Created
December 6, 2014 07:47
-
-
Save jsb9412/b7f9adcf91a12861bcc6 to your computer and use it in GitHub Desktop.
Webtoon Download Program in Python
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import re, sys | |
import urllib2 | |
webtoonURL = 'http://comic.naver.com/webtoon/detail.nhn?titleId=' | |
subWebtoonURL = '&no=' | |
def saveHtml(content, filename) : | |
f = open(filename, 'w') | |
f.write(content) | |
f.close() | |
def getHtml(url) : | |
response = urllib2.urlopen(url) | |
return response.read() | |
def extractImgs(html) : | |
exp = re.compile(r'<img.+?src="(http://imgcomic\.naver\.net/webtoon/[0-9]+/[0-9]+/(.+?\.(jpg|png|gif)))".*?>') | |
imgs = exp.findall(html) | |
return imgs | |
def printImages(images) : | |
for img in images : | |
print images[0] # full link | |
print images[1] # file name | |
print images[2] # extension | |
return 0 | |
def saveImage(imgName, link, index) : | |
print 'start saveImage function!' | |
fileName = imgName + str(index) + '.jpg' | |
f = open(fileName, 'wb') | |
print('link is ' + link) | |
response = urllib2.urlopen(link).read() | |
# content = response.read() | |
#print content | |
f.write(response) | |
f.close() | |
return 0 | |
def saveWebtoon(titleID, htmlName, index, imgName) : | |
fullURL = webtoonURL + str(titleID) + subWebtoonURL + str(index) | |
print ('fullURL : ' + fullURL) | |
html = getHtml(fullURL) | |
htmlFullName = htmlName + '_' +str(index) + '.html' | |
saveHtml(html, htmlFullName) | |
f = open(htmlFullName, 'r') | |
html = f.read() | |
f.close() | |
images = extractImgs(html) | |
if len(images) == 0 : | |
print >> sys.stderr, "No Images!!" | |
return 1 | |
firstURL = images[0][0][:22] | |
secondURL = images[0][0][25:-5] | |
fileNum = 1 | |
for img in images : | |
currentURL = firstURL + 'com' + secondURL + str(fileNum) + '.jpg' | |
saveImage(imgName, currentURL, fileNum) | |
fileNum += 1 | |
return 0 | |
def main() : | |
# input about data | |
titleID = input('Enter TitleID : ') | |
finalEpisode = int(input('Enter Final Episode : ')) | |
htmlName = titleID | |
imgName = input('Enter Image Name : ') | |
for i in range(1, finalEpisode) : | |
fullImageName = imgName + 'Episode' + str(i) + '_' | |
saveWebtoon(titleID, htmlName, i, fullImageName) | |
return 0 | |
if __name__ == '__main__' : | |
sys.exit(main()) | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment