pjht/buku
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Handle multiple reirections. Show URL in BOLD.

Browse Source 
Signed-off-by: Arun Prakash Jana <engineerarun@gmail.com>
This commit is contained in:
Arun Prakash Jana 2016-03-25 01:10:30 +05:30
parent 1a47576649
commit 1ea332ba18
1 changed files with 13 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
buku
View File

@ -172,7 +172,6 @@ def getPageResp(url, redir=False):
# Fetch title from URL
def fetchTitle(url):
global titleData
titleData = None
urlconn = None
@ -183,25 +182,27 @@ def fetchTitle(url):
if resp.status != 200:
# Handle first redirection
if resp.status in (301,302,):
if debug:
print("Location header: %s" % resp.getheader('location', ''))
while 1:
print("\x1b[1mREDIRECTION:\x1b[21m %s" % resp.getheader('location', ''))
redirurl = urljoin(url, resp.getheader('location', ''))
if redirurl.find("sorry/IndexRedirect?") >= 0:
if redirurl.find("sorry/IndexRedirect?") >= 0: # graecefully handle Google blocks
print("ERROR: Connection blocked due to unusual activity.")
else:
if debug:
print("Trying to fetch redirected URL.")
urlconn.close()
resp, urlconn = getPageResp(redirurl, True)
if resp is not None:
if resp.status != 200:
print("ERROR on retry:", str(resp.status), ": ", resp.reason)
else:
getTitleData(resp)
else: # if resp.status in (301,302,):
print("ERROR:", str(resp.status), ": ", resp.reason)
if resp is None:
break
if resp.status in (301,302,):
continue
if resp.status != 200:
print("ERROR on retry:", str(resp.status), ": ", resp.reason)
else:
getTitleData(resp)
break
else: # if resp.status != 200:
getTitleData(resp)
except Exception as e:
@ -219,7 +220,6 @@ def fetchTitle(url):
def AddUpdateEntry(conn, cur, keywords, index):
global titleManual
global online
tags = ','
meta = ''
url = keywords[0]