konjac 0.1 → 0.1.1
Sign up to get free protection for your applications and to get access to all the features.
- data/lib/konjac/version.rb +1 -1
- data/lib/konjac/word.rb +89 -7
- metadata +15 -15
data/lib/konjac/version.rb
CHANGED
data/lib/konjac/word.rb
CHANGED
@@ -46,26 +46,108 @@ module Konjac
|
|
46
46
|
sub_files = Dir.glob(File.expand_path(file))
|
47
47
|
sub_files.each do |sub_file|
|
48
48
|
# Build a list of all the paths we're working with
|
49
|
-
dirname
|
50
|
-
basename
|
51
|
-
xml_path
|
52
|
-
|
49
|
+
dirname = File.dirname(sub_file)
|
50
|
+
basename = File.basename(sub_file, ".*")
|
51
|
+
xml_path = "#{dirname}/#{basename}_orig.xml"
|
52
|
+
clean_path = "#{dirname}/#{basename}.xml"
|
53
|
+
tags_path = "#{dirname}/#{basename}.tags"
|
53
54
|
|
54
55
|
# Unzip the DOCX's word/document.xml file and pipe the output into
|
55
56
|
# an XML with the same base name as the DOCX
|
56
57
|
system "unzip -p #{sub_file} word/document.xml > #{xml_path}"
|
57
58
|
|
58
59
|
# Read in the XML file and extract the content from each <w:t> tag
|
59
|
-
|
60
|
+
cleaner = Nokogiri::XML(File.read(xml_path))
|
60
61
|
File.open(tags_path, "w") do |tags_file|
|
61
|
-
|
62
|
+
# Remove all grammar and spellcheck tags
|
63
|
+
cleaner.xpath("//w:proofErr").remove
|
64
|
+
|
65
|
+
nodes = cleaner.xpath("//w:r")
|
66
|
+
prev = nil
|
67
|
+
nodes.each do |node|
|
68
|
+
unless prev.nil?
|
69
|
+
if (prev.next_sibling == node) && compare_nodes(prev, node)
|
70
|
+
begin
|
71
|
+
node.at_xpath("w:t").content = prev.at_xpath("w:t").content +
|
72
|
+
node.at_xpath("w:t").content
|
73
|
+
prev.remove
|
74
|
+
rescue
|
75
|
+
end
|
76
|
+
end
|
77
|
+
end
|
78
|
+
|
79
|
+
prev = node
|
80
|
+
end
|
81
|
+
|
82
|
+
cleaner.xpath("//w:t").each do |node|
|
62
83
|
tags_file.puts node.content
|
63
84
|
end
|
64
85
|
end
|
86
|
+
|
87
|
+
File.open(clean_path, "w") do |xml|
|
88
|
+
xml.puts cleaner.to_xml
|
89
|
+
end
|
90
|
+
end
|
91
|
+
end
|
92
|
+
end
|
93
|
+
|
94
|
+
private
|
95
|
+
|
96
|
+
# Performs a comparison between two nodes and accepts them as equivalent
|
97
|
+
# if the differences are very minor
|
98
|
+
def compare_nodes(a, b)
|
99
|
+
c = clean_hash(xml_node_to_hash(a))
|
100
|
+
d = clean_hash(xml_node_to_hash(b))
|
101
|
+
c == d
|
102
|
+
end
|
103
|
+
|
104
|
+
def xml_node_to_hash(node)
|
105
|
+
# If we are at the root of the document, start the hash
|
106
|
+
if node.element?
|
107
|
+
result_hash = {}
|
108
|
+
if node.attributes != {}
|
109
|
+
result_hash[:attributes] = {}
|
110
|
+
node.attributes.keys.each do |key|
|
111
|
+
result_hash[:attributes][node.attributes[key].name.to_sym] = prepare(node.attributes[key].value)
|
112
|
+
end
|
113
|
+
end
|
114
|
+
if node.children.size > 0
|
115
|
+
node.children.each do |child|
|
116
|
+
result = xml_node_to_hash(child)
|
117
|
+
|
118
|
+
if child.name == "text"
|
119
|
+
unless child.next_sibling || child.previous_sibling
|
120
|
+
return prepare(result)
|
121
|
+
end
|
122
|
+
elsif result_hash[child.name.to_sym]
|
123
|
+
if result_hash[child.name.to_sym].is_a?(Array)
|
124
|
+
result_hash[child.name.to_sym] << prepare(result)
|
125
|
+
else
|
126
|
+
result_hash[child.name.to_sym] = [result_hash[child.name.to_sym]] << prepare(result)
|
127
|
+
end
|
128
|
+
else
|
129
|
+
result_hash[child.name.to_sym] = prepare(result)
|
130
|
+
end
|
131
|
+
end
|
132
|
+
|
133
|
+
return result_hash
|
134
|
+
else
|
135
|
+
return result_hash
|
65
136
|
end
|
137
|
+
else
|
138
|
+
return prepare(node.content.to_s)
|
66
139
|
end
|
67
140
|
end
|
141
|
+
|
142
|
+
def prepare(data)
|
143
|
+
(data.class == String && data.to_i.to_s == data) ? data.to_i : data
|
144
|
+
end
|
145
|
+
|
146
|
+
# Delete extraneous attributes for comparison
|
147
|
+
def clean_hash(hash)
|
148
|
+
hash.delete :t
|
149
|
+
hash[:rPr][:rFonts][:attributes].delete :hint
|
150
|
+
end
|
68
151
|
end
|
69
152
|
end
|
70
153
|
end
|
71
|
-
# lol
|
metadata
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: konjac
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version:
|
4
|
+
version: 0.1.1
|
5
5
|
prerelease:
|
6
6
|
platform: ruby
|
7
7
|
authors:
|
@@ -13,7 +13,7 @@ date: 2012-01-12 00:00:00.000000000 Z
|
|
13
13
|
dependencies:
|
14
14
|
- !ruby/object:Gem::Dependency
|
15
15
|
name: nokogiri
|
16
|
-
requirement: &
|
16
|
+
requirement: &70314541350300 !ruby/object:Gem::Requirement
|
17
17
|
none: false
|
18
18
|
requirements:
|
19
19
|
- - ! '>='
|
@@ -21,10 +21,10 @@ dependencies:
|
|
21
21
|
version: '0'
|
22
22
|
type: :runtime
|
23
23
|
prerelease: false
|
24
|
-
version_requirements: *
|
24
|
+
version_requirements: *70314541350300
|
25
25
|
- !ruby/object:Gem::Dependency
|
26
26
|
name: bundler
|
27
|
-
requirement: &
|
27
|
+
requirement: &70314541347900 !ruby/object:Gem::Requirement
|
28
28
|
none: false
|
29
29
|
requirements:
|
30
30
|
- - ! '>='
|
@@ -32,10 +32,10 @@ dependencies:
|
|
32
32
|
version: '0'
|
33
33
|
type: :development
|
34
34
|
prerelease: false
|
35
|
-
version_requirements: *
|
35
|
+
version_requirements: *70314541347900
|
36
36
|
- !ruby/object:Gem::Dependency
|
37
37
|
name: rspec
|
38
|
-
requirement: &
|
38
|
+
requirement: &70314541345520 !ruby/object:Gem::Requirement
|
39
39
|
none: false
|
40
40
|
requirements:
|
41
41
|
- - ! '>='
|
@@ -43,10 +43,10 @@ dependencies:
|
|
43
43
|
version: '0'
|
44
44
|
type: :development
|
45
45
|
prerelease: false
|
46
|
-
version_requirements: *
|
46
|
+
version_requirements: *70314541345520
|
47
47
|
- !ruby/object:Gem::Dependency
|
48
48
|
name: sdoc
|
49
|
-
requirement: &
|
49
|
+
requirement: &70314541340780 !ruby/object:Gem::Requirement
|
50
50
|
none: false
|
51
51
|
requirements:
|
52
52
|
- - ! '>='
|
@@ -54,10 +54,10 @@ dependencies:
|
|
54
54
|
version: '0'
|
55
55
|
type: :development
|
56
56
|
prerelease: false
|
57
|
-
version_requirements: *
|
57
|
+
version_requirements: *70314541340780
|
58
58
|
- !ruby/object:Gem::Dependency
|
59
59
|
name: autotest
|
60
|
-
requirement: &
|
60
|
+
requirement: &70314541332480 !ruby/object:Gem::Requirement
|
61
61
|
none: false
|
62
62
|
requirements:
|
63
63
|
- - ! '>='
|
@@ -65,10 +65,10 @@ dependencies:
|
|
65
65
|
version: '0'
|
66
66
|
type: :development
|
67
67
|
prerelease: false
|
68
|
-
version_requirements: *
|
68
|
+
version_requirements: *70314541332480
|
69
69
|
- !ruby/object:Gem::Dependency
|
70
70
|
name: autotest-fsevent
|
71
|
-
requirement: &
|
71
|
+
requirement: &70314541323420 !ruby/object:Gem::Requirement
|
72
72
|
none: false
|
73
73
|
requirements:
|
74
74
|
- - ! '>='
|
@@ -76,10 +76,10 @@ dependencies:
|
|
76
76
|
version: '0'
|
77
77
|
type: :development
|
78
78
|
prerelease: false
|
79
|
-
version_requirements: *
|
79
|
+
version_requirements: *70314541323420
|
80
80
|
- !ruby/object:Gem::Dependency
|
81
81
|
name: autotest-growl
|
82
|
-
requirement: &
|
82
|
+
requirement: &70314541318360 !ruby/object:Gem::Requirement
|
83
83
|
none: false
|
84
84
|
requirements:
|
85
85
|
- - ! '>='
|
@@ -87,7 +87,7 @@ dependencies:
|
|
87
87
|
version: '0'
|
88
88
|
type: :development
|
89
89
|
prerelease: false
|
90
|
-
version_requirements: *
|
90
|
+
version_requirements: *70314541318360
|
91
91
|
description: A Ruby command-line utility for translating files using a YAML wordlist
|
92
92
|
email:
|
93
93
|
- bryan.mckelvey@gmail.com
|