siuying-kindle-highlights 0.0.6
Sign up to get free protection for your applications and to get access to all the features.
- data/lib/kindle-highlights.rb +75 -0
- metadata +70 -0
|
@@ -0,0 +1,75 @@
|
|
|
1
|
+
require 'rubygems'
|
|
2
|
+
require 'mechanize'
|
|
3
|
+
require 'asin'
|
|
4
|
+
|
|
5
|
+
class KindleHighlight
|
|
6
|
+
attr_accessor :highlights
|
|
7
|
+
attr_accessor :next_url, :fetched
|
|
8
|
+
|
|
9
|
+
def initialize(email_address, password)
|
|
10
|
+
@highlights = Array.new
|
|
11
|
+
@agent = Mechanize.new
|
|
12
|
+
@fetched = false
|
|
13
|
+
|
|
14
|
+
page = @agent.get("https://www.amazon.com/ap/signin?openid.return_to=https%3A%2F%2Fkindle.amazon.com%3A443%2Fauthenticate%2Flogin_callback%3Fwctx%3D%252F&pageId=amzn_kindle&openid.mode=checkid_setup&openid.ns=http%3A%2F%2Fspecs.openid.net%2Fauth%2F2.0&openid.identity=http%3A%2F%2Fspecs.openid.net%2Fauth%2F2.0%2Fidentifier_select&openid.pape.max_auth_age=0&openid.assoc_handle=amzn_kindle&openid.claimed_id=http%3A%2F%2Fspecs.openid.net%2Fauth%2F2.0%2Fidentifier_select")
|
|
15
|
+
@amazon_form = page.form('signIn')
|
|
16
|
+
@amazon_form.email = email_address
|
|
17
|
+
@amazon_form.password = password
|
|
18
|
+
end
|
|
19
|
+
|
|
20
|
+
def scrape_highlights
|
|
21
|
+
if self.next_url.nil?
|
|
22
|
+
signin_submission = @agent.submit(@amazon_form)
|
|
23
|
+
highlights_page = @agent.click(signin_submission.link_with(:text => /Your Highlights/))
|
|
24
|
+
self.fetched = true
|
|
25
|
+
else
|
|
26
|
+
highlights_page = @agent.get(self.next_url)
|
|
27
|
+
end
|
|
28
|
+
|
|
29
|
+
new_highlights = Array.new
|
|
30
|
+
highlights_page.search(".//div[@class='highlightRow yourHighlight']").each do |h|
|
|
31
|
+
new_highlights << Highlight.new(h)
|
|
32
|
+
end
|
|
33
|
+
self.highlights = self.highlights + new_highlights
|
|
34
|
+
self.next_url = highlights_page.search("a#nextBookLink").attribute('href') rescue nil
|
|
35
|
+
new_highlights
|
|
36
|
+
end
|
|
37
|
+
|
|
38
|
+
def has_more?
|
|
39
|
+
!self.fetched || (self.fetched && !self.next_url.nil?)
|
|
40
|
+
end
|
|
41
|
+
end
|
|
42
|
+
|
|
43
|
+
class KindleHighlight::Highlight
|
|
44
|
+
include ASIN::Client
|
|
45
|
+
|
|
46
|
+
attr_accessor :annotation_id, :asin, :end_location, :note
|
|
47
|
+
attr_accessor :details_url, :image_url, :title, :author, :content
|
|
48
|
+
|
|
49
|
+
@@amazon_items = Hash.new
|
|
50
|
+
|
|
51
|
+
def initialize(highlight)
|
|
52
|
+
self.annotation_id = highlight.xpath("form/input[@id='annotation_id']").attribute("value").value
|
|
53
|
+
self.asin = highlight.xpath("p/span[@class='hidden asin']").text
|
|
54
|
+
self.content = highlight.xpath("span[@class='highlight']").text rescue ""
|
|
55
|
+
self.note = highlight.xpath("span[@class='noteContent']").text rescue ""
|
|
56
|
+
self.end_location = highlight.xpath("span[@class='end_location']").text
|
|
57
|
+
|
|
58
|
+
amazon_item = lookup_or_get_from_cache(self.asin)
|
|
59
|
+
self.title = amazon_item.title
|
|
60
|
+
self.author = amazon_item.raw.ItemAttributes.Author rescue nil
|
|
61
|
+
self.details_url = amazon_item.details_url
|
|
62
|
+
self.image_url = amazon_item.image_url
|
|
63
|
+
end
|
|
64
|
+
|
|
65
|
+
def lookup_or_get_from_cache(asin)
|
|
66
|
+
unless @@amazon_items.has_key?(asin)
|
|
67
|
+
@@amazon_items[asin] = lookup(asin).first
|
|
68
|
+
end
|
|
69
|
+
@@amazon_items[asin]
|
|
70
|
+
end
|
|
71
|
+
|
|
72
|
+
def to_s
|
|
73
|
+
"<Highlight##{annotation_id}>"
|
|
74
|
+
end
|
|
75
|
+
end
|
metadata
ADDED
|
@@ -0,0 +1,70 @@
|
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
|
2
|
+
name: siuying-kindle-highlights
|
|
3
|
+
version: !ruby/object:Gem::Version
|
|
4
|
+
version: 0.0.6
|
|
5
|
+
prerelease:
|
|
6
|
+
platform: ruby
|
|
7
|
+
authors:
|
|
8
|
+
- Eric Farkas
|
|
9
|
+
- Francis Chong
|
|
10
|
+
autorequire:
|
|
11
|
+
bindir: bin
|
|
12
|
+
cert_chain: []
|
|
13
|
+
date: 2012-10-15 00:00:00.000000000 Z
|
|
14
|
+
dependencies:
|
|
15
|
+
- !ruby/object:Gem::Dependency
|
|
16
|
+
name: mechanize
|
|
17
|
+
requirement: &70358233487240 !ruby/object:Gem::Requirement
|
|
18
|
+
none: false
|
|
19
|
+
requirements:
|
|
20
|
+
- - ! '>='
|
|
21
|
+
- !ruby/object:Gem::Version
|
|
22
|
+
version: 2.0.1
|
|
23
|
+
type: :runtime
|
|
24
|
+
prerelease: false
|
|
25
|
+
version_requirements: *70358233487240
|
|
26
|
+
- !ruby/object:Gem::Dependency
|
|
27
|
+
name: asin
|
|
28
|
+
requirement: &70358233437920 !ruby/object:Gem::Requirement
|
|
29
|
+
none: false
|
|
30
|
+
requirements:
|
|
31
|
+
- - ! '>='
|
|
32
|
+
- !ruby/object:Gem::Version
|
|
33
|
+
version: 1.0.0
|
|
34
|
+
type: :runtime
|
|
35
|
+
prerelease: false
|
|
36
|
+
version_requirements: *70358233437920
|
|
37
|
+
description: Until there is a Kindle API, this will suffice.
|
|
38
|
+
email:
|
|
39
|
+
- eric@prudentiadigital.com
|
|
40
|
+
- francis@ignition.hk
|
|
41
|
+
executables: []
|
|
42
|
+
extensions: []
|
|
43
|
+
extra_rdoc_files: []
|
|
44
|
+
files:
|
|
45
|
+
- lib/kindle-highlights.rb
|
|
46
|
+
homepage: https://github.com/siuying/kindle-highlights
|
|
47
|
+
licenses: []
|
|
48
|
+
post_install_message:
|
|
49
|
+
rdoc_options: []
|
|
50
|
+
require_paths:
|
|
51
|
+
- lib
|
|
52
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
|
53
|
+
none: false
|
|
54
|
+
requirements:
|
|
55
|
+
- - ! '>='
|
|
56
|
+
- !ruby/object:Gem::Version
|
|
57
|
+
version: '0'
|
|
58
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
|
59
|
+
none: false
|
|
60
|
+
requirements:
|
|
61
|
+
- - ! '>='
|
|
62
|
+
- !ruby/object:Gem::Version
|
|
63
|
+
version: '0'
|
|
64
|
+
requirements: []
|
|
65
|
+
rubyforge_project:
|
|
66
|
+
rubygems_version: 1.8.15
|
|
67
|
+
signing_key:
|
|
68
|
+
specification_version: 3
|
|
69
|
+
summary: Kindle highlights
|
|
70
|
+
test_files: []
|