Earony

scrapy clusters great new way scale large, continuous scrapy projects. setup within vm , runs several instances different tools. i've built vm out , installed necessary tools such redis, kafka, , zookeeper. i'm stuck on simple line prior first test crawl , don't know if it's python issue, installation issue, or what. when try python kafka_monitor.py run i get: traceback (most recent call last): file "kafka_monitor.py", line 512, in sys.exit(main()) file "kafka_monitor.py", line 497, in main kafka_monitor.run() file "kafka_monitor.py", line 413, in run self._setup_kafka() file "kafka_monitor.py", line 232, in _setup_kafka ret_val = _hidden_setup() file "/usr/local/lib/python2.7/dist-packages/scutils/method_timer.py", line 46, in f2 retval = f(*args) file "kafka_monitor.py", line 218, in _hidden_setup self.kafka_conn = kafkaclient(self.se...

so having issue alarmmanager. trying run code every 2 minutes, works okay while phone awake, not when device goes sleep - during sleep, intervals 5 minutes apart. since desired interval 2 minutes, 250% off target interval, specific application not acceptable. i aware of changes in api 19, , have followed suggestions re-schedule alarm using setexact() within broadcastreceiver. code below: the code used trigger broadcastreceiver: intent intent = new intent(this, alarmreceiver.class); final pendingintent pendingintent = pendingintent.getbroadcast(this, 0, intent, 0); malarmmanager = (alarmmanager) this.getsystemservice(context.alarm_service); malarmmanager.setexact(alarmmanager.elapsed_realtime_wakeup, systemclock.elapsedrealtime() + 3000, pendingintent); and code in broadcastreceiver re-schedules alarm: @override public void onreceive(context context, intent intent) { toast.maketext(context, "alarmservice triggered.", toast.length_short).sh...

i want fill random bytes files recursivelly root directory. files, folders left untoutched i don´t want delete files, fill them random bytes in research, fill command single file like: dd if=/dev/random of=target-files bs=1m and find files recursivelly should use: find . -name "* .*" my questions are: is possible achieve goal joining 2 commands? (pipe them? how?) is there easier way achieve same results? thanks ;-) you may try pipe find result while loop, , dd /dev/udrandom each files. $ find myfolder -type f |while read fd; dd if=/dev/urandom of=$fd bs=1m count=1;done if want retain file size, may calculation on finding 1k block count , pass args in dd. $ find myfolder -type f|while read fd; fsiz=`stat -c %s $fd`; cnt=`expr $fsiz / 1024`;dd if=/dev/urandom of=$fd bs=1k count=$cnt;done

i have 1 matrix of size 12 , array of size 4 given below: a=[23,45,1;4,6,7;9,5,0;145,65,32]; b=[3,2,4,6]; i want new array c c=[23,45,1,4,6,7,9,5,0,145,65,32,3,2,4,6]; i did following : a=[23,45,1;4,6,7;9,5,0;145,65,32]; b=[3,2,4,6]; a=reshape(a',12,1); b=b(:); c=[a b]; but giving error : error using ==> horzcat cat arguments dimensions not consistent. a = a(:); b = b(:); c = [a; b]; note difference between [a b] or [a, b] , [a; b].

i thought web socket guarantees delivery of data built on top of tcp reliable. reading java web socket api documentation on link https://docs.oracle.com/javaee/7/api/javax/websocket/remoteendpoint.html says there no guarantee of successful delivery of web socket message peer, if action of sending message causes error known container, api throws it. shouldn't tcp gurantee message delivery? reliable != guaranteed. reliable means you'll notified if there failure (or success) - if end users unplugs lan cable, tcp can't 'guarantee delivery.'

i have 2 feature files in project 1 step common in these 2 files need call 2 different step definition function depending upon file on steps written. do have configuration that? so if understand correctly, have 1 line in cucumber needs execute 2 different methods? far know not possible. cucumber keep on telling there multiple methods same name. - title. what explanation different. have 2 feature files common step. 1 feature file getting steps classfile , other feature file getting steps classfile b? in case, can best extend public classes can use code class file in feature file.

i using databasedotcom & databasedotcom-rails gem generate leads salesforce form submissions. working standalone until nested controller under parent, , when press submit routes error: no route matches [post] "/events/516ee9a0421aa9c44e000001/leads/new" here code: resources :events resources :leads end class leadscontroller < applicationcontroller include databasedotcom::rails::controller def new @lead = lead.new respond_to |format| format.html # new.html.erb format.json { render :json => @lead } end end def create @lead = lead.new(params[:lead]) @lead.event_id = params[:event_id] @lead['ownerid'] = '005b0000000wxqe' @lead['isconverted'] = false @lead['isunreadbyowner'] = false respond_to |format| if @event_lead.save format.html { redirect_to @lead, :notice => 'lead created.' } format.json { render :json => @lead, :status => :created, :location ...