Class: TomosiaWallhereCrawl::CrawlImage

Inherits:
Object
  • Object
show all
Defined in:
lib/tomosia_wallhere_crawl.rb

Instance Method Summary collapse

Instance Method Details

#crawldata(key, destination, max = nil) ⇒ Object 



23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
 
# File 'lib/tomosia_wallhere_crawl.rb', line 23

def crawldata(key,destination,max=nil)	
  	sum = 0
		index = 1
		images = []
	while sum != max do
	 	# Open url
	  	url = "https://wallhere.com/en/wallpapers?q=#{key}&page=#{index}"
	  	document = open(url)
	  	content = document.read
	  	parsed_content = Nokogiri::HTML(content)				  
	  	length = parsed_content.css('.item').to_a.length - 1
			total_img = parsed_content.css('div.hub-totalinfo').text.split(' HD Wallpapers')[0].to_i
			if max == nil || max > total_img
				max = total_img
				puts "This tag has #{total_img} pictures"	
			end
				i = 0
		  	for i in i..length
						urlimg = parsed_content.css('.item').to_a[i].children.children.first.to_h['src']
						images.push(urlimg)
							
							print '.'
							sum += 1
				   	 	if max == sum 
				      	break
				end
			end
	    	index += 1
		end
		download(images,destination)
end

#download(images, destination) ⇒ Object 



55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
 
# File 'lib/tomosia_wallhere_crawl.rb', line 55

def download(images,destination)
		data = []
		row = {}
		thread = []
		images.each do |img|	
				thread << Thread.new(img) do
				timeout = 0
				begin 
					open(img) do |image|
						nameimg = File.basename(img,".jpg!s")
						ui = img
						ex = File.extname(img).delete('.!s')
						size = ""
						File.open("#{destination}/#{nameimg}","wb") do |file|
							file.write(image.read)
							size = image.size
						end
						size = size.to_s + " bytes"
						row = {"name"=>nameimg, "url"=>ui, "extension"=> ex, "size" => size}
						data.push(row)
					end
				rescue => exception
					if timeout < 3
						timeout += 1
						retry
					else
						next
					end
				end	
			end
		end
		thread.each {|t| t.join}
		puts " "
		puts "Download successfully"
		savedata(data,destination)
end

#savedata(data = {}, destination) ⇒ Object 



8
9
10
11
12
13
14
15
16
17
18
19
20
21
 
# File 'lib/tomosia_wallhere_crawl.rb', line 8

def savedata (data = {}, destination)
  workbook  = WriteExcel.new("#{destination}/InfoImage.xls")
  worksheet = workbook.add_worksheet
    data.each_with_index do |row, stt|
    row.each do |key, value|
      worksheet.write_string(stt, 0, row['name'])
      worksheet.write_url(stt, 1, row['url'])
      worksheet.write_string(stt, 2, row['extension'])
      worksheet.write_string(stt, 3, row['size'])
      end
    end
	workbook.close
	puts "Save successfully"
end