crawl often

crawler_script/crawl_often.py

@@ -36,11 +36,11 @@ def dictRowGenerator(line):
     except:
         pass
     try:
-        row.append(("url", line['url']))
+        row.append(("text", line['text']))
     except:
         pass
     try:
-        row.append(("text", line['text']))
+        row.append(("url", line['url']))
     except:
         pass
 
@@ -50,7 +50,6 @@ def dictRowGenerator(line):
 today = datetime.datetime.now()
 baseDir = "/home/geoint/virtualHDD/m3/noticias/"
 scrapyDir = "/home/geoint/crawlersNoticias/"
-
 with open(sys.argv[1]) as data_file:
     siteList = json.load(data_file)
     os.chdir(baseDir)
@@ -112,7 +111,7 @@ with open(sys.argv[1]) as data_file:
             for d in range(day, ((datetime.date(y, 12, 31) - datetime.date(y, 1, 1)).days + 1 if today.year != y else today.timetuple().tm_yday) + 1):
                 YESTERDAY = False
                 filename = currentDate.strftime('%Y-%m-%d') + ".json"
-                scrapycommand = "scrapy crawl noticias -t json --nolog -o " + filename
+                scrapycommand = "scrapy crawl noticias --nolog -s filename=" + filename
                 mydir = os.getcwd()
                 print mydir
                 os.chdir(scrapyDir + s['crawler'])
@@ -171,14 +170,21 @@ with open(sys.argv[1]) as data_file:
                         f1 = mydir + '/' + filenameYesterday
                         f2 = filename
                         f3 = baseDir + media + '/' + filenameYesterday
-                        with open(f1) as infile1, open(f2) as infile2, open(f3, 'a') as infile3:
+                        with open(f2) as infile2, open(f3, 'a') as infile3:
+                            try:
+                                infile1 = open(f1)
                                 master = json.load(infile1)
+                                yesterdayFlag = True
+                            except:
+                                yesterdayFlag = False
+
+                            urlSet = set()
                             slave = json.load(infile2)
+                            infile3.write("[")
 
+                            if yesterdayFlag:
                                 urlSet = set([line['url'] for line in master])
                                 counter = 0
-                            infile3.write("[")
-
                                 for line in master:
                                     counter += 1
 
@@ -190,14 +196,20 @@ with open(sys.argv[1]) as data_file:
                                     elif counter > 1:
                                         infile3.write(",\n" + json.dumps(row))
 
+                            counter = 0
                             for line in slave:
                                 lineDate = datetime.datetime.strptime(line['date'][:10], '%Y-%m-%d').date()
                                 if not line['url'] in urlSet and lineDate == currentDate:
+                                    counter += 1
                                     auxRow = dictRowGenerator(line)
                                     row = OrderedDict(auxRow)
+                                    if not yesterdayFlag and counter == 1:
+                                        infile3.write(json.dumps(row))
+                                    else:
                                         infile3.write(",\n" + json.dumps(row))
 
                             infile3.write("]")
+                            if yesterdayFlag: infile1.close()
 
                         os.system("mv " + f3 + " " + mydir)
                     os.system("rm " + f2)