rec10/trunk/src/xml2db_dom.py

   1 #!/usr/bin/python
   2 # coding: UTF-8
   3 # Rec10 TS Recording Tools
   4 # Copyright (C) 2009 Yukikaze
   5
   6 import xml.dom.minidom
   7 import zenhan
   8 import chdb
   9 import auto_rec
  10 import recdb
  11
  12 def getText(elm):
  13     nodelist = elm.childNodes
  14     rc = ""
  15     # 全てのノードに対して
  16     for node in nodelist:
  17         # テキストノードなら値を取得
  18         if node.nodeType == node.TEXT_NODE:
  19             rc = rc + node.data
  20     return rc
  21 def getText_item(obj):
  22     rc=""
  23     for o in obj:
  24         rc=rc+getText(o)
  25     return rc
  26 def writemultitvDB(bctype,tvlists):
  27     rec10d.rec10db.new_epg_timeline(bctype)
  28     for channel,start,stop,title,desc,longdesc,category in tvlists:
  29         rec10d.rec10db.add_epg_timeline(bctype, channel, start, stop, title, desc, longdesc, category)
  30 def writemultichDB(bctype,chlists):
  31     for ontv,dn in chlists:
  32         rec10d.rec10db.update_chname_by_ontv_epg_ch(ontv, dn)
  33 def xml2db_dom(xmlpath, bctype):#bctypeは放送種別で'TE'(地デジ)BS,CSがある。地デジの場合は te数字 が入る
  34     dtb=datetime.datetime.now()
  35     dom=xml.dom.minidom.parse(file(xmlpath))
  36     chlist=[]
  37     tvlist=[]
  38     bayes={}
  39     for ch in dom.getElementsByTagName('channel'):
  40         ontv=ch.getAttribute("id")
  41         chname=ch.getElementsByTagName("display-name").item(0).childNodes[0].data
  42         chlist.append([ontv,chname])
  43     for tv in dom.getElementsByTagName('programme'):
  44         channel=zenhan.toHankaku_ABC123(tv.getAttribute("channel"))
  45         start=tv.getAttribute("start").replace(" +0900", "")
  46         stop=tv.getAttribute("stop").replace(" +0900", "")
  47         title=zenhan.toHankaku_ABC123(getText_item(tv.getElementsByTagName("title"))).replace(",", "_")
  48         desc=zenhan.toHankaku_ABC123(getText_item(tv.getElementsByTagName("desc")))
  49         longdesc=zenhan.toHankaku_ABC123(getText_item(tv.getElementsByTagName("longdesc")))
  50         category=zenhan.toHankaku_ABC123(getText_item(tv.getElementsByTagName("category")))
  51         tch=chdb.ontvsearch(channel)
  52         if tch!=None:
  53             if tch.get('bctype',"")==ubctype:
  54                 bt=bayes.get(tch['chtxt'],["","",0])
  55                 bt2=[bt[0]+title+" ",bt[1]+desc+" "+longdesc+" ",bt[2]+1]
  56                 bayes[tch['chtxt']]=bt2
  57                 tvlist.append([channel,start,stop,title,desc,longdesc,category])
  58                 p=auto_rec.calc_key(tch['chtxt'], title,desc+" "+longdesc)
  59                 if p>2000:
  60                     recdb.add_auto_bayes(tch['chtxt'],title,start,stop,p)
  61     dom.unlink()
  62     if len(chlist) > 0:
  63         writemultichDB(chlist)
  64         chlist = []
  65         if len(tvlist) > 0:
  66             writemultitvDB(tvlist)
  67         tvlist = []
  68         for ct,list in bayes.items():
  69             auto_rec.update_recall(ct,list[0],list[1],list[2])
  70             auto_rec.update_recall("ALL",list[0],list[1],list[2])
  71     dtb=datetime.datetime.now()-dtb
  72     recdblist.printutf8(bctype + u" epg取り出し終了")
  73     recdblist.printutf8(str(dtb.days * 24 * 60 * 60 + dtb.seconds)+u"seconds taken for updating bayes-auto DB .")
  74