doc2text 0.4.9 → 0.4.11
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/lib/doc2text/docx/docx.rb +12 -0
- data/lib/doc2text/xml_based_document_file.rb +4 -4
- metadata +2 -2
checksums.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA256:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: 3c1beb721d04a176cf3c6865d627d1db7ec4d3f0659ccffe0ecc856ce402a09e
|
|
4
|
+
data.tar.gz: a5248bc158bfc3bcac2cb776bb613c0c0c275155997aa71df292a6ecd6459782
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: cce97b69956004f57f17edcdd4276ee3da5d7c5174a4b48138c39705808ccc56c8caae6880f1b41fee4b02ade0b745b6ac0e7e4b43564d614f7e81523ff11a87
|
|
7
|
+
data.tar.gz: cba486e21727b50d6dbeed98374ccbeb5a3cd6494f8db0b4168a5c37612a4c9a13dba1c7986764179fc78657063b414879adff710812d501558ce367a8ac2069
|
data/lib/doc2text/docx/docx.rb
CHANGED
|
@@ -19,6 +19,18 @@ module Doc2Text
|
|
|
19
19
|
end
|
|
20
20
|
end
|
|
21
21
|
|
|
22
|
+
def parse_styles
|
|
23
|
+
styles_parser = Doc2Text::Odt::StylesParser.new
|
|
24
|
+
xml = Nokogiri::XML::SAX::Parser.new(styles_parser)
|
|
25
|
+
xml.parse open File.join('word', 'styles.xml')
|
|
26
|
+
styles_parser.xml_root
|
|
27
|
+
end
|
|
28
|
+
|
|
29
|
+
def parse(markdown)
|
|
30
|
+
parser = Nokogiri::XML::SAX::Parser.new(markdown)
|
|
31
|
+
parser.parse open File.join('word', 'document.xml')
|
|
32
|
+
end
|
|
33
|
+
|
|
22
34
|
def contains_extracted_files?
|
|
23
35
|
File.exist? File.join(extract_path, '[Content_Types].xml')
|
|
24
36
|
end
|
|
@@ -12,9 +12,9 @@ module Doc2Text
|
|
|
12
12
|
|zip_file|
|
|
13
13
|
Dir.mkdir(extract_path)
|
|
14
14
|
zip_file.each do |entry|
|
|
15
|
-
zipped_file_extract_path = File.join extract_path, entry.name
|
|
16
|
-
FileUtils.mkdir_p
|
|
17
|
-
zip_file.extract entry,
|
|
15
|
+
zipped_file_extract_path = File.join extract_path, File.dirname(entry.name)
|
|
16
|
+
FileUtils.mkdir_p zipped_file_extract_path
|
|
17
|
+
zip_file.extract entry, nil, destination_directory: extract_path
|
|
18
18
|
end
|
|
19
19
|
}
|
|
20
20
|
end
|
|
@@ -41,7 +41,7 @@ module Doc2Text
|
|
|
41
41
|
end
|
|
42
42
|
|
|
43
43
|
def extract_path
|
|
44
|
-
File.join File.dirname(@document_path), ".#{File.basename(@document_path)}_#{extract_extension}"
|
|
44
|
+
@extract_path ||= File.join File.dirname(@document_path), ".#{File.basename(@document_path)}_#{extract_extension}"
|
|
45
45
|
end
|
|
46
46
|
end
|
|
47
47
|
end
|
metadata
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
|
2
2
|
name: doc2text
|
|
3
3
|
version: !ruby/object:Gem::Version
|
|
4
|
-
version: 0.4.
|
|
4
|
+
version: 0.4.11
|
|
5
5
|
platform: ruby
|
|
6
6
|
authors:
|
|
7
7
|
- Valentin A.
|
|
@@ -78,7 +78,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
|
|
|
78
78
|
- !ruby/object:Gem::Version
|
|
79
79
|
version: '0'
|
|
80
80
|
requirements: []
|
|
81
|
-
rubygems_version:
|
|
81
|
+
rubygems_version: 3.6.7
|
|
82
82
|
specification_version: 4
|
|
83
83
|
summary: Translates odt to markdown
|
|
84
84
|
test_files: []
|