backup/pywb
1
0
Fork 0
mirror of https://github.com/webrecorder/pywb.git synced 2025-03-24 06:59:52 +01:00
Code Issues Projects Releases Activity

rangecache: don't redirect when using range header, don't cache non-200

Browse Source 
responses
This commit is contained in:
Ilya Kreymer 2014-11-06 22:14:41 -08:00
parent 1a91f514c5
commit 9929737a8e
2 changed files with 11 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
pywb/webapp/rangecache.py
View File

@ -14,7 +14,7 @@ class RangeCache(object):
YOUTUBE_RX = re.compile('.*.googlevideo.com/videoplayback') YOUTUBE_RX = re.compile('.*.googlevideo.com/videoplayback')
YT_EXTRACT_RX = re.compile('&range=([^&]+)') YT_EXTRACT_RX = re.compile('&range=([^&]+)')
DEFAULT_BUFF = 16384 DEFAULT_BUFF = 16384*4
@staticmethod @staticmethod
def match_yt(url): def match_yt(url):
@ -78,9 +78,12 @@ class RangeCache(object):
key = digest key = digest
if not key in self.cache: if not key in self.cache:
wbrequest.custom_params['noredir'] = True
response = wbresponse_func() response = wbresponse_func()
if not response:
return None, None # only cache 200 responses
if not response.status_headers.get_statuscode().startswith('200'):
return response.status_headers, response.body
with NamedTemporaryFile(delete=False) as fh: with NamedTemporaryFile(delete=False) as fh:
for obj in response.body: for obj in response.body:
@ -94,8 +97,6 @@ class RangeCache(object):
self.cache[key] = yaml.dump(spec) self.cache[key] = yaml.dump(spec)
else: else:
spec = yaml.load(self.cache[key]) spec = yaml.load(self.cache[key])
if not spec:
return None, None
spec['headers'] = [tuple(x) for x in spec['headers']] spec['headers'] = [tuple(x) for x in spec['headers']]

6
pywb/webapp/replay_views.py
View File

@ -201,7 +201,11 @@ class ReplayView(object):
if wbrequest.options['is_proxy']: if wbrequest.options['is_proxy']:
return None return None
if range_cache and range_cache.match_yt(wbrequest.wb_url.url): if range_cache:
if range_cache.match_yt(wbrequest.wb_url.url) or wbrequest.env.get('HTTP_RANGE'):
return None
if wbrequest.custom_params.get('noredir'):
return None return None
redir_needed = (wbrequest.options.get('is_timegate', False)) redir_needed = (wbrequest.options.get('is_timegate', False))