RomanNum3ral
/
scrapy
mirror of https://github.com/scrapy/scrapy.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

removed multispider support from logstats

This commit is contained in:
nramirezuy 2013-05-27 17:48:49 -03:00
parent 98aa3efea7
commit 96b8fb69be
1 changed files with 19 additions and 38 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

57
scrapy/contrib/logstats.py
View File

@ -3,20 +3,13 @@ from twisted.internet import task
from scrapy.exceptions import NotConfigured
from scrapy import log, signals
class Slot(object):
def __init__(self):
self.items = 0
self.itemsprev = 0
self.pages = 0
self.pagesprev = 0
class LogStats(object):
"""Log basic scraping stats periodically"""
def __init__(self, interval=60.0):
def __init__(self, stats, interval=60.0):
self.stats = stats
self.interval = interval
self.slots = {}
self.multiplier = 60.0 / self.interval
@classmethod
@ -24,40 +17,28 @@ class LogStats(object):
interval = crawler.settings.getfloat('LOGSTATS_INTERVAL')
if not interval:
raise NotConfigured
o = cls(interval)
crawler.signals.connect(o.item_scraped, signal=signals.item_scraped)
crawler.signals.connect(o.response_received, signal=signals.response_received)
o = cls(crawler.stats, interval)
crawler.signals.connect(o.spider_opened, signal=signals.spider_opened)
crawler.signals.connect(o.spider_closed, signal=signals.spider_closed)
crawler.signals.connect(o.engine_started, signal=signals.engine_started)
crawler.signals.connect(o.engine_stopped, signal=signals.engine_stopped)
return o
def item_scraped(self, spider):
self.slots[spider].items += 1
def response_received(self, spider):
self.slots[spider].pages += 1
def spider_opened(self, spider):
self.slots[spider] = Slot()
self.pagesprev = 0
self.itemsprev = 0
def spider_closed(self, spider):
del self.slots[spider]
self.task = task.LoopingCall(self.log, spider)
self.task.start(self.interval)
def engine_started(self):
self.tsk = task.LoopingCall(self.log)
self.tsk.start(self.interval)
def log(self, spider):
items = self.stats.get_value('item_scraped_count', 0)
pages = self.stats.get_value('response_received_count', 0)
irate = (items - self.itemsprev) * self.multiplier
prate = (pages - self.pagesprev) * self.multiplier
self.pagesprev, self.itemsprev = pages, items
msg = "Crawled %d pages (at %d pages/min), scraped %d items (at %d items/min)" \
% (pages, prate, items, irate)
log.msg(msg, spider=spider)
def log(self):
for spider, slot in self.slots.items():
irate = (slot.items - slot.itemsprev) * self.multiplier
prate = (slot.pages - slot.pagesprev) * self.multiplier
slot.pagesprev, slot.itemsprev = slot.pages, slot.items
msg = "Crawled %d pages (at %d pages/min), scraped %d items (at %d items/min)" \
% (slot.pages, prate, slot.items, irate)
log.msg(msg, spider=spider)
def engine_stopped(self):
if self.tsk.running:
self.tsk.stop()
def spider_closed(self, spider, reason):
if self.task.running:
self.task.stop()