Home » Forums » User Support » Suggestion

Scrape patch

Posted May 18th, 2006 by Anonymous

In order to scrape some private sites correctly, the query parameters of the announce URL must be present in the scrape URL as well. The following patch changes the scrape code to conform to the BitTorrent Protocol Specification 1.0:


--- scrape.py.orig Thu May 18 11:38:37 2006
+++ scrape.py Thu May 18 12:05:59 2006
@@ -17,22 +17,30 @@
del policy

def announce_to_scrape(url):
- items = list(urlparse.urlparse(url))
- path = items[2]
- return urlparse.urljoin(url,
- os.path.basename(path).replace('announce','scrape'))
+ pos = url.rfind("/")
+ if pos == -1:
+ return None
+ if url[pos:(pos + 9)] != '/announce':
+ return None
+
+ return url[0:pos] + '/scrape' + url[(pos + 9):]

def get_scrape_by_announce_url(announce_url,infohash):
seeder,leecher = '?','?'
if not announce_url:
return seeder,leecher
scrape_url = announce_to_scrape(announce_url)
+ if not scrape_url:
+ return seeder,leecher

len_hash = len(infohash)
if len_hash == 40:
infohash = binascii.a2b_hex(infohash)

- scrape_url += '?info_hash=%s' % urllib.quote(infohash)
+ if scrape_url.find('?') >= 0:
+ scrape_url += '&info_hash=%s' % urllib.quote(infohash)
+ else:
+ scrape_url += '?info_hash=%s' % urllib.quote(infohash)

try:
try:

Hope this helps,
Sven

‹ BTQueue dead? Some patches ›
  • 1211 reads
  • nl

applied

On May 21st, 2006 sugree says:

Thanks :)

Post new comment

The content of this field is kept private and will not be shown publicly.