Automatisch erkennen, ob mehrere Episoden den selben Titel haben, und dann fuer die Serie das Datum in den Dateinamen aufnehmen

src/tivomirror/tivomirror

@@ -1,6 +1,6 @@
 #!/usr/local/bin/python
 
-# $Schlepperbande: src/tivomirror/tivomirror,v 1.63 2014/07/05 15:23:02 stb Exp $
+# $Schlepperbande: src/tivomirror/tivomirror,v 1.64 2014/07/07 21:50:28 stb Exp $
 #
 # Stefans Script, um die Sendungen vom Tivo runterzuladen und in MPEG4
 # zu transkodieren.
@@ -138,6 +138,7 @@ class TivoItem:
 		self.sourcesize = int(getTagText(i, "SourceSize"))
 		self.highdef = getTagText(i, "HighDefinition")
 		self.ar = 43
+		self.unique = True
 		if arset.has_key(self.title):
 			self.ar = arset[self.title]
 		elif self.highdef == "Yes":
@@ -149,14 +150,21 @@ class TivoItem:
 				self.episode = self.description
 			else:
 				self.episode = self.datestr
+		self.formatnames()
+	def makeNotUnique(self):
+		self.unique = False
+		self.formatnames()
+	def formatnames(self):
 		if self.episodeNumber and self.episodeNumber != u'0':
 			en = int(self.episodeNumber)
 			if en >= 100:
 				self.name = "%s S%02dE%02d %s" % (self.title, en / 100, en % 100, self.episode)
 			else:
 				self.name = "%s E%s %s" % (self.title, self.episodeNumber, self.episode)
-		else:
+		elif self.unique:
 			self.name = "%s - %s" % (self.title, self.episode)
+		else:
+			self.name = "%s - %s - %s" % (self.title, self.datestr, self.episode)
 		self.dir = "%s/%s" % (targetdir, re.sub("[:/]", "-", self.title))
 		self.file = "%s/%s" % (self.dir, re.sub("[:/]", "-", self.name))
 		self.name = self.name.encode("utf-8");
@@ -168,19 +176,21 @@ class TivoItem:
 
 class TivoToc:
 	def __init__(self):
-		self.toc = None
+		self.dom = None
 		self.filename = "toc.xml"
+		self.uniquedb = anydbm.open("unique.db", "c")
+		self.items = []
 		pass
 
 	def load(self):
 		fd = open(self.filename, "r")
-		self.toc = xml.dom.minidom.parseString(fd.read())
+		self.dom = xml.dom.minidom.parseString(fd.read())
 		fd.close()
-		return self.toc
+		return self.dom
 
 	def save(self):
 		fd = open(self.filename, "w")
-		fd.write(self.toc.toprettyxml())
+		fd.write(self.dom.toprettyxml())
 		fd.close()
 
 	def download_chunk(self, offset):
@@ -195,7 +205,7 @@ class TivoToc:
 		}
 		url = "https://{}/TiVoConnect".format(host)
 		logger.debug("    offset %d" % (offset))
-		r = session.get(url, params=params)
+		r = session.get(url, params=params, timeout=30)
 		if r.status_code != 200:
 			r.raise_for_status()
 		return r.text
@@ -203,22 +213,47 @@ class TivoToc:
 	def download(self):
 		offset = 0
 		itemCount = 1
-		self.toc = None
+		self.dom = None
 		root = None
 		logger.info("*** Getting listing")
 		while itemCount > 0:
 			dom = xml.dom.minidom.parseString(self.download_chunk(offset))
-			if self.toc == None:
-				self.toc = dom
-				root = self.toc.childNodes.item(0)
+			if self.dom == None:
+				self.dom = dom
+				root = self.dom.childNodes.item(0)
 			else:
 				for child in dom.childNodes.item(0).childNodes:
 					if child.nodeName == "Item":
 						root.appendChild(child.cloneNode(True))
 			itemCount = int(getElementText(dom.documentElement.childNodes, "ItemCount"))
 			offset += itemCount
+		return self.dom
 
-		return self.toc
+	def getItems(self):
+		self.titles = {}
+		for node in self.dom.getElementsByTagName("Item"):
+			item = TivoItem(node)
+			self.items.append(item)
+			if item.title not in self.titles:
+				self.titles[item.title] = []
+			self.titles[item.title].append(item)
+		# see if we have items that end up having an identical name; mark
+		# the program title in uniquedb if that's the case
+		for title in self.titles:
+			names = {}
+			for item in self.titles[title]:
+				if item.name not in names:
+					names[item.name] = []
+				names[item.name].append(item)
+			for name in names:
+				if len(names[name]) > 1:
+					self.uniquedb[title.encode("utf-8")] = "1"
+		if getattr(self.uniquedb, "sync", None) and callable(self.uniquedb.sync):
+			self.uniquedb.sync()
+		for item in self.items:
+			if self.uniquedb.has_key(item.title.encode("utf-8")):
+				item.makeNotUnique()
+		return self.items
 
 
 def getText(nodelist):
@@ -418,10 +453,9 @@ def mirror(toc, downloaddb, one=False):
 			(targetdir, avail / gig, minfree / gig))
 		sys.exit(1)
 	
-	items = toc.toc.getElementsByTagName("Item")
-	logger.info("*** %d shows listed" % (items.length))
-	for node in items:
-		item = TivoItem(node)
+	items = toc.getItems()
+	print "*** %d shows listed" % (len(items))
+	for item in items:
 		reason = wantitem(item, downloaddb)
 		if reason != "":
 			logger.debug("*** skipping \"%s\": %s" % (item.name, reason))
@@ -432,7 +466,7 @@ def mirror(toc, downloaddb, one=False):
 
 
 def download_episode(toc, downloaddb, episode):
-	items = toc.toc.getElementsByTagName("Item")
+	items = toc.dom.getElementsByTagName("Item")
 	for node in items:
 		item = TivoItem(node)
 		if item.title == episode or item.name == episode or item.episode == episode:
@@ -440,11 +474,10 @@ def download_episode(toc, downloaddb, episode):
 
 
 def printtoc(toc, downloaddb):
-	items = toc.toc.getElementsByTagName("Item")
-	print "*** %d shows listed" % (items.length)
+	items = toc.getItems()
+	print "*** %d shows listed" % (len(items))
 	shows = {}
-	for node in items:
-		item = TivoItem(node)
+	for item in items:
 		if item.title not in shows:
 			shows[item.title] = []
 		shows[item.title].append(item)
@@ -489,6 +522,7 @@ def main():
 
 		if updateToc or cmd == "mirror":
 			toc.download()
+			toc.save()
 		else:
 			toc.load()