boilerpipe-ruby 0.3.0 → 0.4.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/CHANGELOG.md +5 -0
- data/README.md +4 -2
- data/lib/boilerpipe.rb +5 -0
- data/lib/boilerpipe/extractors/article_sentence_extractor.rb +17 -0
- data/lib/boilerpipe/extractors/keep_everything_with_k_min_words_extractor.rb +21 -0
- data/lib/boilerpipe/filters/min_clause_words_filter.rb +37 -0
- data/lib/boilerpipe/filters/min_words_filter.rb +16 -0
- data/lib/boilerpipe/filters/split_paragraph_blocks_filter.rb +40 -0
- data/lib/boilerpipe/sax/boilerpipe_html_parser.rb +2 -2
- data/lib/boilerpipe/version.rb +1 -1
- metadata +7 -2
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA1:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 68c2ea4ee42a6e1d76e85f7eaa3de9ca95f3a8d3
|
4
|
+
data.tar.gz: 42199704467cb7f20a8fff7c616be67bae1966e1
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 875da6a2ddfdf517509e3ebb7b9c804fadf4a372b812df42bce92fccb82eb8ca31283f11e764c6796c15997bcb8fddf24301d821f4e24a962f40ba6f973f6f17
|
7
|
+
data.tar.gz: 5dfec7323587057c64b931725df2aa5b53dc5a2fadaef52374619fe5798826c98c4e7ae8bb1ea9267b5db9f40249d5822752700e57a145b058231088249d1ac6
|
data/CHANGELOG.md
CHANGED
data/README.md
CHANGED
@@ -16,16 +16,18 @@ Just use either ArticleExtractor, DefaultExtractor or KeepEverythingExtractor -
|
|
16
16
|
|
17
17
|
Presently the follow Extractors are implemented
|
18
18
|
* [x] ArticleExtractor
|
19
|
-
* [
|
19
|
+
* [x] ArticleSentenceExtractor
|
20
20
|
* [x] CanolaExtractor
|
21
21
|
* [x] DefaultExtractor
|
22
22
|
* [x] KeepEverythingExtractor
|
23
|
-
* [
|
23
|
+
* [x] KeepEverythingWithMinKWordsExtractor
|
24
24
|
* [x] LargestContentExtractor
|
25
25
|
* [x] NumWordsRulesExtractor
|
26
26
|
|
27
27
|
[![CircleCI](https://circleci.com/gh/gregors/boilerpipe-ruby/tree/master.svg?style=shield)](https://circleci.com/gh/gregors/boilerpipe-ruby/tree/master)
|
28
28
|
|
29
|
+
[![Gem Version](https://badge.fury.io/rb/boilerpipe-ruby.svg)](https://badge.fury.io/rb/boilerpipe-ruby)
|
30
|
+
|
29
31
|
## Installation
|
30
32
|
|
31
33
|
Add this line to your application's Gemfile:
|
data/lib/boilerpipe.rb
CHANGED
@@ -6,9 +6,11 @@ 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/article_sentence_extractor'
|
9
10
|
require 'boilerpipe/extractors/canola_extractor'
|
10
11
|
require 'boilerpipe/extractors/default_extractor'
|
11
12
|
require 'boilerpipe/extractors/keep_everything_extractor'
|
13
|
+
require 'boilerpipe/extractors/keep_everything_with_k_min_words_extractor'
|
12
14
|
require 'boilerpipe/extractors/largest_content_extractor'
|
13
15
|
require 'boilerpipe/extractors/num_words_rules_extractor'
|
14
16
|
|
@@ -24,8 +26,11 @@ require 'boilerpipe/filters/keep_largest_block_filter'
|
|
24
26
|
require 'boilerpipe/filters/large_block_same_tag_level_to_content_filter'
|
25
27
|
require 'boilerpipe/filters/list_at_end_filter'
|
26
28
|
require 'boilerpipe/filters/mark_everything_content_filter'
|
29
|
+
require 'boilerpipe/filters/min_clause_words_filter'
|
30
|
+
require 'boilerpipe/filters/min_words_filter'
|
27
31
|
require 'boilerpipe/filters/num_words_rules_classifier'
|
28
32
|
require 'boilerpipe/filters/simple_block_fusion_processor'
|
33
|
+
require 'boilerpipe/filters/split_paragraph_blocks_filter'
|
29
34
|
require 'boilerpipe/filters/terminating_blocks_finder'
|
30
35
|
require 'boilerpipe/filters/trailing_headline_to_boilerplate_filter'
|
31
36
|
|
@@ -0,0 +1,17 @@
|
|
1
|
+
# A full-text extractor which is tuned towards extracting sentences from news articles.
|
2
|
+
|
3
|
+
module Boilerpipe::Extractors
|
4
|
+
class ArticleSentenceExtractor
|
5
|
+
def self.text(contents)
|
6
|
+
doc = ::Boilerpipe::SAX::BoilerpipeHTMLParser.parse(contents)
|
7
|
+
::Boilerpipe::Extractors::ArticleSentenceExtractor.process(doc)
|
8
|
+
doc.content
|
9
|
+
end
|
10
|
+
|
11
|
+
def self.process(doc)
|
12
|
+
::Boilerpipe::Extractors::ArticleExtractor.process doc
|
13
|
+
::Boilerpipe::Filters::SplitParagraphBlocksFilter.process doc
|
14
|
+
::Boilerpipe::Filters::MinClauseWordsFilter.process doc
|
15
|
+
end
|
16
|
+
end
|
17
|
+
end
|
@@ -0,0 +1,21 @@
|
|
1
|
+
|
2
|
+
# A full-text extractor which extracts the largest text component of a page.
|
3
|
+
# For news articles, it may perform better than the DefaultExtractor, but
|
4
|
+
# usually worse than ArticleExtractor.
|
5
|
+
|
6
|
+
module Boilerpipe::Extractors
|
7
|
+
class KeepEverythingWithKMinWordsExtractor
|
8
|
+
def self.text(min, contents)
|
9
|
+
doc = ::Boilerpipe::SAX::BoilerpipeHTMLParser.parse(contents)
|
10
|
+
::Boilerpipe::Extractors::KeepEverythingWithKMinWordsExtractor.process min, doc
|
11
|
+
doc.content
|
12
|
+
end
|
13
|
+
|
14
|
+
def self.process(min, doc)
|
15
|
+
::Boilerpipe::Filters::SimpleBlockFusionProcessor.process doc
|
16
|
+
::Boilerpipe::Filters::MarkEverythingContentFilter.process doc
|
17
|
+
::Boilerpipe::Filters::MinWordsFilter.process min, doc
|
18
|
+
doc
|
19
|
+
end
|
20
|
+
end
|
21
|
+
end
|
@@ -0,0 +1,37 @@
|
|
1
|
+
#
|
2
|
+
# Keeps only blocks that have at least one segment fragment ("clause") with at least k
|
3
|
+
# words (default: 5).
|
4
|
+
#
|
5
|
+
# NOTE: You might consider using the SplitParagraphBlocksFilter upstream.
|
6
|
+
#
|
7
|
+
# SplitParagraphBlocksFilter
|
8
|
+
|
9
|
+
module Boilerpipe::Filters
|
10
|
+
class MinClauseWordsFilter
|
11
|
+
|
12
|
+
def self.process(doc, min_words=5)
|
13
|
+
|
14
|
+
doc.text_blocks.each do |tb|
|
15
|
+
next if tb.is_not_content?
|
16
|
+
|
17
|
+
clause_delimiter = /[\p{L}\d \u00a0]+[\,.:;!?]+(?:[ \n\r]+|$)/
|
18
|
+
tb.text.scan(clause_delimiter).each do |possible_clause|
|
19
|
+
if is_clause? possible_clause
|
20
|
+
break
|
21
|
+
else
|
22
|
+
tb.content = false
|
23
|
+
end
|
24
|
+
end
|
25
|
+
end
|
26
|
+
|
27
|
+
doc
|
28
|
+
end
|
29
|
+
|
30
|
+
def self.is_clause?(text, min_words=5)
|
31
|
+
return false if text.nil?
|
32
|
+
whitespace = /[ \n\r]+/
|
33
|
+
text.scan(whitespace).size >= min_words
|
34
|
+
end
|
35
|
+
|
36
|
+
end
|
37
|
+
end
|
@@ -0,0 +1,16 @@
|
|
1
|
+
|
2
|
+
# Keeps only those content blocks which contain at least k words.
|
3
|
+
|
4
|
+
module Boilerpipe::Filters
|
5
|
+
class MinWordsFilter
|
6
|
+
|
7
|
+
def self.process(min_words, doc)
|
8
|
+
doc.text_blocks.each do |tb|
|
9
|
+
next if tb.is_not_content?
|
10
|
+
tb.content = false if tb.num_words < min_words
|
11
|
+
end
|
12
|
+
doc
|
13
|
+
end
|
14
|
+
|
15
|
+
end
|
16
|
+
end
|
@@ -0,0 +1,40 @@
|
|
1
|
+
|
2
|
+
# Splits TextBlocks at paragraph boundaries.
|
3
|
+
#
|
4
|
+
# NOTE: This is not fully supported (i.e., it will break highlighting support via
|
5
|
+
# #getContainedTextElements()), but this one probably is necessary for some other filters.
|
6
|
+
#
|
7
|
+
# see MinClauseWordsFilter
|
8
|
+
|
9
|
+
module Boilerpipe::Filters
|
10
|
+
class SplitParagraphBlocksFilter
|
11
|
+
|
12
|
+
def self.process(doc)
|
13
|
+
tbs = doc.text_blocks
|
14
|
+
new_blocks = []
|
15
|
+
changes = false
|
16
|
+
tbs.each do |tb|
|
17
|
+
paragraphs = tb.text.split(/[\n\r]+/)
|
18
|
+
|
19
|
+
if paragraphs.size < 2
|
20
|
+
new_blocks << tb
|
21
|
+
next
|
22
|
+
end
|
23
|
+
|
24
|
+
is_content = tb.is_content?
|
25
|
+
labels = tb.labels
|
26
|
+
paragraphs.each do |paragraph|
|
27
|
+
tbP = ::Boilerpipe::Document::TextBlock.new(paragraph)
|
28
|
+
tbP.content = is_content
|
29
|
+
tbP.add_labels(labels)
|
30
|
+
new_blocks << tbP
|
31
|
+
changes = true
|
32
|
+
end
|
33
|
+
end
|
34
|
+
|
35
|
+
doc.replace_text_blocks!(new_blocks) if changes
|
36
|
+
doc
|
37
|
+
end
|
38
|
+
|
39
|
+
end
|
40
|
+
end
|
@@ -4,11 +4,11 @@ module Boilerpipe::SAX
|
|
4
4
|
def self.parse(text)
|
5
5
|
|
6
6
|
#script bug - delete script tags
|
7
|
-
text
|
7
|
+
text.gsub!(/\<script>.+?<\/script>/i, '')
|
8
8
|
|
9
9
|
# nokogiri uses libxml for mri and nekohtml for jruby
|
10
10
|
# mri doesn't remove when missing the semicolon
|
11
|
-
text
|
11
|
+
text.gsub!(/( ) /, '\1; ')
|
12
12
|
|
13
13
|
|
14
14
|
# use nokogiri to fix any bad tags, errors - keep experimenting with this
|
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: 0.4.0
|
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: 2017-09-15 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: bundler
|
@@ -103,9 +103,11 @@ 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/article_sentence_extractor.rb
|
106
107
|
- lib/boilerpipe/extractors/canola_extractor.rb
|
107
108
|
- lib/boilerpipe/extractors/default_extractor.rb
|
108
109
|
- lib/boilerpipe/extractors/keep_everything_extractor.rb
|
110
|
+
- lib/boilerpipe/extractors/keep_everything_with_k_min_words_extractor.rb
|
109
111
|
- lib/boilerpipe/extractors/largest_content_extractor.rb
|
110
112
|
- lib/boilerpipe/extractors/num_words_rules_extractor.rb
|
111
113
|
- lib/boilerpipe/filters/block_proximity_fusion.rb
|
@@ -120,8 +122,11 @@ files:
|
|
120
122
|
- lib/boilerpipe/filters/large_block_same_tag_level_to_content_filter.rb
|
121
123
|
- lib/boilerpipe/filters/list_at_end_filter.rb
|
122
124
|
- lib/boilerpipe/filters/mark_everything_content_filter.rb
|
125
|
+
- lib/boilerpipe/filters/min_clause_words_filter.rb
|
126
|
+
- lib/boilerpipe/filters/min_words_filter.rb
|
123
127
|
- lib/boilerpipe/filters/num_words_rules_classifier.rb
|
124
128
|
- lib/boilerpipe/filters/simple_block_fusion_processor.rb
|
129
|
+
- lib/boilerpipe/filters/split_paragraph_blocks_filter.rb
|
125
130
|
- lib/boilerpipe/filters/terminating_blocks_finder.rb
|
126
131
|
- lib/boilerpipe/filters/trailing_headline_to_boilerplate_filter.rb
|
127
132
|
- lib/boilerpipe/labels/default.rb
|