def fetch_site():
url = request.args['url']
r = requests.get(url)
if "UTF-8" not in r.encoding:
r.encoding = r.apparent_encoding
htmlcontent = r.text
#htmlcontent = htmlcontent.replace('href="/', 'href="' + url + '/')
baseurl = re.sub('(https*://)([^/]+)/*.*', r'\1'+ r'\2', url,flags=re.IGNORECASE)
baseurl = baseurl.replace('','')
#print('Base URL')
#print(baseurl)
htmlcontent = convert(request.args['source'], request.args['target'], htmlcontent, json.loads(request.args['nativize']),
json.loads(request.args['preOptions']), json.loads(request.args['postOptions']))
# Replace relative paths with absolute paths
htmlcontent=re.sub("(\")/",r"\1"+baseurl+"/",htmlcontent)
htmlcontent=re.sub("(\.\")/",r"\1"+baseurl+"/",htmlcontent)
htmlcontent=re.sub("(url\()\/",r"\1"+baseurl+"/",htmlcontent)
## Parameters
params = 'source=' + request.args['source'] + '&target=' + request.args['target'] + '&preOptions=' + request.args['preOptions'] + '&postOptions=' + request.args['postOptions'] + '&nativize=' + request.args['nativize']
transurl = html.escape("http://aksharamukha.appspot.com/api/website?"+params+'&url=')
# fix double dot
urlparts = url.split("/")
doubledot =""
for i in range(0, len(urlparts)-2):
doubledot = doubledot + urlparts[i]+ "/"
htmlcontent=htmlcontent.replace("../",doubledot)
## Replace links
htmlcontent=re.sub("(<a href\=\"?)",r"\1"+transurl,htmlcontent)
htmlcontent=re.sub("(<a class=.*? href\=\"?)",r"\1"+transurl,htmlcontent)
htmlcontent=re.sub("(<a target\=\"\_blank\" href\=\")",r"\1"+transurl,htmlcontent)
htmlcontent=re.sub("(<a target\=\"\_self\" href\=\")",r"\1"+transurl,htmlcontent)
## Replace with native numerals
htmlcontent = PostProcess.RetainIndicNumerals(htmlcontent, request.args['target'], True)
## Retain Dandas
htmlcontent = PostProcess.RetainDandasIndic(htmlcontent, request.args['target'], True)
return htmlcontent
