easy_captcha_solver 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/easy_captcha_solver.rb +40 -41
- metadata +1 -1
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA1:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 0e812e7f2b327ace5217bdb9f41d62b36d20dafc
|
4
|
+
data.tar.gz: 25e4e3575a6a4e7cb8e0d0c9fa592780b18e8c41
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: b1a34ef70f02a8324523be85414bcce4c0a6f68caa5442d32dc6b98597fb37f7059820f970d64b08829d4d32c8d15d6c0d61711e00c9bfef570728c3fe773f05
|
7
|
+
data.tar.gz: 2bb91e8f255736e6b4703dc82b71761b632f11336356d409c54ae692b57558d64b02cfea9bcdd8ff94fb220c5add6b140f954299c8778b55eeafe99f87e93020
|
data/lib/easy_captcha_solver.rb
CHANGED
@@ -5,59 +5,58 @@ class EasyCaptchaSolver
|
|
5
5
|
attr_reader :captcha
|
6
6
|
|
7
7
|
def initialize ( options = {} )
|
8
|
-
|
9
|
-
|
10
|
-
agent.user_agent_alias = 'Mac Safari'
|
11
|
-
}
|
8
|
+
image = options[:image_path] if options[:image_path]
|
9
|
+
image = get_captcha_image(options[:image_url]) if options[:image_url]
|
12
10
|
|
13
|
-
|
11
|
+
throw Exception.new "Image path or image URL must be provided.
|
12
|
+
Example: easy_c = EasyCaptcha.new( image_url: 'http://www.example.com/captcha' )
|
13
|
+
or easy_c = EasyCaptcha.new( image_path: './captcha.jpg' )" unless image
|
14
14
|
|
15
|
-
#
|
16
|
-
if options[:image_url]
|
17
|
-
|
18
|
-
|
19
|
-
# Guess image extension and rename tmp file
|
20
|
-
@image = "./tmp_image.#{get_image_extension(image)}"
|
21
|
-
File.rename( image, @image)
|
22
|
-
end
|
15
|
+
# Try to solve the captcha and delete temp img if necessary
|
16
|
+
solve_captcha(image) ensure File.delete(image) if options[:image_url] && File.exist?(image)
|
17
|
+
end
|
23
18
|
|
24
|
-
|
25
|
-
|
26
|
-
|
19
|
+
private
|
20
|
+
def get_captcha_image( image_url )
|
21
|
+
agent = Mechanize.new { |agent|
|
22
|
+
agent.user_agent_alias = 'Mac Safari'
|
23
|
+
}
|
27
24
|
|
28
|
-
|
25
|
+
# Save a file instead of trying to solve the captcha from memory
|
26
|
+
# because of tesseract limitations with .png images
|
27
|
+
image = agent.get(image_url).save! "./tmp_image"
|
28
|
+
image_extension = "./tmp_image.#{get_image_extension(image)}"
|
29
|
+
File.rename( image, image_extension)
|
29
30
|
|
31
|
+
image_extension
|
30
32
|
end
|
31
33
|
|
32
|
-
|
33
|
-
|
34
|
-
def solve_captcha()
|
35
|
-
# Solve captcha using Tesseract-ocr
|
34
|
+
def solve_captcha(image)
|
36
35
|
e = Tesseract::Engine.new {|e|
|
37
36
|
e.language = :eng
|
38
37
|
e.blacklist = '|'
|
39
38
|
}
|
40
|
-
|
41
|
-
@captcha = e.text_for(@image).strip # => 'ABC'
|
39
|
+
@captcha = e.text_for(image).strip # => 'ABC'
|
42
40
|
end
|
43
|
-
end
|
44
41
|
|
45
|
-
def get_image_extension(local_file_path)
|
46
|
-
|
47
|
-
|
48
|
-
|
49
|
-
|
50
|
-
|
51
|
-
|
52
|
-
|
53
|
-
|
54
|
-
|
55
|
-
|
56
|
-
|
57
|
-
|
58
|
-
|
59
|
-
|
60
|
-
|
61
|
-
|
42
|
+
def get_image_extension(local_file_path)
|
43
|
+
png = Regexp.new("\x89PNG".force_encoding("binary"))
|
44
|
+
jpg = Regexp.new("\xff\xd8\xff\xe0\x00\x10JFIF".force_encoding("binary"))
|
45
|
+
jpg2 = Regexp.new("\xff\xd8\xff\xe1(.*){2}Exif".force_encoding("binary"))
|
46
|
+
case IO.read(local_file_path, 10)
|
47
|
+
when /^GIF8/
|
48
|
+
'gif'
|
49
|
+
when /^#{png}/
|
50
|
+
'png'
|
51
|
+
when /^#{jpg}/
|
52
|
+
'jpg'
|
53
|
+
when /^#{jpg2}/
|
54
|
+
'jpg'
|
55
|
+
else
|
56
|
+
mime_type = `file #{local_file_path} --mime-type`.gsub("\n", '') # Works on linux and mac
|
57
|
+
raise UnprocessableEntity, "unknown file type" if !mime_type
|
58
|
+
mime_type.split(':')[1].split('/')[1].gsub('x-', '').gsub(/jpeg/, 'jpg').gsub(/text/, 'txt').gsub(/x-/, '')
|
59
|
+
end
|
62
60
|
end
|
61
|
+
|
63
62
|
end
|