#version 0.6

import os,time, re
import urllib
 
class imgur_downloader():
	
	images = []
	subdomain = ''
	counter = False
	
	def __init__(self):
		return None
	
	def parseSubdomain(self, url):
		page = urllib.urlopen(url).read()
		links = []
		last = 0
		
		tag = '"cover"'
		
		while 1:
			
			last = page.find(tag, last)
			
			if last == -1:
				break
			
			links.append( "http:"+page[page.find('href=', last)+6:page.find('">', last+9)]+"/all" )
			
			last = last + 9
		
		return links

	def parseAlbum(self, url):
		page = urllib.urlopen(url).read()
		
		
		titleStart = page.find("data-title")+12
		albumname = page[titleStart:page.find('"',titleStart)]
		
		print "parsing album: %s" % albumname
		
		albumimages = [albumname]
		
		start = page.find("images:", page.find("ImgurAlbum"))+8
		rawAlbumdata = page[start: page.find("]}", start)+2]
		
		null=False
		albumdata = eval(rawAlbumdata)
		
		for i in albumdata["items"]:
			albumimages.append( "http://i.imgur.com/"+i["hash"]+i["ext"] )
		
		return albumimages

	def parseGallery(self, url):
		gallery = urllib.urlopen(url).read()
		
		baseUrl = gallery.find("baseURL:")
		url = "http://www.imgur.com"+gallery[baseUrl+8:gallery.find(",", baseUrl)].replace(' ','').replace("'",'')
		
		galleryname = gallery[baseUrl+8:gallery.find(",", baseUrl)].replace(' ','').replace('/','').replace("'",'')
		
		print "parsing gallery: %s" % galleryname
		
		galleryimages = [galleryname]
		
		maxpage = gallery.find("maxPage:")
		pagecount = gallery[maxpage+8:gallery.find(",", maxpage)].replace(' ','')
		
		for page in range(eval(pagecount)):
			
			if url[-1:] == "/":
				xmlurl = url + "hot/page/"+str(page)+".xml"
			else:
				xmlurl = url + "/hot/page/"+str(page)+".xml"
			
			xml = urllib.urlopen(xmlurl).read()
			
			print "Page %s" % page
			
			last = 0
			
			xml.count("/hash")
			
			while 1:
				hash = xml.find("<hash>", last)
				
				if hash == -1:
					break
				
				link =  xml[ hash+6: xml.find("</", hash) ] 
				
				extPos = xml.find("<ext>", hash)
				ext = xml[ extPos+5 : xml.find("</", extPos) ] 
				
				galleryimages.append( "http://i.imgur.com/"+link+ext )
				
				last = hash+1
		
		return galleryimages

	def parsePage(self, url):
		
		if url.find("http://") == -1:
			url = "http://"+url
		
		page = urllib.urlopen(url).read()
		
		if page.find("subdomain_css")!=-1:
			links = self.parseSubdomain(url)
			
			self.subdomain = url[url.find("://")+3 : url.find(".", url.find("://")+3)]+"/"
			print self.subdomain
			
			for i in links:
				self.parsePage(i)
			
		elif page.find("album_css")!=-1:
			self.subdomain = page[page.find("By <a href=")+14 : page.find(".", page.find("By <a href=")+14)] +"/"
			print self.subdomain
			
			self.counter = True
			self.images.append( self.parseAlbum(url) )
			
		elif page.find("gallery_css")!=-1:
			self.counter = False
			self.images.append( self.parseGallery(url) )
		


	def saveImages(self):
		
		print 
		print "saving images:"
		
		for album in self.images:
			nr = 0
			for img in album[1:]:
				nr = nr+1
				origName = img[img.rfind("/")+1:]
				albumname = re.sub('[^A-Za-z0-9]+', '', album[0])
				
				if self.counter == False:
					fileName = "images/" + self.subdomain + albumname+"/"+origName
				else:
					fileName = "images/" + self.subdomain + albumname+"/"+str(nr).zfill(4)+"_"+origName
				
				dir = os.path.dirname(fileName)
				if not os.path.exists(dir):
					os.makedirs(dir)
				
				
				
				if os.path.exists(fileName)!= True :
					print "downloading %s" % origName
					try:
						imgData = urllib.urlopen(img).read()
						output = open(fileName,'wb')
						output.write(imgData)
						output.close()
					except:
						print "file reading error"
						continue
					time.sleep(1)
				else:
					print "file exists!"
					continue
				


if __name__ == '__main__':
	myurl = raw_input("input imgur gallery/subdomain/album adress:")
	newDownload = imgur_downloader()
	newDownload.parsePage(myurl)
	newDownload.saveImages()