EDAP15 Program Analysis

The homework exercise schedule is as follows:

Number	Start	Recommended Deadline	Link	Description	Notes
1	2020-11-14	2020-12-03	repository	description
2	2020-11-27	2020-12-10	repository	description	See Test Scripts below
3	2020-12-04	2020-12-17	repository	description	Use the `extended` branch (see "Extended Branch" below); also see Test Scripts below
4	2020-12-11	2021-01-12	repository	description	No presentation to TA; instead send e-mail with results to Christoph who will check / ask questions / sign off

Process:

Follow homework instructions.
Git tag your solution as instructed in the homework.
Book a time with the teaching assistant and present solution.
- You can only book one slot per week directly.
- You may ask the TA for an additional slot the same week after presenting to him, but slots are limited.
- You can only present one exercise to the TA at a time.
- You can present more than once.
- You do not have to meet the recommended deadline, but we strongly recommend that you do.
- We will not generally make slots available more than a week in advance (excepting the January slots).
- If you book a slot before or immediately after the recommended deadline, you are guaranteed enough slots to present all solutions.
If you have successfully presented all solutions by 2021-01-13, you are guaranteed that you can take the oral exam.

Extended Branch

To simplify homework 3, we have released an additional git branch that gives more structure in NonNullPointerAnalysis.jrag. You can merge this branch in the same way as explained here, except by using origin/extended instead of origin/master:

git remote add upstream https://git.cs.lth.se/creichen/edap15-exercise-3.git
git fetch upstream
git checkout master
git merge upstream/extended
git push origin master

If you are an early group who has already started on the exercise, you may still want to look at the extended skeleton skeleton implementation to take whatever functionality you find useful. We will factor in the extra work you did when determining whether you pass the exercise.

Test Scripts for Exercises 2 and 3

To run the test scripts for exercises 2 and 3, you can use the Python scripts compare-hw2.py and compare-hw3.py (should work for both Python 2 and Python 3), as follows:

python compare-hw2.py F.expected F.out

(for Homework 2, or compare-hw3.py for Homework 3). Here, F.out is the output produced by your tool. (On many operating systems, you can also use the hw2-compare.sh and hw3-compare.sh scripts to automate running your tool and comparing.)

If you have already passed, you do not need to re-run these tests to pass. If Noric or Christoph gave you different instructions, then those override the output of these scripts.

The .expected files now contain not only expected output, but also some meta-specifications. The script handles these automatically, so as long as you rely on the script to check your output, you don't need to understand them.

maybe: N 1 5 ?
choice: N 2 5 +  |  N 2 5 ?
merge: 3 10 20 30

If you do want to investigate by hand, here is what they mean:

maybe: This line is optional (i.e., it is not an error if you report this, but also not required).
choice: X | Y You should report either X or Y.
merge: L C0 C1 ... Cn Treat (line,column) pairs (L, C0), ..., (L, Cn) as equivalent. We use this for field-insensitive analysis to merge/normalise all field names.