[Mypyspace-developer] SF.net SVN: mypyspace:[319] myspaceCrawler/trunk

[<gea...@us...>]

Revision: 319
          http://mypyspace.svn.sourceforge.net/mypyspace/?rev=319&view=rev
Author:   gearmonkey
Date:     2009-01-27 13:55:06 +0000 (Tue, 27 Jan 2009)

Log Message:
-----------
fixed the image url problem, now actually point to song images not myspace default filler images.

added a bit more inline documentation to the api examples code.

Modified Paths:
--------------
    myspaceCrawler/trunk/examples.py
    myspaceCrawler/trunk/mpsUser.py

Modified: myspaceCrawler/trunk/examples.py
===================================================================
--- myspaceCrawler/trunk/examples.py	2009-01-27 12:43:42 UTC (rev 318)
+++ myspaceCrawler/trunk/examples.py	2009-01-27 13:55:06 UTC (rev 319)
@@ -6,7 +6,7 @@
 some simple functions demonstrating mpsUser and mpsSong functionality
 
 Created by Benjamin Fields on 2008-11-09.
-Copyright (c) 2008 __MyCompanyName__. All rights reserved.
+Copyright (c) 2008 Goldsmiths. All rights reserved.
 """
 
 import sys
@@ -36,7 +36,7 @@
 		return 0
 	
 def socialCharts(initArtist, radius, chartLength=1):
-	'''breadth first crawl of width radius to find most chartLength popular songs from the center initArtist.'''
+	'''breadth first crawl of width radius to find at most chartLength popular songs from the center initArtist.'''
 	songQueue = []
 	visitedArtists = []
 	artistsInThisLevel = [initArtist]

Modified: myspaceCrawler/trunk/mpsUser.py
===================================================================
--- myspaceCrawler/trunk/mpsUser.py	2009-01-27 12:43:42 UTC (rev 318)
+++ myspaceCrawler/trunk/mpsUser.py	2009-01-27 13:55:06 UTC (rev 319)
@@ -46,11 +46,13 @@
         isArtist         --     Boolean, True means instance describes a MySpace artist with media
         rdfprefix        --     prefix for all rdf UIRs
         page             --     locally loaded copy of html pointed to by source
+			The following are only set if user is found to be an artist
         mediaXML         --     locally loaded (via miniDom) copy of xml describing playlist of media assciated 
-                                    with myspace Artist (not set in non artists)
+                                    with myspace Artist 
         totalPlays       --     sum of playcounts of all songs associated with myspace Artist 
-                                    (not set in non Artist)
-        artist           --     self declared name of artist (not set in non Artist)
+        artist           --     self declared name of artist
+		artistID         --     unique ID possessed by artists only, needed to retrieve media and  media related meta data
+		playlistID       --     unique ID used to retrieve playlist found on page
 
     
 	'''
@@ -342,7 +344,7 @@
 		else:
 			self.extractionprefix = extractionprefix
 		self.title = self.exhaustiveXML.getElementsByTagName('title')[0].firstChild.nodeValue
-		self.image = self.exhaustiveXML.getElementsByTagName('small')[0].firstChild.nodeValue
+		self.getimage()
 		self.playcount = xmlNode.getElementsByTagName('stats')[0].getAttribute('plays')
 		self.comments = "" #this is a blank string hold for the comments fields.  Might be used later.
 		self.trackNum, self.totalTracks = None, None
@@ -357,10 +359,31 @@
 		try:
 			self.uri = self.exhaustiveXML.getElementsByTagName('link')[0].firstChild.nodeValue
 		except AttributeError, err:
-			logging.info("mpsUser::getUri ran into a problem finding the download link for a song by artist with uid: " + 
+			logging.info("mpsUser::mpsSong::getUri ran into a problem finding the download link for a song by artist with uid: " + 
 				str(self.parent().uid) + " link will be left blank.\n\tError msg: " + str(err))
 			self.uri = ''
-
+	def getimage(self):
+		'''find an image associated with the song, getting the largest resolution available'''
+		try:
+			self.image = self.exhaustiveXML.getElementsByTagName('track')[0].getElementsByTagName('large')[0].firstChild.nodeValue
+		except AttributeError:
+			try:
+				self.image = self.exhaustiveXML.getElementsByTagName('track')[0].getElementsByTagName('medium')[0].firstChild.nodeValue
+			except AttributeError:
+				try:
+					self.image = self.exhaustiveXML.getElementsByTagName('track')[0].getElementsByTagName('small')[0].firstChild.nodeValue
+				except Exception, err:
+					logging.info("mpsUser::mpsSong::getimage ran into a problem finding an image for a song by artist with uid: " + 
+						str(self.parent().uid) + " image will be left blank.\n\tError msg: " + str(err))
+					self.image = ''
+			except Exception, err:
+				logging.info("mpsUser::mpsSong::getimage ran into a problem finding an image for a song by artist with uid: " + 
+					str(self.parent().uid) + " image will be left blank.\n\tError msg: " + str(err))
+				self.image = ''
+		except Exception, err:
+			logging.info("mpsUser::mpsSong::getimage ran into a problem finding an image for a song by artist with uid: " + 
+				str(self.parent().uid) + " image will be left blank.\n\tError msg: " + str(err))
+			self.image = ''
 	def setTrackNum(self, trackNumber, totalTracks):
 		'''set the track number for this song and the number of tracks in the album it is in.'''
 		self.trackNum = trackNumber


This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.