CSC111  Lab #10

For this lab you need the amazon library. It is a library written by Mark Pilgrim. It allows Python programmers to write programs that get information from Amazon.com's database without using a browser. To see it at work, get a copy of this library

getcopy amazon.py

Next, create and run this program:

import amazon # import amazon.com library def getBooksWithKeywords( keyword ): pythonBooks = amazon.searchByKeyword( keyword ) print "Number of books found: %d" % ( len( pythonBooks ) ) for i in range( len( pythonBooks ) ): print "name: %s" % (pythonBooks[i].ProductName) print "URL: %s" % (pythonBooks[i].URL) print "Price: %s" % (pythonBooks[i].OurPrice) print "ISBN: %s" % (pythonBooks[i].Isbn) print "Image: %s" % (pythonBooks[i].ImageUrlMedium) print "\n\n" def main(): getBooksWithKeywords( raw_input( "keywords in book title? " ) )

The amazon.searchByKeyword( keyword ) is one of the functions provided by the amazon.py library. You give it a string representing a keyword, and it interrogates www.amazon.com for a list of books that match this keyword (most likely appearing in the title). The function returns a list of objects that have several instance variables associated with them, such as its ProductName, the url at Amazon where the book is showcased, or the link to the image showing the book cover (ImageUrlMedium). In class we would have defined inspector or accessor methods to lookup all these variables, such as getURL() or for example getX() for the Point objects when using the graphics.py module. Mark Pilgrim decided not to use this method and allows us direct access to the variables. That's fine. That's why we see pythonBooks[i].URL rather than a method call like pythonBooks[i].getURL().

• Copy and past or retype the above program into a file called lab10.py.
  Then run it, using the keyword

  guitar
  

  Notice how the formatting in the print statement makes the output look so tidy.

  Unexpected crashes

  Sometimes the information returned by Amazon will not contain some of the fields that are included at other times. For example, if you search books with the keywords "Python", or "guitar", the first program above runs fine.
  If, however, you run the program and provide the keyword "neural", then you will find that the program crashes because the attribute "OurPrice" is not included in some of the entries.

  The solution is to make our program aware that there might be exceptions. This is easy to do, and we saw a demonstration of it in class. Because the error we get when the program crashes is defined as an "AttributeError", we can rewrite the statement that causes the program to crash:

  
          print "Price:      %s" % (pythonBooks[i].OurPrice)
  

  as follows:

          try:
              print "Price:      %s" % (pythonBooks[i].OurPrice)
  
          except AttributeError:
              pass        
  

  You can read the code above as follows: "Try to get the attribute OurPrice from the i-th entry in the list pythonBooks. If successful, print it and continue. If not successful, and the error encountered is of type AttributeError, then do nothing (pass), and continue on."

  You can use this technique to protect your python statements that will access the list of books, music, or movies that Amazon will return to you. To get more information on the subject, read Section 8.3 of this page from the Python.org site.

• Read the comments in the amazon.py library on the search() function; it has many variants, allowing one to search for books, for music, or for movies, by keywords, by artist, and many other characteristics. Do not worry about the requirements for getting a license from amazon. I already did this part.
• Here is a variation on the program above that will search for music CD's given the name of an artist specified by the user:

  import amazon def getCDsFromArtist( artist ): musicList = amazon.searchByArtist( artist, product_line="music" ) print "Number of items found: %d" % ( len( musicList ) ) for i in range( len( musicList ) ): print "name: %s" % (musicList[i].ProductName) print "URL: %s" % (musicList[i].URL) print "Price: %s" % (musicList[i].OurPrice) print "Media: %s" % (musicList[i].Media) print "Image: %s" % (musicList[i].ImageUrlLarge) print "Released: %s" % (musicList[i].ReleaseDate) print "\n\n" def main2(): getCDsFromArtist( raw_input( "artist? " ) )

  Read Section 4.7.2 on Keyword Arguments of the Python tutorial to understand the way the parameter product_line="music" is passed.
  Make sure to read this now and to understand it before moving on.

• Type the above program into your file called lab10.py. Noticed that I used main2 as the name of the main function so I could call either one.
• Run the program and type in

  Rufus Wainwright
  

  Oops, there goes that AttributeError again! Fix it up with the try-except statement and rerun with the same Artist name.
• When you rerun the program and type in

  Rufus Wainwright
  

  you may now get

  AttributeError: Bag instance has no attribute 'ReleaseDate'
  

  Put a try-except statement around the print statement for ReleaseDate and continue. You may wish to add try-except to all the prints.
  Bag instance refers to the amazon.py class Bag. Bag objects are used to store the values retrieved from Amazon.com.
• Now run it one more time and when prompted for Artist Name, type in

  yduJ nilknarF
  

  You should see a long Traceback and this error explanation:

  amazon.AmazonError: There are no exact matches for the search.
  

  Oops. There were no CDs by artist yduJ nilknarF. The error is amazon.AmazonError.
  This is an error that is defined in amazon.py by the programmer. Yes, the programmer can define errors. We can catch it too. Put a try-except around the statement:

   musicList = amazon.searchByArtist( artist,  product_line="music" )
  

  to catch this error. What should you do after the except? If you can't figure that out, try running it with the try-except and use the pass statement in the body of the except: part.
• You have now graduated to Hw10!