boilerpipe-ruby 0.2.0 → 0.3.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/CHANGELOG.md +7 -0
- data/README.md +19 -5
- data/lib/boilerpipe.rb +6 -0
- data/lib/boilerpipe/extractors/canola_extractor.rb +16 -0
- data/lib/boilerpipe/extractors/keep_everything_extractor.rb +16 -0
- data/lib/boilerpipe/extractors/largest_content_extractor.rb +18 -0
- data/lib/boilerpipe/extractors/num_words_rules_extractor.rb +15 -0
- data/lib/boilerpipe/filters/canola_classifier.rb +28 -0
- data/lib/boilerpipe/filters/mark_everything_content_filter.rb +14 -0
- data/lib/boilerpipe/version.rb +1 -1
- metadata +8 -2
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA1:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 64511e224c1ec186c5c7cde62568dd0ba8cf1005
|
4
|
+
data.tar.gz: 455a8bac8eaadda62706d8f507854c6f6ccb6dba
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 8f2f769627e15ca0b8293122143304ef397ed789567cff61ceb48f45a657336442711c17d29d25fa8cfdf0dbcfd557030063e7b2a132550921ce601e2573e71f
|
7
|
+
data.tar.gz: 630f46c4d3a6e71933be0ba5f3bc98dc97fd34d10fb05d9e656e8f5f2f20fb4c28197122ef349398c79de7e57c795721307a5dadc7b4a796f907b2904d4816d0
|
data/CHANGELOG.md
CHANGED
data/README.md
CHANGED
@@ -10,13 +10,19 @@ I saw other gems making use of boilerpipe via the [free api](http://boilerpipe-w
|
|
10
10
|
|
11
11
|
This solution works great if you're using Jruby but I wanted a pure ruby solution to use on MRI. Open vim - start coding...
|
12
12
|
|
13
|
-
|
13
|
+
# TLDR
|
14
|
+
|
15
|
+
Just use either ArticleExtractor, DefaultExtractor or KeepEverythingExtractor - try out the others when you feel like experimenting...
|
14
16
|
|
15
17
|
Presently the follow Extractors are implemented
|
16
18
|
* [x] ArticleExtractor
|
19
|
+
* [ ] ArticleSentenceExtractor
|
20
|
+
* [x] CanolaExtractor
|
17
21
|
* [x] DefaultExtractor
|
18
|
-
* [
|
19
|
-
* [ ]
|
22
|
+
* [x] KeepEverythingExtractor
|
23
|
+
* [ ] KeepEverythingWithMinKWordsExtractor
|
24
|
+
* [x] LargestContentExtractor
|
25
|
+
* [x] NumWordsRulesExtractor
|
20
26
|
|
21
27
|
[![CircleCI](https://circleci.com/gh/gregors/boilerpipe-ruby/tree/master.svg?style=shield)](https://circleci.com/gh/gregors/boilerpipe-ruby/tree/master)
|
22
28
|
|
@@ -44,10 +50,18 @@ Or install it yourself as:
|
|
44
50
|
> require 'open-uri'
|
45
51
|
=> true
|
46
52
|
> content = open('https://blog.carbonfive.com/2017/08/28/always-squash-and-rebase-your-git-commits/').read; true;
|
47
|
-
|
53
|
+
|
54
|
+
> Boilerpipe::Extractors::ArticleExtractor.text(content).slice(0..40)
|
48
55
|
=> "Always Squash and Rebase your Git Commits"
|
49
|
-
|
56
|
+
|
57
|
+
> Boilerpipe::Extractors::DefaultExtractor.text(content).slice(0..40)
|
50
58
|
=> "Posted on\nWhat is the squash rebase workf"
|
59
|
+
|
60
|
+
> Boilerpipe::Extractors::LargestContentExtractor.text(content).slice(0, 40)
|
61
|
+
=> "git push origin master\nWhy should you ad"
|
62
|
+
|
63
|
+
> Boilerpipe::Extractors::KeepEverythingExtractor.text(content).slice(0..40)
|
64
|
+
=> "Toggle Navigation\nCarbon Five\nAbout\nWork\n"
|
51
65
|
|
52
66
|
## Development
|
53
67
|
|
data/lib/boilerpipe.rb
CHANGED
@@ -6,10 +6,15 @@ require 'boilerpipe/document/text_document'
|
|
6
6
|
require 'boilerpipe/document/text_block'
|
7
7
|
|
8
8
|
require 'boilerpipe/extractors/article_extractor'
|
9
|
+
require 'boilerpipe/extractors/canola_extractor'
|
9
10
|
require 'boilerpipe/extractors/default_extractor'
|
11
|
+
require 'boilerpipe/extractors/keep_everything_extractor'
|
12
|
+
require 'boilerpipe/extractors/largest_content_extractor'
|
13
|
+
require 'boilerpipe/extractors/num_words_rules_extractor'
|
10
14
|
|
11
15
|
require 'boilerpipe/filters/block_proximity_fusion'
|
12
16
|
require 'boilerpipe/filters/boilerplate_block_filter'
|
17
|
+
require 'boilerpipe/filters/canola_classifier'
|
13
18
|
require 'boilerpipe/filters/density_rules_classifier'
|
14
19
|
require 'boilerpipe/filters/document_title_match_classifier'
|
15
20
|
require 'boilerpipe/filters/expand_title_to_content_filter'
|
@@ -18,6 +23,7 @@ require 'boilerpipe/filters/ignore_blocks_after_content_filter'
|
|
18
23
|
require 'boilerpipe/filters/keep_largest_block_filter'
|
19
24
|
require 'boilerpipe/filters/large_block_same_tag_level_to_content_filter'
|
20
25
|
require 'boilerpipe/filters/list_at_end_filter'
|
26
|
+
require 'boilerpipe/filters/mark_everything_content_filter'
|
21
27
|
require 'boilerpipe/filters/num_words_rules_classifier'
|
22
28
|
require 'boilerpipe/filters/simple_block_fusion_processor'
|
23
29
|
require 'boilerpipe/filters/terminating_blocks_finder'
|
@@ -0,0 +1,16 @@
|
|
1
|
+
module Boilerpipe::Extractors
|
2
|
+
class CanolaExtractor
|
3
|
+
|
4
|
+
def self.text(contents)
|
5
|
+
doc = ::Boilerpipe::SAX::BoilerpipeHTMLParser.parse(contents)
|
6
|
+
::Boilerpipe::Extractors::CanolaExtractor.process doc
|
7
|
+
doc.content
|
8
|
+
end
|
9
|
+
|
10
|
+
def self.process(doc)
|
11
|
+
::Boilerpipe::Filters::CanolaClassifier.process doc
|
12
|
+
|
13
|
+
doc
|
14
|
+
end
|
15
|
+
end
|
16
|
+
end
|
@@ -0,0 +1,16 @@
|
|
1
|
+
# Marks all blocks as content.
|
2
|
+
|
3
|
+
module Boilerpipe::Extractors
|
4
|
+
class KeepEverythingExtractor
|
5
|
+
def self.text(contents)
|
6
|
+
doc = ::Boilerpipe::SAX::BoilerpipeHTMLParser.parse(contents)
|
7
|
+
::Boilerpipe::Extractors::KeepEverythingExtractor.process doc
|
8
|
+
doc.content
|
9
|
+
end
|
10
|
+
|
11
|
+
def self.process(doc)
|
12
|
+
::Boilerpipe::Filters::MarkEverythingContentFilter.process doc
|
13
|
+
doc
|
14
|
+
end
|
15
|
+
end
|
16
|
+
end
|
@@ -0,0 +1,18 @@
|
|
1
|
+
module Boilerpipe::Extractors
|
2
|
+
class LargestContentExtractor
|
3
|
+
def self.text(contents)
|
4
|
+
doc = ::Boilerpipe::SAX::BoilerpipeHTMLParser.parse(contents)
|
5
|
+
::Boilerpipe::Extractors::LargestContentExtractor.process doc
|
6
|
+
doc.content
|
7
|
+
end
|
8
|
+
|
9
|
+
def self.process(doc)
|
10
|
+
filters = ::Boilerpipe::Filters
|
11
|
+
filters::NumWordsRulesClassifier.process doc
|
12
|
+
filters::BlockProximityFusion::MAX_DISTANCE_1.process doc
|
13
|
+
filters::KeepLargestBlockFilter::INSTANCE.process doc
|
14
|
+
|
15
|
+
doc
|
16
|
+
end
|
17
|
+
end
|
18
|
+
end
|
@@ -0,0 +1,15 @@
|
|
1
|
+
module Boilerpipe::Extractors
|
2
|
+
class NumWordsRulesExtractor
|
3
|
+
|
4
|
+
def self.text(contents)
|
5
|
+
doc = ::Boilerpipe::SAX::BoilerpipeHTMLParser.parse(contents)
|
6
|
+
::Boilerpipe::Extractors::NumWordsRulesExtractor.process doc
|
7
|
+
doc.content
|
8
|
+
end
|
9
|
+
|
10
|
+
def self.process(doc)
|
11
|
+
::Boilerpipe::Filters::NumWordsRulesClassifier.process doc
|
12
|
+
doc
|
13
|
+
end
|
14
|
+
end
|
15
|
+
end
|
@@ -0,0 +1,28 @@
|
|
1
|
+
# A full-text extractor trained on http://krdwrd.org/
|
2
|
+
# https://krdwrd.org/trac/attachment/wiki/Corpora/Canola/CANOLA.pdf
|
3
|
+
# Works well with SimpleEstimator, too.
|
4
|
+
|
5
|
+
module Boilerpipe::Filters
|
6
|
+
class CanolaClassifier
|
7
|
+
|
8
|
+
def self.process(doc)
|
9
|
+
return doc if doc.text_blocks.size < 1
|
10
|
+
|
11
|
+
empty = Boilerpipe::Document::TextBlock.empty_start
|
12
|
+
text_blocks = [empty] + doc.text_blocks + [empty]
|
13
|
+
|
14
|
+
text_blocks.each_cons(3) do |slice|
|
15
|
+
prev, current, nxt = *slice
|
16
|
+
current.content = classify(prev, current, nxt)
|
17
|
+
end
|
18
|
+
|
19
|
+
doc
|
20
|
+
end
|
21
|
+
|
22
|
+
def self.classify(prev, current, nxt)
|
23
|
+
current.link_density > 0 && nxt.num_words > 11 \
|
24
|
+
|| current.num_words > 19 \
|
25
|
+
|| nxt.num_words > 6 && nxt.link_density == 0 && prev.link_density == 0 && ( current.num_words > 6 || prev.num_words > 7 || nxt.num_words > 19 )
|
26
|
+
end
|
27
|
+
end
|
28
|
+
end
|
data/lib/boilerpipe/version.rb
CHANGED
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: boilerpipe-ruby
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.
|
4
|
+
version: 0.3.0
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Gregory Ostermayr
|
8
8
|
autorequire:
|
9
9
|
bindir: exe
|
10
10
|
cert_chain: []
|
11
|
-
date: 2017-09-
|
11
|
+
date: 2017-09-12 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: bundler
|
@@ -103,9 +103,14 @@ files:
|
|
103
103
|
- lib/boilerpipe/document/text_document.rb
|
104
104
|
- lib/boilerpipe/errors.rb
|
105
105
|
- lib/boilerpipe/extractors/article_extractor.rb
|
106
|
+
- lib/boilerpipe/extractors/canola_extractor.rb
|
106
107
|
- lib/boilerpipe/extractors/default_extractor.rb
|
108
|
+
- lib/boilerpipe/extractors/keep_everything_extractor.rb
|
109
|
+
- lib/boilerpipe/extractors/largest_content_extractor.rb
|
110
|
+
- lib/boilerpipe/extractors/num_words_rules_extractor.rb
|
107
111
|
- lib/boilerpipe/filters/block_proximity_fusion.rb
|
108
112
|
- lib/boilerpipe/filters/boilerplate_block_filter.rb
|
113
|
+
- lib/boilerpipe/filters/canola_classifier.rb
|
109
114
|
- lib/boilerpipe/filters/density_rules_classifier.rb
|
110
115
|
- lib/boilerpipe/filters/document_title_match_classifier.rb
|
111
116
|
- lib/boilerpipe/filters/expand_title_to_content_filter.rb
|
@@ -114,6 +119,7 @@ files:
|
|
114
119
|
- lib/boilerpipe/filters/keep_largest_block_filter.rb
|
115
120
|
- lib/boilerpipe/filters/large_block_same_tag_level_to_content_filter.rb
|
116
121
|
- lib/boilerpipe/filters/list_at_end_filter.rb
|
122
|
+
- lib/boilerpipe/filters/mark_everything_content_filter.rb
|
117
123
|
- lib/boilerpipe/filters/num_words_rules_classifier.rb
|
118
124
|
- lib/boilerpipe/filters/simple_block_fusion_processor.rb
|
119
125
|
- lib/boilerpipe/filters/terminating_blocks_finder.rb
|