text_extractor 0.0.2 → 0.1.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/lib/text_extractor.rb +9 -9
- data/lib/text_extractor/filldown.rb +1 -1
- data/lib/text_extractor/record.rb +15 -3
- data/lib/text_extractor/version.rb +1 -1
- metadata +56 -15
checksums.yaml
ADDED
@@ -0,0 +1,7 @@
|
|
1
|
+
---
|
2
|
+
SHA1:
|
3
|
+
metadata.gz: 80ea0027acddf0b1b3cd96a78737c6e990159a71
|
4
|
+
data.tar.gz: 4938b707b639b5efde6013e2c06f1c04a57567a6
|
5
|
+
SHA512:
|
6
|
+
metadata.gz: a9d49fa0d38752dbe50969db22489740000ecce854c4e1105e8dd6be0b6a24328f799c6d3594e9a7a3fcab3c7d6f96512ca233030b20bf35461227bbd4351efd
|
7
|
+
data.tar.gz: 1281833781015336ae886392e07d884ed82787cdba4a5d2eb57f2f1740c5df40ab95d3de88ed38be73034879025c9be2998f62119479fbee86e41a59aeb5960a
|
data/lib/text_extractor.rb
CHANGED
@@ -1,14 +1,14 @@
|
|
1
|
-
require_relative
|
2
|
-
require_relative
|
3
|
-
require_relative
|
4
|
-
require_relative
|
1
|
+
require_relative 'text_extractor/extraction'
|
2
|
+
require_relative 'text_extractor/filldown'
|
3
|
+
require_relative 'text_extractor/record'
|
4
|
+
require_relative 'text_extractor/value'
|
5
5
|
|
6
6
|
# represents an extractor definition
|
7
7
|
class TextExtractor
|
8
8
|
attr_reader :records, :values
|
9
9
|
|
10
10
|
def initialize(&block)
|
11
|
-
|
11
|
+
raise "#{self.class}.new requires a block" unless block
|
12
12
|
@values = {}
|
13
13
|
@fill = {}
|
14
14
|
@values = {}
|
@@ -21,7 +21,7 @@ class TextExtractor
|
|
21
21
|
module Patterns
|
22
22
|
INTEGER = /\d+/
|
23
23
|
FLOAT = /\d+\.?|\d*\.\d+/
|
24
|
-
RATIONAL = %r
|
24
|
+
RATIONAL = %r{\d+/\d+}
|
25
25
|
IPV4 = /[0-9.]{7,15}/
|
26
26
|
IPV6 = /[:a-fA-F0-9\.]{2,45}/
|
27
27
|
IPADDR = Regexp.union(IPV4, IPV6)
|
@@ -68,12 +68,12 @@ class TextExtractor
|
|
68
68
|
def strip_record(regexp)
|
69
69
|
lines = regexp.source.lines
|
70
70
|
prefix = lines.last
|
71
|
-
lines.map! { |s| s.gsub(
|
71
|
+
lines.map! { |s| s.gsub(prefix.to_s, '') } if prefix =~ /\A\s*\z/
|
72
72
|
Regexp.new(lines.join.strip, regexp.options)
|
73
73
|
end
|
74
74
|
|
75
75
|
def record(klass = Record, **kwargs, &block)
|
76
|
-
|
76
|
+
raise "#{self.class}.record requires a block" unless block
|
77
77
|
@current_record_values = []
|
78
78
|
regexp = strip_record(instance_exec(&block))
|
79
79
|
kwargs[:values] = @current_record_values
|
@@ -81,7 +81,7 @@ class TextExtractor
|
|
81
81
|
end
|
82
82
|
|
83
83
|
def filldown(**kwargs, &block)
|
84
|
-
|
84
|
+
raise "#{self.class}.filldown requires a block" unless block
|
85
85
|
record(Filldown, **kwargs, &block)
|
86
86
|
end
|
87
87
|
|
@@ -12,9 +12,21 @@ class TextExtractor
|
|
12
12
|
|
13
13
|
def extraction(match, fill)
|
14
14
|
extracted = {}.merge!(@default_values)
|
15
|
-
|
16
|
-
|
17
|
-
|
15
|
+
.merge!(extract_fills fill)
|
16
|
+
.merge!(extract_values match)
|
17
|
+
build_extraction(extracted)
|
18
|
+
end
|
19
|
+
|
20
|
+
def build_extraction(extracted)
|
21
|
+
case factory
|
22
|
+
when Hash
|
23
|
+
klass, params = factory.first
|
24
|
+
klass.new(*extracted.values_at(*params))
|
25
|
+
when Class
|
26
|
+
factory.new(*extracted.values)
|
27
|
+
else
|
28
|
+
extracted
|
29
|
+
end
|
18
30
|
end
|
19
31
|
|
20
32
|
def match(string, pos = 0)
|
metadata
CHANGED
@@ -1,16 +1,57 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: text_extractor
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.0
|
5
|
-
prerelease:
|
4
|
+
version: 0.1.0
|
6
5
|
platform: ruby
|
7
6
|
authors:
|
8
7
|
- Ben Miller
|
9
8
|
autorequire:
|
10
9
|
bindir: bin
|
11
10
|
cert_chain: []
|
12
|
-
date:
|
13
|
-
dependencies:
|
11
|
+
date: 2016-03-11 00:00:00.000000000 Z
|
12
|
+
dependencies:
|
13
|
+
- !ruby/object:Gem::Dependency
|
14
|
+
name: bundler
|
15
|
+
requirement: !ruby/object:Gem::Requirement
|
16
|
+
requirements:
|
17
|
+
- - "~>"
|
18
|
+
- !ruby/object:Gem::Version
|
19
|
+
version: '1.0'
|
20
|
+
type: :development
|
21
|
+
prerelease: false
|
22
|
+
version_requirements: !ruby/object:Gem::Requirement
|
23
|
+
requirements:
|
24
|
+
- - "~>"
|
25
|
+
- !ruby/object:Gem::Version
|
26
|
+
version: '1.0'
|
27
|
+
- !ruby/object:Gem::Dependency
|
28
|
+
name: rake
|
29
|
+
requirement: !ruby/object:Gem::Requirement
|
30
|
+
requirements:
|
31
|
+
- - "~>"
|
32
|
+
- !ruby/object:Gem::Version
|
33
|
+
version: '10.0'
|
34
|
+
type: :development
|
35
|
+
prerelease: false
|
36
|
+
version_requirements: !ruby/object:Gem::Requirement
|
37
|
+
requirements:
|
38
|
+
- - "~>"
|
39
|
+
- !ruby/object:Gem::Version
|
40
|
+
version: '10.0'
|
41
|
+
- !ruby/object:Gem::Dependency
|
42
|
+
name: minitest
|
43
|
+
requirement: !ruby/object:Gem::Requirement
|
44
|
+
requirements:
|
45
|
+
- - "~>"
|
46
|
+
- !ruby/object:Gem::Version
|
47
|
+
version: '5.0'
|
48
|
+
type: :development
|
49
|
+
prerelease: false
|
50
|
+
version_requirements: !ruby/object:Gem::Requirement
|
51
|
+
requirements:
|
52
|
+
- - "~>"
|
53
|
+
- !ruby/object:Gem::Version
|
54
|
+
version: '5.0'
|
14
55
|
description:
|
15
56
|
email: bjmllr@gmail.com
|
16
57
|
executables: []
|
@@ -18,34 +59,34 @@ extensions: []
|
|
18
59
|
extra_rdoc_files: []
|
19
60
|
files:
|
20
61
|
- lib/text_extractor.rb
|
21
|
-
- lib/text_extractor/version.rb
|
22
62
|
- lib/text_extractor/extraction.rb
|
23
|
-
- lib/text_extractor/value.rb
|
24
|
-
- lib/text_extractor/record.rb
|
25
63
|
- lib/text_extractor/filldown.rb
|
64
|
+
- lib/text_extractor/record.rb
|
65
|
+
- lib/text_extractor/value.rb
|
66
|
+
- lib/text_extractor/version.rb
|
26
67
|
homepage: https://github.com/bjmllr/text_extractor
|
27
68
|
licenses:
|
28
|
-
-
|
69
|
+
- GPL-3.0
|
70
|
+
metadata: {}
|
29
71
|
post_install_message:
|
30
72
|
rdoc_options: []
|
31
73
|
require_paths:
|
32
74
|
- lib
|
33
75
|
required_ruby_version: !ruby/object:Gem::Requirement
|
34
|
-
none: false
|
35
76
|
requirements:
|
36
|
-
- -
|
77
|
+
- - ">="
|
37
78
|
- !ruby/object:Gem::Version
|
38
|
-
version: 2.
|
79
|
+
version: 2.1.0
|
39
80
|
required_rubygems_version: !ruby/object:Gem::Requirement
|
40
|
-
none: false
|
41
81
|
requirements:
|
42
|
-
- -
|
82
|
+
- - ">="
|
43
83
|
- !ruby/object:Gem::Version
|
44
84
|
version: '0'
|
45
85
|
requirements: []
|
46
86
|
rubyforge_project:
|
47
|
-
rubygems_version:
|
87
|
+
rubygems_version: 2.5.1
|
48
88
|
signing_key:
|
49
|
-
specification_version:
|
89
|
+
specification_version: 4
|
50
90
|
summary: Easily extract data from text
|
51
91
|
test_files: []
|
92
|
+
has_rdoc:
|