terc_parser 0.0.7
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/.gitignore +6 -0
- data/.rspec +1 -0
- data/Gemfile +4 -0
- data/README.md +72 -0
- data/Rakefile +1 -0
- data/lib/terc_parser.rb +2 -0
- data/lib/terc_parser/terc_parser.rb +54 -0
- data/lib/terc_parser/version.rb +3 -0
- data/spec/spec_helper.rb +1 -0
- data/spec/terc_parser/terc_parser_spec.rb +28 -0
- data/terc_parser.gemspec +22 -0
- data/vendor/.gitkeep +0 -0
- metadata +97 -0
checksums.yaml
ADDED
@@ -0,0 +1,7 @@
|
|
1
|
+
---
|
2
|
+
SHA1:
|
3
|
+
metadata.gz: e0891392b8c802d19f1a84bf41de1af2d97860a3
|
4
|
+
data.tar.gz: 50930c5ddb4b6f418fc47f09767591a05879f8a3
|
5
|
+
SHA512:
|
6
|
+
metadata.gz: 905bcdcd042aba9b888284712787314ab2d8ecb0da6842f1b017e731dcbb717a72301f14247d071a963a64f226a5321c225503497077d521d4fc800f62106315
|
7
|
+
data.tar.gz: bbcac2b5488cc3134c8ba736015cd027d9fdb4b3a9a8e000cda1f7e7df9edc0b6ac01896e312ef9261884653a396358ff0910e21d3d1d663761f86810d59497c
|
data/.gitignore
ADDED
data/.rspec
ADDED
@@ -0,0 +1 @@
|
|
1
|
+
--color
|
data/Gemfile
ADDED
data/README.md
ADDED
@@ -0,0 +1,72 @@
|
|
1
|
+
#TERC parser
|
2
|
+
|
3
|
+
Gem składa się praktycznie z jednej metody: `TercParser::parse_terc`. Po jej wywołaniu, opcjonalnie podając jako argument ścieżkę do pliku `TERC.xml`, zwróci ona hash zawierający cztery klucze:
|
4
|
+
|
5
|
+
* **:stan_na** - zawiera datę (jako string) utworzenia pliku `TERC.xml`
|
6
|
+
* **:wojewodztwa** - zawiera tablice hashy, każdy hash zawiera informacje o jednym województwie
|
7
|
+
* **:powiaty** - jw. ale dane o powiatach
|
8
|
+
* **:gminy** - jw. ale dane o gminach
|
9
|
+
|
10
|
+
Aktualny plik `TERC.xml` można pobrać ze strony GUSu:
|
11
|
+
http://www.stat.gov.pl/broker/access/prefile/listPreFiles.jspa
|
12
|
+
|
13
|
+
Plik `TERC.xml` jest cyfrową wersją [rejestru identyfikatorów i nazw jednostek podziału administracyjnego Polski](http://pl.wikipedia.org/wiki/TERC) udostępnionym przez [GUS](http://pl.wikipedia.org/wiki/GUS) i jest częścią rejestru [TERYT](http://pl.wikipedia.org/wiki/TERYT).
|
14
|
+
|
15
|
+
## Instalacja
|
16
|
+
|
17
|
+
``` bash
|
18
|
+
gem install terc_parser
|
19
|
+
```
|
20
|
+
|
21
|
+
## Przykład
|
22
|
+
|
23
|
+
Wynik działania:
|
24
|
+
|
25
|
+
``` bash
|
26
|
+
STAN NA: 2011-01-01
|
27
|
+
-------- WOJEWODZTWA (16) ---------
|
28
|
+
2 - dolnośląskie
|
29
|
+
4 - kujawsko-pomorskie
|
30
|
+
6 - lubelskie
|
31
|
+
8 - lubuskie
|
32
|
+
10 - łódzkie
|
33
|
+
12 - małopolskie
|
34
|
+
14 - mazowieckie
|
35
|
+
16 - opolskie
|
36
|
+
18 - podkarpackie
|
37
|
+
20 - podlaskie
|
38
|
+
22 - pomorskie
|
39
|
+
24 - śląskie
|
40
|
+
26 - świętokrzyskie
|
41
|
+
28 - warmińsko-mazurskie
|
42
|
+
30 - wielkopolskie
|
43
|
+
32 - zachodniopomorskie
|
44
|
+
-------- POWIATY (379) -------------
|
45
|
+
-------- GMINY (2479) -------------
|
46
|
+
```
|
47
|
+
|
48
|
+
Kod:
|
49
|
+
|
50
|
+
``` ruby
|
51
|
+
require "terc_parser"
|
52
|
+
|
53
|
+
terc = TercParser::parse_terc "/path_to_terc/TERC.xml"
|
54
|
+
|
55
|
+
puts "STAN NA: #{terc[:stan_na]}"
|
56
|
+
|
57
|
+
puts "-------- WOJEWODZTWA (#{terc[:wojewodztwa].count}) ---------"
|
58
|
+
terc[:wojewodztwa].each do |w|
|
59
|
+
puts "#{w[:id]} - #{w[:name]}"
|
60
|
+
end
|
61
|
+
|
62
|
+
puts "-------- POWIATY (#{terc[:powiaty].count}) -------------"
|
63
|
+
terc[:powiaty].each do |p|
|
64
|
+
#puts "#{p[:id]} - #{p[:woj_id]} - #{p[:name]}"
|
65
|
+
end
|
66
|
+
|
67
|
+
puts "-------- GMINY (#{terc[:gminy].count}) -------------"
|
68
|
+
terc[:gminy].each do |g|
|
69
|
+
#puts "#{g[:id]} - #{g[:woj_id]} - #{g[:pow_id]} - #{g[:name]}"
|
70
|
+
end
|
71
|
+
|
72
|
+
```
|
data/Rakefile
ADDED
@@ -0,0 +1 @@
|
|
1
|
+
require "bundler/gem_tasks"
|
data/lib/terc_parser.rb
ADDED
@@ -0,0 +1,54 @@
|
|
1
|
+
#encoding: utf-8
|
2
|
+
require "nokogiri"
|
3
|
+
require "active_support/core_ext/string/multibyte"
|
4
|
+
|
5
|
+
module TercParser
|
6
|
+
class << self
|
7
|
+
def parse_terc(terc_file_path = "TERC.xml")
|
8
|
+
stan_na = wojewodztwa = powiaty = gminy = nil
|
9
|
+
|
10
|
+
process_xml_file terc_file_path do |doc|
|
11
|
+
stan_na = doc.xpath("/teryt/catalog/@date").text
|
12
|
+
|
13
|
+
### WOJEWODZTWA ## ----------------------------------
|
14
|
+
wojewodztwa_nodeset = doc.xpath("//col[@name='NAZDOD' and text()='województwo']/..")
|
15
|
+
|
16
|
+
wojewodztwa = wojewodztwa_nodeset.map do |woj|
|
17
|
+
id = woj.xpath("col[@name = 'WOJ']").text.to_i
|
18
|
+
name = woj.xpath("col[@name = 'NAZWA']/text()").text.mb_chars.downcase.to_s
|
19
|
+
{id: id, name: name}
|
20
|
+
end
|
21
|
+
|
22
|
+
### POWIATY ## --------------------------------------
|
23
|
+
powiaty_nodeset = doc.xpath("//col[@name='NAZDOD' and contains(text(),'powiat')]/..")
|
24
|
+
|
25
|
+
powiaty = powiaty_nodeset.map do |woj|
|
26
|
+
id = woj.xpath("col[@name = 'POW']").text.to_i
|
27
|
+
woj_id = woj.xpath("col[@name = 'WOJ']").text.to_i
|
28
|
+
name = woj.xpath("col[@name = 'NAZWA']/text()").text.mb_chars.downcase.to_s
|
29
|
+
{id: id, woj_id: woj_id, name: name}
|
30
|
+
end
|
31
|
+
|
32
|
+
### GMINY ## --------------------------------------
|
33
|
+
gminy_nodeset = doc.xpath("//col[@name='NAZDOD' and contains(text(),'gmina')]/..")
|
34
|
+
|
35
|
+
gminy = gminy_nodeset.map do |woj|
|
36
|
+
id = woj.xpath("col[@name = 'GMI']").text.to_i
|
37
|
+
woj_id = woj.xpath("col[@name = 'WOJ']").text.to_i
|
38
|
+
pow_id = woj.xpath("col[@name = 'POW']").text.to_i
|
39
|
+
name = woj.xpath("col[@name = 'NAZWA']/text()").text.mb_chars.downcase.to_s
|
40
|
+
{id: id, woj_id: woj_id, pow_id: pow_id, name: name}
|
41
|
+
end
|
42
|
+
end
|
43
|
+
|
44
|
+
return {stan_na: stan_na, wojewodztwa: wojewodztwa, powiaty: powiaty, gminy: gminy}
|
45
|
+
end
|
46
|
+
|
47
|
+
def process_xml_file(file_name, &block)
|
48
|
+
f = File.open(file_name)
|
49
|
+
doc = Nokogiri::XML(f) { |config| config.strict }
|
50
|
+
yield doc
|
51
|
+
f.close
|
52
|
+
end
|
53
|
+
end
|
54
|
+
end
|
data/spec/spec_helper.rb
ADDED
@@ -0,0 +1 @@
|
|
1
|
+
require 'terc_parser'
|
@@ -0,0 +1,28 @@
|
|
1
|
+
#encoding: utf-8
|
2
|
+
require 'spec_helper'
|
3
|
+
|
4
|
+
TERC_XML_FILE_PATH = File.expand_path('../../../vendor/TERC.xml', __FILE__)
|
5
|
+
|
6
|
+
describe TercParser do
|
7
|
+
let(:terc) { TercParser::parse_terc TERC_XML_FILE_PATH }
|
8
|
+
|
9
|
+
it "should collect data that are correct type" do
|
10
|
+
terc.should be_instance_of Hash
|
11
|
+
terc[:stan_na].class.should == String
|
12
|
+
|
13
|
+
terc[:wojewodztwa].should be_instance_of Array
|
14
|
+
terc[:wojewodztwa][0][:id].should be_instance_of Fixnum
|
15
|
+
terc[:wojewodztwa][0][:name].should be_instance_of String
|
16
|
+
|
17
|
+
terc[:powiaty].should be_instance_of Array
|
18
|
+
terc[:powiaty][0][:id].should be_instance_of Fixnum
|
19
|
+
terc[:powiaty][0][:woj_id].should be_instance_of Fixnum
|
20
|
+
terc[:powiaty][0][:name].should be_instance_of String
|
21
|
+
|
22
|
+
terc[:gminy].should be_instance_of Array
|
23
|
+
terc[:gminy][0][:id].should be_instance_of Fixnum
|
24
|
+
terc[:gminy][0][:woj_id].should be_instance_of Fixnum
|
25
|
+
terc[:gminy][0][:pow_id].should be_instance_of Fixnum
|
26
|
+
terc[:gminy][0][:name].should be_instance_of String
|
27
|
+
end
|
28
|
+
end
|
data/terc_parser.gemspec
ADDED
@@ -0,0 +1,22 @@
|
|
1
|
+
# coding: utf-8
|
2
|
+
lib = File.expand_path('../lib', __FILE__)
|
3
|
+
$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
|
4
|
+
require "terc_parser/version"
|
5
|
+
|
6
|
+
Gem::Specification.new do |s|
|
7
|
+
s.name = "terc_parser"
|
8
|
+
s.version = TercParser::VERSION
|
9
|
+
s.authors = ["TeWu"]
|
10
|
+
s.email = ["tewu.dev@gmail.com"]
|
11
|
+
s.homepage = "https://github.com/TeWu/TERC_parser"
|
12
|
+
s.summary = %q{Nazwy województw, powiatów i gmin w jednym Hashu}
|
13
|
+
|
14
|
+
s.files = `git ls-files`.split("\n")
|
15
|
+
s.test_files = `git ls-files -- {test,spec,features}/*`.split("\n")
|
16
|
+
s.executables = `git ls-files -- bin/*`.split("\n").map { |f| File.basename(f) }
|
17
|
+
s.require_paths = ["lib"]
|
18
|
+
|
19
|
+
s.add_development_dependency "rspec", "~> 3.5"
|
20
|
+
s.add_runtime_dependency "nokogiri", "~> 1.6"
|
21
|
+
s.add_runtime_dependency "activesupport", "~> 5.0"
|
22
|
+
end
|
data/vendor/.gitkeep
ADDED
File without changes
|
metadata
ADDED
@@ -0,0 +1,97 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: terc_parser
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 0.0.7
|
5
|
+
platform: ruby
|
6
|
+
authors:
|
7
|
+
- TeWu
|
8
|
+
autorequire:
|
9
|
+
bindir: bin
|
10
|
+
cert_chain: []
|
11
|
+
date: 2016-10-25 00:00:00.000000000 Z
|
12
|
+
dependencies:
|
13
|
+
- !ruby/object:Gem::Dependency
|
14
|
+
name: rspec
|
15
|
+
requirement: !ruby/object:Gem::Requirement
|
16
|
+
requirements:
|
17
|
+
- - "~>"
|
18
|
+
- !ruby/object:Gem::Version
|
19
|
+
version: '3.5'
|
20
|
+
type: :development
|
21
|
+
prerelease: false
|
22
|
+
version_requirements: !ruby/object:Gem::Requirement
|
23
|
+
requirements:
|
24
|
+
- - "~>"
|
25
|
+
- !ruby/object:Gem::Version
|
26
|
+
version: '3.5'
|
27
|
+
- !ruby/object:Gem::Dependency
|
28
|
+
name: nokogiri
|
29
|
+
requirement: !ruby/object:Gem::Requirement
|
30
|
+
requirements:
|
31
|
+
- - "~>"
|
32
|
+
- !ruby/object:Gem::Version
|
33
|
+
version: '1.6'
|
34
|
+
type: :runtime
|
35
|
+
prerelease: false
|
36
|
+
version_requirements: !ruby/object:Gem::Requirement
|
37
|
+
requirements:
|
38
|
+
- - "~>"
|
39
|
+
- !ruby/object:Gem::Version
|
40
|
+
version: '1.6'
|
41
|
+
- !ruby/object:Gem::Dependency
|
42
|
+
name: activesupport
|
43
|
+
requirement: !ruby/object:Gem::Requirement
|
44
|
+
requirements:
|
45
|
+
- - "~>"
|
46
|
+
- !ruby/object:Gem::Version
|
47
|
+
version: '5.0'
|
48
|
+
type: :runtime
|
49
|
+
prerelease: false
|
50
|
+
version_requirements: !ruby/object:Gem::Requirement
|
51
|
+
requirements:
|
52
|
+
- - "~>"
|
53
|
+
- !ruby/object:Gem::Version
|
54
|
+
version: '5.0'
|
55
|
+
description:
|
56
|
+
email:
|
57
|
+
- tewu.dev@gmail.com
|
58
|
+
executables: []
|
59
|
+
extensions: []
|
60
|
+
extra_rdoc_files: []
|
61
|
+
files:
|
62
|
+
- ".gitignore"
|
63
|
+
- ".rspec"
|
64
|
+
- Gemfile
|
65
|
+
- README.md
|
66
|
+
- Rakefile
|
67
|
+
- lib/terc_parser.rb
|
68
|
+
- lib/terc_parser/terc_parser.rb
|
69
|
+
- lib/terc_parser/version.rb
|
70
|
+
- spec/spec_helper.rb
|
71
|
+
- spec/terc_parser/terc_parser_spec.rb
|
72
|
+
- terc_parser.gemspec
|
73
|
+
- vendor/.gitkeep
|
74
|
+
homepage: https://github.com/TeWu/TERC_parser
|
75
|
+
licenses: []
|
76
|
+
metadata: {}
|
77
|
+
post_install_message:
|
78
|
+
rdoc_options: []
|
79
|
+
require_paths:
|
80
|
+
- lib
|
81
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
82
|
+
requirements:
|
83
|
+
- - ">="
|
84
|
+
- !ruby/object:Gem::Version
|
85
|
+
version: '0'
|
86
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
87
|
+
requirements:
|
88
|
+
- - ">="
|
89
|
+
- !ruby/object:Gem::Version
|
90
|
+
version: '0'
|
91
|
+
requirements: []
|
92
|
+
rubyforge_project:
|
93
|
+
rubygems_version: 2.4.8
|
94
|
+
signing_key:
|
95
|
+
specification_version: 4
|
96
|
+
summary: Nazwy województw, powiatów i gmin w jednym Hashu
|
97
|
+
test_files: []
|