
4
|
+
addons:
|
5
|
+
code_climate:
|
6
|
+
repo_token:
|
7
|
+
secure: "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"
|
8
|
+
notifications:
|
9
|
+
slack:
|
10
|
+
secure: "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"
|
11
|
+
jdk:
|
12
|
+
- oraclejdk8
|
data/CHANGELOG.md
ADDED
@@ -0,0 +1,13 @@
|
|
1
|
+
## 0.0.2 - 2015-07-08
|
2
|
+
|
3
|
+
The name of this plugin is changed to "embulk-parser-query_string" from "embulk-parser-query-string".
|
4
|
+
|
5
|
+
* [maintenance] Fix example config in README and sample config [#7](https://github.com/treasure-data/embulk-parser-query_string/pull/7)
|
6
|
+
* [fixed] Decode line correctly [#6](https://github.com/treasure-data/embulk-parser-query_string/pull/6)
|
7
|
+
* [fixed] fall back to guess csv [#5](https://github.com/treasure-data/embulk-parser-query_string/pull/5)
|
8
|
+
* [enhancement] Error handling for parser [#4](https://github.com/treasure-data/embulk-parser-query_string/pull/4)
|
9
|
+
* [maintenance] Use underscore for plugin name [#3](https://github.com/treasure-data/embulk-parser-query_string/pull/3)
|
10
|
+
|
11
|
+
## 0.0.1 - 2015-07-07
|
12
|
+
|
13
|
+
The first release!!
|
data/Gemfile
ADDED
data/LICENSE
ADDED
@@ -0,0 +1,13 @@
|
|
1
|
+
Copyright 2015 Everyleaf Corporation
|
2
|
+
|
3
|
+
Licensed under the Apache License, Version 2.0 (the "License");
|
4
|
+
you may not use this file except in compliance with the License.
|
5
|
+
You may obtain a copy of the License at
|
6
|
+
|
7
|
+
http://www.apache.org/licenses/LICENSE-2.0
|
8
|
+
|
9
|
+
Unless required by applicable law or agreed to in writing, software
|
10
|
+
distributed under the License is distributed on an "AS IS" BASIS,
|
11
|
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
12
|
+
See the License for the specific language governing permissions and
|
13
|
+
limitations under the License.
|
data/LICENSE.txt
ADDED
@@ -0,0 +1,21 @@
|
|
1
|
+
|
2
|
+
MIT License
|
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
|
18
|
+
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
|
19
|
+
LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
|
20
|
+
OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
|
21
|
+
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
data/README.md
ADDED
@@ -0,0 +1,84 @@
|
|
1
|
+
[](https://travis-ci.org/treasure-data/embulk-parser-query_string)
|
2
|
+
[](https://codeclimate.com/github/treasure-data/embulk-parser-query_string)
|
3
|
+
[](https://codeclimate.com/github/treasure-data/embulk-parser-query_string/coverage)
|
4
|
+
|
5
|
+
# Query String parser plugin for [Embulk](http://www.embulk.org)
|
6
|
+
|
7
|
+
Transform `key=value&key2=value2` line to `{key: "value", key2: "value2"}`. (HTTP Query String to Hash)
|
8
|
+
|
9
|
+
Currently, this plugin supports minimum case, some edge cases are unsupported as below.
|
10
|
+
|
11
|
+
- Duplicated key (e.g. `key=1&key=2`)
|
12
|
+
- Array parameter (e.g. `key[]=1&key[]=2`)
|
13
|
+
|
14
|
+
## Overview
|
15
|
+
|
16
|
+
* **Plugin type**: parser
|
17
|
+
* **Guess supported**: yes
|
18
|
+
|
19
|
+
## Configuration
|
20
|
+
|
21
|
+
- **strip_quote**: If you have quoted lines file such as `"foo=FOO&bar=BAR"`, should be true for strip their quotes. (bool, default: true)
|
22
|
+
- **strip_whitespace**: Strip whitespace before parsing lines for any indented line parse correctly such as ' foo=FOO'. (bool, default: true)
|
23
|
+
|
24
|
+
## Example
|
25
|
+
|
26
|
+
You have such text file (`target_file.txt`) as below:
|
27
|
+
|
28
|
+
```text
|
29
|
+
"user_id=42&some_param=ABC"
|
30
|
+
"user_id=43&some_param=EFG"
|
31
|
+
"user_id=44&some_param=XYZ"
|
32
|
+
```
|
33
|
+
|
34
|
+
And you have `partial-config.yml` as below:
|
35
|
+
|
36
|
+
```yaml
|
37
|
+
in:
|
38
|
+
type: file
|
39
|
+
path_prefix: ./target_file
|
40
|
+
parser:
|
41
|
+
strip_quote: true
|
42
|
+
strip_whitespace: true
|
43
|
+
exec: {}
|
44
|
+
out: {type: stdout}
|
45
|
+
```
|
46
|
+
|
47
|
+
Run `embulk guess`.
|
48
|
+
|
49
|
+
```
|
50
|
+
$ embulk guess -g query_string partial-config.yml -o guessed.yml
|
51
|
+
```
|
52
|
+
|
53
|
+
You got guessed.yml as below:
|
54
|
+
|
55
|
+
```yaml
|
56
|
+
in:
|
57
|
+
type: file
|
58
|
+
path_prefix: ./target_file
|
59
|
+
parser:
|
60
|
+
strip_quote: true
|
61
|
+
strip_whitespace: true
|
62
|
+
charset: ISO-8859-2
|
63
|
+
newline: CRLF
|
64
|
+
type: query_string
|
65
|
+
schema:
|
66
|
+
- {name: user_id, type: long}
|
67
|
+
- {name: some_param, type: string}
|
68
|
+
exec: {}
|
69
|
+
out: {type: stdout}
|
70
|
+
```
|
71
|
+
|
72
|
+
Finally, `embulk run` with generated guessed.yml.
|
73
|
+
|
74
|
+
```
|
75
|
+
$ embulk run guessed.yml
|
76
|
+
```
|
77
|
+
|
78
|
+
You can see the parsed records on STDOUT.
|
79
|
+
|
80
|
+
## Install plugin
|
81
|
+
|
82
|
+
```
|
83
|
+
$ embulk gem install embulk-parser-query_string
|
84
|
+
```
|
data/Rakefile
ADDED
@@ -0,0 +1,60 @@
|
|
1
|
+
require "bundler/gem_tasks"
|
2
|
+
require "json"
|
3
|
+
|
4
|
+
task default: :test
|
5
|
+
|
6
|
+
desc "Run tests"
|
7
|
+
task :test do
|
8
|
+
ruby("test/run-test.rb", "--use-color=yes")
|
9
|
+
end
|
10
|
+
|
11
|
+
namespace :release do
|
12
|
+
desc "Add header of now version release to ChangeLog and bump up version"
|
13
|
+
task :prepare do
|
14
|
+
root_dir = Pathname.new(File.expand_path("../", __FILE__))
|
15
|
+
changelog_file = root_dir.join("CHANGELOG.md")
|
16
|
+
gemspec_file = root_dir.join("embulk-parser-query_string.gemspec")
|
17
|
+
|
18
|
+
system("git fetch origin")
|
19
|
+
|
20
|
+
# detect merged PR
|
21
|
+
old_version = gemspec_file.read[/spec\.version += *"([0-9]+\.[0-9]+\.[0-9]+)"/, 1]
|
22
|
+
pr_numbers = `git log v#{old_version}..origin/master --oneline`.scan(/#[0-9]+/)
|
23
|
+
|
24
|
+
if !$?.success? || pr_numbers.empty?
|
25
|
+
puts "Detecting PR failed. Please confirm if any PR were merged after the latest release."
|
26
|
+
exit(false)
|
27
|
+
end
|
28
|
+
|
29
|
+
# Generate new version
|
30
|
+
major, minor, patch = old_version.split(".").map(&:to_i)
|
31
|
+
new_version = "#{major}.#{minor}.#{patch + 1}"
|
32
|
+
|
33
|
+
# Update ChangeLog
|
34
|
+
pr_descriptions = pr_numbers.map do |number|
|
35
|
+
body = open("https://api.github.com/repos/treasure-data/embulk-parser-query_string/issues/#{number.gsub("#", "")}").read
|
36
|
+
payload = JSON.parse(body)
|
37
|
+
"* [] #{payload["title"]} [#{number}](https://github.com/treasure-data/embulk-parser-query_string/pull/#{number.gsub('#', '')})"
|
38
|
+
end.join("\n")
|
39
|
+
|
40
|
+
new_changelog = <<-HEADER
|
41
|
+
## #{new_version} - #{Time.now.strftime("%Y-%m-%d")}
|
42
|
+
#{pr_descriptions}
|
43
|
+
|
44
|
+
#{changelog_file.read.chomp}
|
45
|
+
HEADER
|
46
|
+
|
47
|
+
File.open(changelog_file, "w") {|f| f.write(new_changelog) }
|
48
|
+
|
49
|
+
# Update version.rb
|
50
|
+
old_content = gemspec_file.read
|
51
|
+
File.open(gemspec_file, "w") do |f|
|
52
|
+
f.write old_content.gsub(/(spec\.version += *)".*?"/, %Q!\\1"#{new_version}"!)
|
53
|
+
end
|
54
|
+
|
55
|
+
# Update Gemfile.lock
|
56
|
+
system("bundle install")
|
57
|
+
|
58
|
+
puts "ChangeLog, version and Gemfile.lock were updated. New version is #{new_version}."
|
59
|
+
end
|
60
|
+
end
|
@@ -0,0 +1,23 @@
|
|
1
|
+
|
2
|
+
Gem::Specification.new do |spec|
|
3
|
+
spec.name = "embulk-parser-query_string"
|
4
|
+
spec.version = "0.0.2"
|
5
|
+
spec.authors = ["yoshihara", "uu59"]
|
6
|
+
spec.summary = "Query String parser plugin for Embulk"
|
7
|
+
spec.description = "Parses Query String files read by other file input plugins."
|
8
|
+
spec.email = ["h.yoshihara@everyleaf.com", "k@uu59.org"]
|
9
|
+
spec.licenses = ["Apache2"]
|
10
|
+
spec.homepage = "https://github.com/treasure-data/embulk-parser-query_string"
|
11
|
+
|
12
|
+
spec.files = `git ls-files`.split("\n") + Dir["classpath/*.jar"]
|
13
|
+
spec.test_files = spec.files.grep(%r{^(test|spec)/})
|
14
|
+
spec.require_paths = ["lib"]
|
15
|
+
|
16
|
+
spec.add_development_dependency 'embulk', [">= 0.6.13", "< 1.0"]
|
17
|
+
spec.add_development_dependency 'bundler', ['~> 1.0']
|
18
|
+
spec.add_development_dependency 'rake', ['>= 10.0']
|
19
|
+
spec.add_development_dependency 'pry'
|
20
|
+
spec.add_development_dependency 'test-unit'
|
21
|
+
spec.add_development_dependency 'test-unit-rr'
|
22
|
+
spec.add_development_dependency 'codeclimate-test-reporter'
|
23
|
+
end
|
@@ -0,0 +1,39 @@
|
|
1
|
+
require "embulk/parser/query_string"
|
2
|
+
|
3
|
+
module Embulk
|
4
|
+
module Guess
|
5
|
+
# $ embulk guess -g "query_string" partial-config.yml
|
6
|
+
|
7
|
+
class QueryString < LineGuessPlugin
|
8
|
+
Plugin.register_guess("query_string", self)
|
9
|
+
|
10
|
+
def guess_lines(config, sample_lines)
|
11
|
+
return {} unless config.fetch("parser", {}).fetch("type", "query_string") == "query_string"
|
12
|
+
|
13
|
+
options = {
|
14
|
+
strip_quote: config.param("strip_quote", :bool, default: true),
|
15
|
+
strip_whitespace: config.param("strip_whitespace", :bool, default: true)
|
16
|
+
}
|
17
|
+
records = sample_lines.map do |line|
|
18
|
+
Parser::QueryString.parse(line, options) || {}
|
19
|
+
end
|
20
|
+
format = records.inject({}) do |result, record|
|
21
|
+
record.each_pair do |key, value|
|
22
|
+
(result[key] ||= []) << value
|
23
|
+
end
|
24
|
+
result
|
25
|
+
end
|
26
|
+
guessed = {type: "query_string", schema: []}
|
27
|
+
format.each_pair do |key, values|
|
28
|
+
if values.any? {|value| value.match(/[^0-9]/) }
|
29
|
+
guessed[:schema] << {name: key, type: :string}
|
30
|
+
else
|
31
|
+
guessed[:schema] << {name: key, type: :long}
|
32
|
+
end
|
33
|
+
end
|
34
|
+
return {"parser" => guessed}
|
35
|
+
end
|
36
|
+
end
|
37
|
+
|
38
|
+
end
|
39
|
+
end
|
@@ -0,0 +1,79 @@
|
|
1
|
+
require "uri"
|
2
|
+
|
3
|
+
module Embulk
|
4
|
+
module Parser
|
5
|
+
|
6
|
+
class QueryString < ParserPlugin
|
7
|
+
Plugin.register_parser("query_string", self)
|
8
|
+
|
9
|
+
def self.transaction(config, &control)
|
10
|
+
decoder_task = config.load_config(Java::LineDecoder::DecoderTask)
|
11
|
+
|
12
|
+
task = {
|
13
|
+
decoder: DataSource.from_java(decoder_task.dump),
|
14
|
+
strip_quote: config.param("strip_quote", :bool, default: true),
|
15
|
+
strip_whitespace: config.param("strip_whitespace", :bool, default: true),
|
16
|
+
}
|
17
|
+
|
18
|
+
columns = []
|
19
|
+
schema = config.param(:schema, :array, default: [])
|
20
|
+
schema.each do |column|
|
21
|
+
name = column["name"]
|
22
|
+
type = column["type"].to_sym
|
23
|
+
|
24
|
+
columns << Column.new(nil, name, type)
|
25
|
+
end
|
26
|
+
|
27
|
+
yield(task, columns)
|
28
|
+
end
|
29
|
+
|
30
|
+
def init
|
31
|
+
@options = {
|
32
|
+
strip_quote: task[:strip_quote],
|
33
|
+
strip_whitespace: task[:strip_whitespace],
|
34
|
+
}
|
35
|
+
|
36
|
+
@decoder = task.param(:decoder, :hash).load_task(Java::LineDecoder::DecoderTask)
|
37
|
+
end
|
38
|
+
|
39
|
+
def run(file_input)
|
40
|
+
decoder = Java::LineDecoder.new(file_input.instance_variable_get(:@java_file_input), @decoder)
|
41
|
+
|
42
|
+
while decoder.nextFile
|
43
|
+
while line = decoder.poll
|
44
|
+
process_line(line)
|
45
|
+
end
|
46
|
+
end
|
47
|
+
|
48
|
+
page_builder.finish
|
49
|
+
end
|
50
|
+
|
51
|
+
def self.parse(line, options = {})
|
52
|
+
line.chomp!
|
53
|
+
line.strip! if options[:strip_whitespace]
|
54
|
+
if options[:strip_quote]
|
55
|
+
line = line[/\A(?:["'])?(.*?)(?:["'])?\z/, 1]
|
56
|
+
end
|
57
|
+
|
58
|
+
begin
|
59
|
+
Hash[URI.decode_www_form(line)]
|
60
|
+
rescue ArgumentError
|
61
|
+
nil
|
62
|
+
end
|
63
|
+
end
|
64
|
+
|
65
|
+
private
|
66
|
+
|
67
|
+
def process_line(line)
|
68
|
+
record = self.class.parse(line, @options)
|
69
|
+
|
70
|
+
return unless record
|
71
|
+
|
72
|
+
records = schema.map do |column|
|
73
|
+
record[column.name]
|
74
|
+
end
|
75
|
+
page_builder.add(records)
|
76
|
+
end
|
77
|
+
end
|
78
|
+
end
|
79
|
+
end
|
data/partial-config.yml
ADDED
@@ -0,0 +1,122 @@
|
|
1
|
+
require "prepare_embulk"
|
2
|
+
require "embulk/guess/query_string"
|
3
|
+
require "embulk/data_source"
|
4
|
+
|
5
|
+
module Embulk
|
6
|
+
module Guess
|
7
|
+
class QueryStringTest < Test::Unit::TestCase
|
8
|
+
class TestGuessLines < self
|
9
|
+
data do
|
10
|
+
{
|
11
|
+
same_keys: [sample_lines_with_same_keys, schema_with_same_keys],
|
12
|
+
different_keys: [sample_lines_with_different_keys, schema_with_different_keys],
|
13
|
+
invalid: [sample_lines_with_invalid, schema_with_invalid],
|
14
|
+
|
15
|
+
}
|
16
|
+
end
|
17
|
+
|
18
|
+
def test_schema(data)
|
19
|
+
sample_lines, schema = data
|
20
|
+
actual = QueryString.new.guess_lines(config, sample_lines)
|
21
|
+
expected = {
|
22
|
+
"parser" => {
|
23
|
+
type: "query_string",
|
24
|
+
schema: schema
|
25
|
+
}
|
26
|
+
}
|
27
|
+
assert_equal(expected, actual)
|
28
|
+
end
|
29
|
+
|
30
|
+
data do
|
31
|
+
valid_schema = {
|
32
|
+
"parser" => {
|
33
|
+
type: "query_string",
|
34
|
+
schema: schema_with_same_keys,
|
35
|
+
}
|
36
|
+
}
|
37
|
+
|
38
|
+
{
|
39
|
+
"query_string" => ["query_string", valid_schema],
|
40
|
+
"other" => ["other", {}],
|
41
|
+
}
|
42
|
+
end
|
43
|
+
|
44
|
+
def test_type(data)
|
45
|
+
type, expected = data
|
46
|
+
sample_lines = self.class.sample_lines_with_same_keys
|
47
|
+
config = DataSource[{parser: {type: type}}]
|
48
|
+
|
49
|
+
actual = QueryString.new.guess_lines(config, sample_lines)
|
50
|
+
assert_equal(expected, actual)
|
51
|
+
end
|
52
|
+
end
|
53
|
+
|
54
|
+
private
|
55
|
+
|
56
|
+
class << self
|
57
|
+
def sample_lines_with_same_keys
|
58
|
+
[
|
59
|
+
%Q(foo=1&bar=vv&baz=3),
|
60
|
+
%Q(foo=2&bar=ss&baz=a),
|
61
|
+
]
|
62
|
+
end
|
63
|
+
|
64
|
+
def schema_with_same_keys
|
65
|
+
[
|
66
|
+
{name: "foo", type: :long},
|
67
|
+
{name: "bar", type: :string},
|
68
|
+
{name: "baz", type: :string},
|
69
|
+
]
|
70
|
+
end
|
71
|
+
|
72
|
+
def sample_lines_with_different_keys
|
73
|
+
[
|
74
|
+
%Q(foo=1&bar=vv&baz=3&hoge=999),
|
75
|
+
%Q(foo=2&bar=ss&baz=a&xxx=ABC),
|
76
|
+
]
|
77
|
+
end
|
78
|
+
|
79
|
+
def schema_with_different_keys
|
80
|
+
[
|
81
|
+
{name: "foo", type: :long},
|
82
|
+
{name: "bar", type: :string},
|
83
|
+
{name: "baz", type: :string},
|
84
|
+
{name: "hoge", type: :long},
|
85
|
+
{name: "xxx", type: :string},
|
86
|
+
]
|
87
|
+
end
|
88
|
+
|
89
|
+
def sample_lines_with_invalid
|
90
|
+
[
|
91
|
+
%Q(foo=1&bar=vv&baz=3),
|
92
|
+
%Q(this=line=is=invalid),
|
93
|
+
%Q(foo=2&bar=ss&baz=a),
|
94
|
+
]
|
95
|
+
end
|
96
|
+
|
97
|
+
def schema_with_invalid
|
98
|
+
schema_with_same_keys
|
99
|
+
end
|
100
|
+
end
|
101
|
+
|
102
|
+
def task
|
103
|
+
{
|
104
|
+
strip_quote: true,
|
105
|
+
strip_whitespace: true,
|
106
|
+
schema: columns,
|
107
|
+
}
|
108
|
+
end
|
109
|
+
|
110
|
+
def columns
|
111
|
+
[
|
112
|
+
{"name" => "foo", "type" => "string"},
|
113
|
+
{"name" => "bar", "type" => "string"},
|
114
|
+
]
|
115
|
+
end
|
116
|
+
|
117
|
+
def config
|
118
|
+
DataSource[task.to_a]
|
119
|
+
end
|
120
|
+
end
|
121
|
+
end
|
122
|
+
end
|
@@ -0,0 +1,120 @@
|
|
1
|
+
require "prepare_embulk"
|
2
|
+
require "embulk/parser/query_string"
|
3
|
+
require "embulk/data_source"
|
4
|
+
|
5
|
+
module Embulk
|
6
|
+
module Parser
|
7
|
+
class QueryStringPluginTest < Test::Unit::TestCase
|
8
|
+
class TestParse < self
|
9
|
+
def test_without_options
|
10
|
+
result = QueryString.parse(line)
|
11
|
+
assert_equal(expected, result)
|
12
|
+
end
|
13
|
+
|
14
|
+
def test_with_strip_quote
|
15
|
+
result = QueryString.parse(quoted_line, strip_quote: true)
|
16
|
+
assert_equal(expected, result)
|
17
|
+
end
|
18
|
+
|
19
|
+
def test_with_strip_whitespace
|
20
|
+
result = QueryString.parse(indented_line, strip_whitespace: true)
|
21
|
+
assert_equal(expected, result)
|
22
|
+
end
|
23
|
+
|
24
|
+
def test_with_invalid
|
25
|
+
result = QueryString.parse(invalid_line)
|
26
|
+
assert_nil(result)
|
27
|
+
end
|
28
|
+
|
29
|
+
private
|
30
|
+
|
31
|
+
def expected
|
32
|
+
{"foo" => "FOO", "bar" => "3"}
|
33
|
+
end
|
34
|
+
|
35
|
+
def line
|
36
|
+
%Q(foo=FOO&bar=3)
|
37
|
+
end
|
38
|
+
|
39
|
+
def quoted_line
|
40
|
+
%Q("#{line}")
|
41
|
+
end
|
42
|
+
|
43
|
+
def indented_line
|
44
|
+
%Q( #{line})
|
45
|
+
end
|
46
|
+
|
47
|
+
def invalid_line
|
48
|
+
"invalid=www=form"
|
49
|
+
end
|
50
|
+
end
|
51
|
+
|
52
|
+
class TestProcessBuffer < self
|
53
|
+
def test_process_line
|
54
|
+
records.each do |record|
|
55
|
+
mock(page_builder).add(record.values)
|
56
|
+
end
|
57
|
+
|
58
|
+
plugin.send(:process_line, line)
|
59
|
+
end
|
60
|
+
|
61
|
+
private
|
62
|
+
|
63
|
+
def records
|
64
|
+
[
|
65
|
+
{"foo" => "FOO", "bar" => "1"},
|
66
|
+
]
|
67
|
+
end
|
68
|
+
|
69
|
+
def line
|
70
|
+
"foo=FOO&bar=1"
|
71
|
+
end
|
72
|
+
end
|
73
|
+
|
74
|
+
def test_transaction
|
75
|
+
QueryString.transaction(config) do |actual_task, actual_columns|
|
76
|
+
t = task.dup
|
77
|
+
t.delete(:schema)
|
78
|
+
assert_equal(t, actual_task)
|
79
|
+
assert_equal(schema, actual_columns)
|
80
|
+
end
|
81
|
+
end
|
82
|
+
|
83
|
+
private
|
84
|
+
|
85
|
+
def plugin
|
86
|
+
@plugin ||= QueryString.new(DataSource[task], schema, page_builder)
|
87
|
+
end
|
88
|
+
|
89
|
+
def page_builder
|
90
|
+
@page_builder ||= Object.new
|
91
|
+
end
|
92
|
+
|
93
|
+
def task
|
94
|
+
{
|
95
|
+
decoder: {"Charset" => "UTF-8", "Newline" => "CRLF"},
|
96
|
+
strip_quote: true,
|
97
|
+
strip_whitespace: true,
|
98
|
+
schema: columns,
|
99
|
+
}
|
100
|
+
end
|
101
|
+
|
102
|
+
def columns
|
103
|
+
[
|
104
|
+
{"name" => "foo", "type" => "string"},
|
105
|
+
{"name" => "bar", "type" => "string"},
|
106
|
+
]
|
107
|
+
end
|
108
|
+
|
109
|
+
def schema
|
110
|
+
columns.map do |column|
|
111
|
+
Column.new(nil, column["name"], column["type"].to_sym)
|
112
|
+
end
|
113
|
+
end
|
114
|
+
|
115
|
+
def config
|
116
|
+
DataSource[task.to_a]
|
117
|
+
end
|
118
|
+
end
|
119
|
+
end
|
120
|
+
end
|
@@ -0,0 +1,16 @@
|
|
1
|
+
require "embulk/command/embulk_run"
|
2
|
+
|
3
|
+
classpath_dir = Embulk.home("classpath")
|
4
|
+
jars = Dir.entries(classpath_dir).select{|f| f =~ /\.jar$/ }.sort
|
5
|
+
jars.each do |jar|
|
6
|
+
require File.join(classpath_dir, jar)
|
7
|
+
end
|
8
|
+
|
9
|
+
props = java.util.Properties.new
|
10
|
+
props.setProperty("embulk.use_global_ruby_runtime", "true")
|
11
|
+
|
12
|
+
bootstrap_model_manager = org.embulk.config.ModelManager.new(nil, com.fasterxml.jackson.databind.ObjectMapper.new)
|
13
|
+
system_config = org.embulk.config.ConfigLoader.new(bootstrap_model_manager).fromPropertiesYamlLiteral(props, "embulk.")
|
14
|
+
org.embulk.EmbulkService.new(system_config).injector.getInstance(java.lang.Class.forName('org.jruby.embed.ScriptingContainer'))
|
15
|
+
|
16
|
+
require "embulk"
|
data/test/run-test.rb
ADDED
@@ -0,0 +1,18 @@
|
|
1
|
+
#!/usr/bin/env ruby
|
2
|
+
|
3
|
+
base_dir = File.expand_path(File.join(File.dirname(__FILE__), ".."))
|
4
|
+
lib_dir = File.join(base_dir, "lib")
|
5
|
+
test_dir = File.join(base_dir, "test")
|
6
|
+
|
7
|
+
require "test-unit"
|
8
|
+
require "test/unit/rr"
|
9
|
+
require "codeclimate-test-reporter"
|
10
|
+
|
11
|
+
$LOAD_PATH.unshift(lib_dir)
|
12
|
+
$LOAD_PATH.unshift(test_dir)
|
13
|
+
|
14
|
+
ENV["TEST_UNIT_MAX_DIFF_TARGET_STRING_SIZE"] ||= "5000"
|
15
|
+
|
16
|
+
CodeClimate::TestReporter.start
|
17
|
+
|
18
|
+
exit Test::Unit::AutoRunner.run(true, test_dir)
|
metadata
ADDED
@@ -0,0 +1,170 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: embulk-parser-query_string
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 0.0.2
|
5
|
+
platform: ruby
|
6
|
+
authors:
|
7
|
+
- yoshihara
|
8
|
+
- uu59
|
9
|
+
autorequire:
|
10
|
+
bindir: bin
|
11
|
+
cert_chain: []
|
12
|
+
date: 2015-07-08 00:00:00.000000000 Z
|
13
|
+
dependencies:
|
14
|
+
- !ruby/object:Gem::Dependency
|
15
|
+
requirement: !ruby/object:Gem::Requirement
|
16
|
+
requirements:
|
17
|
+
- - '>='
|
18
|
+
- !ruby/object:Gem::Version
|
19
|
+
version: 0.6.13
|
20
|
+
- - <
|
21
|
+
- !ruby/object:Gem::Version
|
22
|
+
version: '1.0'
|
23
|
+
name: embulk
|
24
|
+
prerelease: false
|
25
|
+
type: :development
|
26
|
+
version_requirements: !ruby/object:Gem::Requirement
|
27
|
+
requirements:
|
28
|
+
- - '>='
|
29
|
+
- !ruby/object:Gem::Version
|
30
|
+
version: 0.6.13
|
31
|
+
- - <
|
32
|
+
- !ruby/object:Gem::Version
|
33
|
+
version: '1.0'
|
34
|
+
- !ruby/object:Gem::Dependency
|
35
|
+
requirement: !ruby/object:Gem::Requirement
|
36
|
+
requirements:
|
37
|
+
- - ~>
|
38
|
+
- !ruby/object:Gem::Version
|
39
|
+
version: '1.0'
|
40
|
+
name: bundler
|
41
|
+
prerelease: false
|
42
|
+
type: :development
|
43
|
+
version_requirements: !ruby/object:Gem::Requirement
|
44
|
+
requirements:
|
45
|
+
- - ~>
|
46
|
+
- !ruby/object:Gem::Version
|
47
|
+
version: '1.0'
|
48
|
+
- !ruby/object:Gem::Dependency
|
49
|
+
requirement: !ruby/object:Gem::Requirement
|
50
|
+
requirements:
|
51
|
+
- - '>='
|
52
|
+
- !ruby/object:Gem::Version
|
53
|
+
version: '10.0'
|
54
|
+
name: rake
|
55
|
+
prerelease: false
|
56
|
+
type: :development
|
57
|
+
version_requirements: !ruby/object:Gem::Requirement
|
58
|
+
requirements:
|
59
|
+
- - '>='
|
60
|
+
- !ruby/object:Gem::Version
|
61
|
+
version: '10.0'
|
62
|
+
- !ruby/object:Gem::Dependency
|
63
|
+
requirement: !ruby/object:Gem::Requirement
|
64
|
+
requirements:
|
65
|
+
- - '>='
|
66
|
+
- !ruby/object:Gem::Version
|
67
|
+
version: '0'
|
68
|
+
name: pry
|
69
|
+
prerelease: false
|
70
|
+
type: :development
|
71
|
+
version_requirements: !ruby/object:Gem::Requirement
|
72
|
+
requirements:
|
73
|
+
- - '>='
|
74
|
+
- !ruby/object:Gem::Version
|
75
|
+
version: '0'
|
76
|
+
- !ruby/object:Gem::Dependency
|
77
|
+
requirement: !ruby/object:Gem::Requirement
|
78
|
+
requirements:
|
79
|
+
- - '>='
|
80
|
+
- !ruby/object:Gem::Version
|
81
|
+
version: '0'
|
82
|
+
name: test-unit
|
83
|
+
prerelease: false
|
84
|
+
type: :development
|
85
|
+
version_requirements: !ruby/object:Gem::Requirement
|
86
|
+
requirements:
|
87
|
+
- - '>='
|
88
|
+
- !ruby/object:Gem::Version
|
89
|
+
version: '0'
|
90
|
+
- !ruby/object:Gem::Dependency
|
91
|
+
requirement: !ruby/object:Gem::Requirement
|
92
|
+
requirements:
|
93
|
+
- - '>='
|
94
|
+
- !ruby/object:Gem::Version
|
95
|
+
version: '0'
|
96
|
+
name: test-unit-rr
|
97
|
+
prerelease: false
|
98
|
+
type: :development
|
99
|
+
version_requirements: !ruby/object:Gem::Requirement
|
100
|
+
requirements:
|
101
|
+
- - '>='
|
102
|
+
- !ruby/object:Gem::Version
|
103
|
+
version: '0'
|
104
|
+
- !ruby/object:Gem::Dependency
|
105
|
+
requirement: !ruby/object:Gem::Requirement
|
106
|
+
requirements:
|
107
|
+
- - '>='
|
108
|
+
- !ruby/object:Gem::Version
|
109
|
+
version: '0'
|
110
|
+
name: codeclimate-test-reporter
|
111
|
+
prerelease: false
|
112
|
+
type: :development
|
113
|
+
version_requirements: !ruby/object:Gem::Requirement
|
114
|
+
requirements:
|
115
|
+
- - '>='
|
116
|
+
- !ruby/object:Gem::Version
|
117
|
+
version: '0'
|
118
|
+
description: Parses Query String files read by other file input plugins.
|
119
|
+
email:
|
120
|
+
- h.yoshihara@everyleaf.com
|
121
|
+
- k@uu59.org
|
122
|
+
executables: []
|
123
|
+
extensions: []
|
124
|
+
extra_rdoc_files: []
|
125
|
+
files:
|
126
|
+
- .gitignore
|
127
|
+
- .travis.yml
|
128
|
+
- CHANGELOG.md
|
129
|
+
- Gemfile
|
130
|
+
- LICENSE
|
131
|
+
- LICENSE.txt
|
132
|
+
- README.md
|
133
|
+
- Rakefile
|
134
|
+
- embulk-parser-query_string.gemspec
|
135
|
+
- lib/embulk/guess/query_string.rb
|
136
|
+
- lib/embulk/parser/query_string.rb
|
137
|
+
- partial-config.yml
|
138
|
+
- test/embulk/guess/test_query_string.rb
|
139
|
+
- test/embulk/parser/test_query_string_plugin.rb
|
140
|
+
- test/prepare_embulk.rb
|
141
|
+
- test/run-test.rb
|
142
|
+
homepage: https://github.com/treasure-data/embulk-parser-query_string
|
143
|
+
licenses:
|
144
|
+
- Apache2
|
145
|
+
metadata: {}
|
146
|
+
post_install_message:
|
147
|
+
rdoc_options: []
|
148
|
+
require_paths:
|
149
|
+
- lib
|
150
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
151
|
+
requirements:
|
152
|
+
- - '>='
|
153
|
+
- !ruby/object:Gem::Version
|
154
|
+
version: '0'
|
155
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
156
|
+
requirements:
|
157
|
+
- - '>='
|
158
|
+
- !ruby/object:Gem::Version
|
159
|
+
version: '0'
|
160
|
+
requirements: []
|
161
|
+
rubyforge_project:
|
162
|
+
rubygems_version: 2.4.6
|
163
|
+
signing_key:
|
164
|
+
specification_version: 4
|
165
|
+
summary: Query String parser plugin for Embulk
|
166
|
+
test_files:
|
167
|
+
- test/embulk/guess/test_query_string.rb
|
168
|
+
- test/embulk/parser/test_query_string_plugin.rb
|
169
|
+
- test/prepare_embulk.rb
|
170
|
+
- test/run-test.rb
|