pdftdx 0.2.0 → 0.3.1
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/Gemfile.lock +1 -1
- data/lib/pdftdx/parser.rb +3 -1
- data/lib/pdftdx/version.rb +1 -1
- metadata +1 -2
- data/bin/pdftdx +0 -26
checksums.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA1:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: b6ac984f258036c4d20985529d0fdcda9fd3254f
|
|
4
|
+
data.tar.gz: 9b325fba42e742f0521317f4202a8b2a783114da
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: 0ee08da850b0cb3ee593c7bbcbb0dd4f0d1244619e512018a83964fa103797d108ba2a4f5eca28f8e19ff2901c2d26d20bf36ca26d3f4787cd20a6286ef50513
|
|
7
|
+
data.tar.gz: 8d230f85707d4820ce341222e95a6f71fb553a3d7d57b7622a70b8668a966f8e4626e37538a2fe563855c0c3c50cdd846563ba3f6408318d0c39879c9ab2d710
|
data/Gemfile.lock
CHANGED
data/lib/pdftdx/parser.rb
CHANGED
|
@@ -59,6 +59,8 @@ module PDFTDX
|
|
|
59
59
|
|
|
60
60
|
# DEBUG
|
|
61
61
|
puts "=============> #{table}"
|
|
62
|
+
|
|
63
|
+
table
|
|
62
64
|
end
|
|
63
65
|
|
|
64
66
|
# HTML Filter
|
|
@@ -74,7 +76,7 @@ module PDFTDX
|
|
|
74
76
|
|
|
75
77
|
# Collect & Process File Data
|
|
76
78
|
off = 0
|
|
77
|
-
|
|
79
|
+
process_data page_data.collect { |_idx, page| off = off + PAGE_OFF; page.select { |l| LINE_REGEX =~ l }.collect { |l| LINE_REGEX.match l }.collect { |d| { top: off + d[1].to_i, left: d[2].to_i, data: hfilter(coder.decode(d[3])) } } }.flatten
|
|
78
80
|
end
|
|
79
81
|
end
|
|
80
82
|
end
|
data/lib/pdftdx/version.rb
CHANGED
metadata
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
|
2
2
|
name: pdftdx
|
|
3
3
|
version: !ruby/object:Gem::Version
|
|
4
|
-
version: 0.
|
|
4
|
+
version: 0.3.1
|
|
5
5
|
platform: ruby
|
|
6
6
|
authors:
|
|
7
7
|
- Eresse
|
|
@@ -72,7 +72,6 @@ files:
|
|
|
72
72
|
- LICENSE.txt
|
|
73
73
|
- README.md
|
|
74
74
|
- Rakefile
|
|
75
|
-
- bin/pdftdx
|
|
76
75
|
- lib/pdftdx.rb
|
|
77
76
|
- lib/pdftdx/parser.rb
|
|
78
77
|
- lib/pdftdx/version.rb
|
data/bin/pdftdx
DELETED
|
@@ -1,26 +0,0 @@
|
|
|
1
|
-
#!/usr/bin/env ruby
|
|
2
|
-
|
|
3
|
-
# PDF Table Data Extractor
|
|
4
|
-
# by Eresse <eresse@eresse.net>
|
|
5
|
-
|
|
6
|
-
# Internal Includes
|
|
7
|
-
require 'pdftdx'
|
|
8
|
-
|
|
9
|
-
# Usage
|
|
10
|
-
def usage
|
|
11
|
-
puts "Usage: #{$0} <PDF_FILE>"
|
|
12
|
-
exit 1
|
|
13
|
-
end
|
|
14
|
-
|
|
15
|
-
# Main
|
|
16
|
-
def main args
|
|
17
|
-
|
|
18
|
-
# Check Args
|
|
19
|
-
usage unless args.length == 1
|
|
20
|
-
|
|
21
|
-
# Extract Data from provided PDF File\
|
|
22
|
-
PDFTDX.extract_data args[0]
|
|
23
|
-
end
|
|
24
|
-
|
|
25
|
-
# Call Main
|
|
26
|
-
main ARGV
|