How to get the credit and how the practicals work
To get the credit you need at least 80 points from homework assignments. There will be at least 12 assignments, 10 points each. Presence at the practicals is not mandatory, practicals basically work as consultation sessions. But it is highly recommended to attend the first session where the rules and system is thoroughly explained. Usually, new assignment will be present together with solution of the previous one. Rest of the tutorial is Q&A session -- we discuss questions and issues concerning the lecture, assignments etc. For example, you may work on the assignment during the practicals so you can immediately consult any problems you encounter with the instructor.
Do not be afraid to ask questions! (either during the practicals or via e-mail). In the worst case, you just learn nothing new :-).
Assignments
All assignments are submitted using ReCodEx system (event the experimental ones). Please join the ReCodEx group corresponding to these practicals. Instructor's feedback to your solutions will also be in ReCodEx (can be useful to set up mail notifications). Materials for assignments (source codes etc.) will be published to the git repository.
The rules are here. Read the rules very carefully and consult the instructor if anything is not clear! Not knowing the rules is no excuse if you accidentally break them.Few suggestions:
- Start early. All-nighters just before deadline have a certain charm but you can be quite sure that the instructor won't answer your question at 3 a.m. Also, programming under the influence of sleep deprivation and litres of caffeine has usually quite negative effect on the quality and functionality of the code.
- Read the assignment instructions carefully and do not start coding right away -- begin by analyzing the problem using pen and paper. Do you really understand how the data structure works? And what about the edge cases?
- Write a clean and nice code. Do not be fooled by the automated test
system -- the instructor is going to read your code. There is an old programmer's saying
Always code as if the guy who ends up maintaining your code will be a violent psychopath who knows where you live.
The instructor is probably not a violent psychopath, nor he knows where you live, but he can certainly give you negative points for nasty hacks (and positive points for particularly nice implementation). You may use the code templates for the assignments as an inspiration. By the way, more comments or long variable names does not necessarily mean nicer code -- sometimes less is more. - Test you code! Do you change the data structure during the operation? What about some assertions or conditions to check the invariants of the data structure? Is your heap falling apart? Why not draw its structure during the operations by some tool such as DOT?
- Do not over-optimize your code. Time and memory constraints should be generous enough so that your implementation should pass as long as it has correct asymptotic complexity and does not do any crazy stuff. It is usually better to have a nice clean code that does small amount of extra work than super-duper optimized spaghetti that not even you can really understand.
If you get stuck you can try:
Language intermezzo
Implementation assignments can be submitted either in Python 3 or in C++17. More detailed parameters, such as compiler options, can be found in ReCodEx. If you do not know C++ but you know C, do not despair! It is recommended to use C++ as if it was plain old C (perhaps with exception of C++ vectors), so you can ignore most of the features the C++ has.
If you know neither C++ nor Python there are references to documentation and some tutorials below. And, as always, you can ask the instructor.
If you are unsure which language to choose I personally would recommend Python. I consider it to be more beginner-friendly and perhaps more useful in general. Either way, I would definitely recommend using the language you feel comfortable with.
Python
- Documentation and various references
- List of tutorials
- PyCharm: IDE for Python (if text editor is not enough)
C++
- Documentation and various references
- One tutorial and another tutorial
- QtCreator: IDE for C++ (it can be also be found as package in some of the linux distributions)