My proposed Programming Paradigms course has been unanimously approved 
by the faculty, and even "upgraded" from a CIT course to a CIS course, so that it is available to MSE majors. Here's a writeup.
The course is currently listed on the Registrar's page as CIS 700-002 (that's just a placeholder--the actual number will change). It will be MW 10:30-12:00 (the time will not change).
Would you like to help grade your Interpreter assignment?
Here's the deal. Kompally will be using this test1.logo program to test your interpreter. It should produce the picture shown on the right. (The five linked boxes test the various commands, while the green dots test procedure calls, parameters, and local/global variables.) If your program doesn't work at all, he will try to fix it so that it does. However, if your program just crashes, or says there's a syntax error but doesn't say where (I think my program is syntactically correct), he may not be able to make it work. If he can't make your interpreter work, you will get a bad grade. |
 |
If your interpreter doesn't work with my test program, you can find and fix the errors, and send the corrections to Kompally (not the entire program, just the parts that need to be changed). That way he can get the program to work more easily, it's less work for him, and almost certainly a better grade for you.
If your original program is broken to the extent that Kompally cannot fix it, and you have to resubmit, this will cost at least 20 points (out of the 150 total).
No class Tuesday, April 20. Please attend the Saul Gorn lecture (3:00, Wu & Chen).
 |
Average (mean): 75.7 Standard deviation: 15.2 Great job, everybody! |
Pizza party!
Thursday, April 15, 6 p.m.
Levine 307
Who is invited: All current and former MCIT students and faculty, and students taking MCIT classes.
Please email to let her know if you are likely to attend, so that we can order enough pizza! (OK to bring a guest, just let us know.)
My office hours this week:
Wednesday 11:00-12:00 and 1:30-3:00.
Thursday 3:00-4:00
Friday 11:00-12:00
As you probably know, Diana will be leaving us next year.
This week we have two applicants coming in to interview for her position. I would really appreciate it if you could attend their talks and give me your opinions. (Sorry, no, this doesn't mean extra points in CIT594 ... but it does mean free food.)
Chris Murphy, Wednesday, March 24, 307 Levine--Italian dinner (for MCIT students and faculty) at 5:30, and talk at 6.
David Nellans, Thursday, March 25, talk at 1:30-3:00, Skirkanich Auditorium.
Pizza party at 6:00.
I have posted DavesRecognizerTest.java.
Some of you failed a large number of tests by only following the "happy path" and seldom or never checking for possible parse errors. This time, we have only taken off 5 points for failing to throw SyntaxExceptions, regardless of how many tests you failed. However, failing to check input for errors does not lead to very robust programs.
Another correction. My <procedure> example didn't follow the grammar. As usual, in case of a conflict, follow the grammar as given in the Recognizer.
There's someplace in provided code that I treat list as a NAME rather than a KEYWORD. It should be a KEYWORD. I'll try to fix it when I have a chance.
Back from my conference. Thankfully, only a few minor issues have arisen regarding the parser.
- As noted in the Parser writeup, where there is a conflict, it is the grammar in the Recognizer writeup that you should use. There was a conflict in the rule for
<procedure>. Fortunately, this does not affect the tree that should be constructed. - In the tree examples I used a
tonode as the root of a procedure, but in another place I useddefinefor the same purpose. Please use"to". - Because
xis a keyword and not a variable, I should have used a different name in the<set>and<procedure>examples; other places it is okay.
The code for Parser.java and ParserTest.java has now been posted.
I had to make a lot of corrections and updates that I didn't expect--that's why it took so long. I hope I got them all.
This is starter code, intended to help you get started. You may modify it in any way you please (including ignoring it altogether).
All CIS/MSE, MCIT, CGGT, EMBS, & Robotics students and all faculty are invited to a lunch tomorrow, 2/18, 11:30-1:15, in Levine 307. Stop by for something to eat and a chat.
take care,
Mike
Special MCIT talk
AFTER THE REVOLUTION:
Coping with success in an Open Source World
Eric S. Raymond
Heilmeier Hall (Towne 100)
6:00-8:00pm, Wednesday,
February 24
|
Pizza party! Thursday, February 18, 6 p.m. Who is invited: All current and former MCIT students and faculty, and students taking MCIT classes. OK to bring a guest. |
Additions to Tree Generation assignment:
I've added a method to the Grammar class: public static boolean isTerminal(String token). This is a one-liner: A string is a token nonterminal if it begins with a '<', ends with a '>', and is at least 3 characters long. This could be done outside the Grammar class, but it really belongs in the Grammar class. (Thanks, Kapil.) Also, getDefinitions should return null if given a terminal instead of a nonterminal.
In Grammar.read() you can, if you prefer, use the split method of String, instead of StringTokenizer (thanks, George). The form to use is line.trim().split("\\s+").
Please make the
read() method of the Grammar class static. There isn't any reason you should have to already have a Grammar object before you can read one in. If it's static, you can just say Grammar g = Grammar.read(); (Thanks, John!)You are not required to use the parse method I supplied. If you find it convenient, use it; if not, don't. (However, if it has syntax errors or raises exceptions when copied into your
Tree class, there's probably something wrong with your code.) Cracking the Technical Interview:
Advice for CS students on interviewing for software engineer positions
Gayle Laakmann
Thursday, February 4, 2010, 4:00pm (CIT 594 will not meet this day)
Wu & Chen Auditorium
Abstract
CS interviews are a different breed from other interviews and, as such, require specialized skills and techniques. /Cracking the Technical Interview/ will teach you how to prepare for technical interviews, what top companies like Google and Microsoft really look for, and how to tackle the toughest programming and algorithm problems. This talk will include stories from the speaker's extensive interviewing experience as well as a live "demo" of how to tackle a technical problem
About the Speaker
Gayle Laakmann (Penn CS BSE & MSE '05) is the founder and CEO of CareerCup.com and the author of "Cracking the Technical Interview" (http://www.careercup.com/book). CareerCup is the leading source of technical interview preparation and provides a free forum with 3000+ technical interview questions, a book, a video, and mock interviews. Gayle has worked for Google, Microsoft and Apple and has extensive interviewing experience on both sides of the table. She has interviewed and received offers from Google, Microsoft, Apple, Amazon, IBM, Goldman Sachs and a variety of other firms, and she has interviewed over 120 candidates at Google and served on its hiring committee. She is now an MBA student at Wharton.
Email: gayle@careercup.com
Personal Website: http://www.technologywoman.com
Company Website:http://www.careercup.com
parse method is like your toString method in reverse; that is, it accepts the kind of strings that toString should produce, and constructs Tree<String> objects from them.