【Python】实现从AWR 报表上抓取指定数据

因为写关于数据库性能周报要查找和计算AWR 报表上的一些关键指标的值，每次手工收集数据都花很长时间，写了一个python 工具来获取自己想要的值，并做了计算！（现在看来还不太完善，以后会更贴近写周报的需求）
import sys
import urllib
import HTMLParser
import string

sum_Logical= 0
sum_Physical_reads= 0
sum_Physical_writes= 0
sum_Executes= 0
sum_Transactions= 0

##因为从awr获取的数值为unicode类型的，必须把值转换为数字！
def utof(s1):
   s2=s1.strip()
   s3=s2.encode( 'utf-8')
   s4=s3.split( ',')
   length= len(s4)
   if length <= 1 :
      t1= string.atof(s4[ 0])
      return t1
   elif length == 2:
      t1=string.atof(s4[ 1])+string.atof(s4[ 0])* 1000
      return t1
   elif length == 3:           t1=string.atof(s4[ 2])+string.atoi(s4[ 1])* 1000+string.atoi(s4[ 0])* 1000000
      return t1
   else:
      return 0
##类是解析html并且从html上获取想要的数据
urltext = []
class CustomParser(HTMLParser.HTMLParser):
  selected=( 'table', 'h1', 'font', 'ul', 'li', 'tr', 'td', 'a')
  def reset( self):
      HTMLParser.HTMLParser.reset( self)
      self._level_stack = []

  def handle_starttag( self, tag, attrs):
       if tag in CustomParser.selected:
         self._level_stack.append(tag)

  def handle_endtag( self, tag):
      if self._level_stack \
         and tag in CustomParser.selected \
         and tag == self._level_stack[- 1]:
         self._level_stack.pop()
  ##获取html上出去标签之后的文本数据
  def handle_data( self, data):
     if "/".join( self._level_stack) in ( 'table/tr/td', 'table/tr/td/h1/font', 'table/tr/td/ul/li') and data != ' \n ':
        urltext.append(data)    
##对传入的url 进行解析并获取数据
def gethtml(url):
    content = unicode(urllib.urlopen(url,params).read(), 'GB2312')
    parser = CustomParser()
    parser.feed(content)
    parser.close()

Logical=[]
Physical_reads=[]
Physical_writes=[]
Executes=[]
Transactions=[]
###计算想要的数据
def calucate(urltext):
    print '-----------------------------------------'
    global sum_Logical
    global sum_Physical_reads
    global sum_Physical_writes
    global sum_Executes
    global sum_Transactions
    k= 0
    for item in urltext:
       k=k+ 1
       if k< 50 :
         continue
       elif item == 'Logical reads:' :
         sum_Logical +=utof(urltext[k]) 
         print 'Logical reads:     ' ,urltext[k].strip()
       elif item == 'Physical reads:' :
         sum_Physical_reads +=utof(urltext[k])
         print 'Physical reads:    ',urltext[k].strip()
       elif item == 'Physical writes:' :
         sum_Physical_writes +=utof(urltext[k])
         print 'Physical writes:   ' ,urltext[k].strip()
       elif item == 'Executes:':
         sum_Executes += utof(urltext[k])
         print 'Executes:          ' ,urltext[k].strip()
       elif item == 'Transactions:' :
         sum_Transactions += utof(urltext[k])
         print 'Transactions:      ',urltext[k].strip()
       elif k> 86:
        break


if len(sys.argv) > 1:
  params = urllib.urlencode({ 'ip': sys.argv[ 1], 'action': 2})
else:
  params = None

um_Logical= 0
sum_Physical_reads= 0
sum_Physical_writes= 0
sum_Executes= 0
sum_Transactions= 0
     
url=[ 'http://127.0.0.1/cacti/spreport/rac3.yangql.com/sp_yangdb_20111211_10_16119_16120.html',
'http://127.0.0.1/cacti/spreport/rac3.yangql.com/sp_yangdb_20111211_17_16126_16127.html',
'http://127.0.0.1/cacti/spreport/rac3.yangql.com/sp_yangdb_20111210_17_16102_16103.html',
'http://127.0.0.1/cacti/spreport/rac3.yangql.com/sp_yangdb_20111210_10_16095_16096.html',
'http://127.0.0.1/cacti/spreport/rac3.yangql.com/sp_yangdb_20111209_17_16078_16079.html',
'http://127.0.0.1/cacti/spreport/rac3.yangql.com/sp_yangdb_20111208_17_16054_16055.html',
'http://127.0.0.1/cacti/spreport/rac3.yangql.com/sp_yangdb_20111209_10_16071_16072.html',
'http://127.0.0.1/cacti/spreport/rac3.yangql.com/sp_yangdb_20111208_10_16047_16048.html',
'http://127.0.0.1/cacti/spreport/rac3.yangql.com/sp_yangdb_20111207_17_16030_16031.html',
'http://127.0.0.1/cacti/spreport/rac3.yangql.com/sp_yangdb_20111207_10_16023_16024.html',
'http://127.0.0.1/cacti/spreport/rac3.yangql.com/sp_yangdb_20111206_17_16006_16007.html',
'http://127.0.0.1/cacti/spreport/rac3.yangql.com/sp_yangdb_20111206_10_15999_16000.html',
'http://127.0.0.1/cacti/spreport/rac3.yangql.com/sp_yangdb_20111205_17_15982_15983.html',
'http://127.0.0.1/cacti/spreport/rac3.yangql.com/sp_yangdb_20111205_10_15975_15976.html'
]

for val in url:
  print ' '
  gethtml(val)
  calucate(urltext)
  urltext = []
 
length= len(url)
print '-----------------------------------------'
print 'avg_Logical:       ',sum_Logical/length
print 'avg_Physical_reads:',sum_Physical_reads/length
print 'avg_Physical_writes',sum_Physical_writes/length
print 'avg_Executes       ',sum_Executes/length
print 'avg_Transactions   ',sum_Transactions/length
效果截图：