parsefile 0.0.4 → 0.0.5
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/lib/ocrfile.rb +27 -4
- data/lib/parsefile.rb +9 -2
- metadata +3 -2
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA1:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: c9b7c462a9aee22375232c5d0a2533cfae0ea46b
|
4
|
+
data.tar.gz: d9aba7f3809b24b4d21cceee8f9d55e957e4128c
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: d172ebfaa962e386490b1012a929f3c20d19f9926dbecfe59defeff19202932d2744b5bb1f1e4bca7acf7956563bdcb14a991092293c3aac5a5429041b130e0b
|
7
|
+
data.tar.gz: da667040f692c10b9da0098a52abcb6e7b972a004d0e84313cc84544b930022687a0bbab18de8cdbecc07d7aa535d5ea7e4494eecfb5db65a9599b336cbc8d0a
|
data/lib/ocrfile.rb
CHANGED
@@ -1,13 +1,15 @@
|
|
1
1
|
require 'fileutils'
|
2
2
|
require 'docsplit'
|
3
3
|
require 'curb'
|
4
|
+
require 'mimemagic'
|
4
5
|
|
5
6
|
class OCRFile
|
6
|
-
def initialize(file, input_dir, output_dir, rel_path)
|
7
|
+
def initialize(file, input_dir, output_dir, rel_path, tika)
|
7
8
|
@path = file
|
8
9
|
@input_dir = input_dir
|
9
10
|
@output_dir = output_dir
|
10
11
|
@rel_path = rel_path
|
12
|
+
@tika = tika
|
11
13
|
@text = ""
|
12
14
|
end
|
13
15
|
|
@@ -19,7 +21,11 @@ class OCRFile
|
|
19
21
|
elsif @path.include?(".pdf")
|
20
22
|
ocr_pdf
|
21
23
|
else
|
22
|
-
|
24
|
+
if @tika
|
25
|
+
give_me_text_local
|
26
|
+
else
|
27
|
+
give_me_text
|
28
|
+
end
|
23
29
|
end
|
24
30
|
rescue # Detect errors
|
25
31
|
binding.pry
|
@@ -44,10 +50,26 @@ class OCRFile
|
|
44
50
|
c = Curl::Easy.new("http://givemetext.okfnlabs.org/tika/tika/form")
|
45
51
|
c.multipart_form_post = true
|
46
52
|
c.http_post(Curl::PostField.file('file', @path))
|
47
|
-
|
53
|
+
|
54
|
+
@text = c.body_str
|
48
55
|
gotten_text_ok?(@text)
|
49
56
|
end
|
50
57
|
|
58
|
+
def give_me_text_local
|
59
|
+
c = Curl::Easy.new(@tika + "/tika")
|
60
|
+
# TODO: move this mime filtering to a higher global level
|
61
|
+
mime_magic = MimeMagic.by_path(@path)
|
62
|
+
file_data = File.read(@path)
|
63
|
+
c.headers['Content-Type'] = mime_magic.type
|
64
|
+
c.headers['Accept'] = "text/plain"
|
65
|
+
c.http_put(file_data)
|
66
|
+
|
67
|
+
#binding.pry
|
68
|
+
|
69
|
+
@text = c.body_str
|
70
|
+
gotten_text_ok?(@text)
|
71
|
+
end
|
72
|
+
|
51
73
|
# Checks if text was successfully extracted
|
52
74
|
def gotten_text_ok?(text)
|
53
75
|
throw :extraction_error if text.include?("java.io.IOException: Stream Closed")
|
@@ -74,7 +96,8 @@ class OCRFile
|
|
74
96
|
# Extract text and save
|
75
97
|
Docsplit.extract_text(docs_no_spaces, :output => base+'text')
|
76
98
|
text_files = Dir[base+'text/'+filename+'*']
|
77
|
-
sorted_text = text_files.sort_by {|f|
|
99
|
+
sorted_text = text_files.sort_by {|f|
|
100
|
+
f.split(filename).last.gsub("_", "").gsub(".txt", "").to_i }
|
78
101
|
sorted_text.each do |f|
|
79
102
|
@text += File.read(f)
|
80
103
|
end
|
data/lib/parsefile.rb
CHANGED
@@ -7,10 +7,17 @@ load 'ocrfile.rb'
|
|
7
7
|
load 'extractmetadata.rb'
|
8
8
|
|
9
9
|
class ParseFile
|
10
|
-
def initialize(file, input_dir, output_dir)
|
10
|
+
def initialize(file, input_dir, output_dir, tika)
|
11
11
|
@path = file
|
12
12
|
@input_dir = input_dir
|
13
13
|
@output_dir = output_dir
|
14
|
+
# Pass the url for a custom (or local) Tika server
|
15
|
+
# Else use OKFNs service over normal HTTP... ZOMG... O.o
|
16
|
+
if tika
|
17
|
+
@tika = tika
|
18
|
+
else
|
19
|
+
@tika = nil
|
20
|
+
end
|
14
21
|
end
|
15
22
|
|
16
23
|
# Parse the file
|
@@ -21,7 +28,7 @@ class ParseFile
|
|
21
28
|
@metadata = m.extract
|
22
29
|
|
23
30
|
# OCR File
|
24
|
-
o = OCRFile.new(@path, @input_dir, @output_dir, @metadata[:rel_path])
|
31
|
+
o = OCRFile.new(@path, @input_dir, @output_dir, @metadata[:rel_path], @tika)
|
25
32
|
@text = o.ocr
|
26
33
|
|
27
34
|
# Generate output and return
|
metadata
CHANGED
@@ -1,16 +1,17 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: parsefile
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.0.
|
4
|
+
version: 0.0.5
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- M. C. McGrath
|
8
|
+
- Brennan Novak
|
8
9
|
autorequire:
|
9
10
|
bindir: bin
|
10
11
|
cert_chain: []
|
11
12
|
date: 2016-05-16 00:00:00.000000000 Z
|
12
13
|
dependencies: []
|
13
|
-
description: OCR file and extract metadata
|
14
|
+
description: OCR file and extract metadata using Apache Tika and Tesseract
|
14
15
|
email: shidash@shidash.com
|
15
16
|
executables: []
|
16
17
|
extensions: []
|