New format for (not_)viewed pages/hits and bandwidth that are now recorded by day (in a dictionnary were only element 0 is initialized). Element 0 is the total. WARNING : not backward compatible with previous databases.
This commit is contained in:
@@ -83,14 +83,17 @@ class IWLAPreAnalysisPageToHit(IPlugin):
|
||||
|
||||
uri = request['extract_request']['extract_uri']
|
||||
|
||||
day = request['time_decoded'].tm_mday
|
||||
if request['is_page']:
|
||||
# Page to hit
|
||||
for regexp in self.ph_regexps:
|
||||
if regexp.match(uri):
|
||||
self.logger.debug('%s changed from page to hit' % (uri))
|
||||
request['is_page'] = False
|
||||
super_hit['viewed_pages'] -= 1
|
||||
super_hit['viewed_hits'] += 1
|
||||
super_hit['viewed_pages'][day] -= 1
|
||||
super_hit['viewed_hits'][day] = super_hit['viewed_hits'].get(day, 0) + 1
|
||||
super_hit['viewed_pages'][0] -= 1
|
||||
super_hit['viewed_hits'][0] += 1
|
||||
break
|
||||
else:
|
||||
# Hit to page
|
||||
@@ -98,6 +101,8 @@ class IWLAPreAnalysisPageToHit(IPlugin):
|
||||
if regexp.match(uri):
|
||||
self.logger.debug('%s changed from hit to page' % (uri))
|
||||
request['is_page'] = True
|
||||
super_hit['viewed_pages'] += 1
|
||||
super_hit['viewed_hits'] -= 1
|
||||
super_hit['viewed_pages'][day] = super_hit['viewed_pages'].get(day, 0) + 1
|
||||
super_hit['viewed_hits'][day] -= 1
|
||||
super_hit['viewed_pages'][0] += 1
|
||||
super_hit['viewed_hits'][0] -= 1
|
||||
break
|
||||
|
@@ -104,12 +104,12 @@ class IWLAPreAnalysisRobots(IPlugin):
|
||||
continue
|
||||
|
||||
# 1) no pages view --> robot
|
||||
# if not super_hit['viewed_pages']:
|
||||
# if not super_hit['viewed_pages'][0]:
|
||||
# super_hit['robot'] = 1
|
||||
# continue
|
||||
|
||||
# 2) pages without hit --> robot
|
||||
if not super_hit['viewed_hits']:
|
||||
if not super_hit['viewed_hits'][0] and super_hit['viewed_pages'][0]:
|
||||
self.logger.debug(super_hit)
|
||||
self._setRobot(k, super_hit)
|
||||
continue
|
||||
@@ -137,7 +137,7 @@ class IWLAPreAnalysisRobots(IPlugin):
|
||||
self._setRobot(k, super_hit)
|
||||
continue
|
||||
|
||||
if not super_hit['viewed_pages'] and \
|
||||
(super_hit['viewed_hits'] and not referers):
|
||||
if not super_hit['viewed_pages'][0] and \
|
||||
(super_hit['viewed_hits'][0] and not referers):
|
||||
self._setRobot(k, super_hit)
|
||||
continue
|
||||
|
Reference in New Issue
Block a user