:From the timeline…

Darwin went pretty smoothly overall, I feel. I was working with Ting on this one, which resulted in a nice makefile being used as well. I don’t believe our creature regularly beats out the Rover, unfortunately, but the instruction set is really limited. Estimated and time actually spent were both around 8 hours, I believe.

C++ and inheritance are.. special. Some of the weird things regarding binding and using things from a parent class have made me come to the conclusion that, when asked, if something looks like it shouldn’t compile, it probably will compile into perfectly legitimate C++. Being able to define something in an abstract class is useful, though, compared to not being able to do anything. It does seem odd however that a language that forces someone into doing very little (fully unchecked exceptions, for instance) does have a mechanism for forced abstract classes. Just one more confusing thing about it, I suppose.

No Comments

I was out of town last weekend and hardly around the internet, oops. Darwin looks like a fun project, I’m working with Ting on it. Should be finished with it fairly quickly. Pair and tuple are useful enough structures, though there isn’t a whole lot to really say about either of them. Pairs by value, tuples by reference.

No Comments

Functions inside of classes, more coverage of operators. Both were pretty useful, and some things that I had been wondering about exactly how to do.  Hard to really think of what else to comment on, I’m feeling kind of sick at the moment.

Worked with Alex Espinosa on project 6. Expected about 10 hours, final time was around 6-7 hours.

No Comments

Delay caused by spending large portions of the weekend replacing harddrives. Fun times.

Last week we finally got “properly” started on classes. Some of the things with friends and how C++ handles what I can only assume is backwards compatibility is just ridiculous. Specifically, classes, methods, and functions all handling templating differently in the context of friends really stuck out to me. Classes and methods don’t use <U> but functions do?

Probably some obscure reason. Grade on test 2 was about what I expected, I’m satisfied with it enough.

No Comments

Test 2 went well, I believe. I won’t know until I see my results, obviously, but I’m pretty sure I made a decent enough grade on it. Lots of questions similar to those on the quizzes over the past two weeks with the A(), ~A(), and so forth. Queue was fairly easy and should work. Etc.

Allocator went well. I was working with Gilbert Cardenas on this one. The expected and actual times are about 5 and a half hours, would’ve gone faster if I didn’t have a silly bug in the deallocate method I wrote. Duur, the opening of the prior block isn’t at 8-(priorsentinel’s number) it’s at 12-(priorsentinel’s number). Similarly with having an extra -4 on the following block. TestAllocator.out is going up in just a second.

No Comments

Almost forgot–the projects had kept me constantly updating, I suppose.

Further differences between arrays on the heap and on the stack. I’m liking vector and vector of vector even more, the code required to handle a 2D array on the heap otherwise is just nasty. The last method shown in class, using the array to point to a single array, that was useful and less complicated than most of the other things, but still ridiculous.

Allocator looks fun, and the test this week, eh. We’ll see.

No Comments

Ahh, C++.

I’m so glad that we’re finally getting into using the heap, it makes life so much easier on some things, even if we’re using vectors to get around having to actually manually manage the memory ourselves.  That, along with finally starting a bit more on objects with the object arrays. Not really a whole lot to report about the class for the week, though. I apparently couldn’t remember much of anything from the readings, that was kinda annoying.

Test coming up. Ah well, it’s still kinda far off.

As far as Matlab went…

predicted number of hours to complete: 2 hours
actual number of hours to complete: 3 hours

Not really a whole lot to say. It was a pretty simple project, no real bugs encountered that lasted for more than 30 seconds to need to be reported (wrong variable for indexing, oops, that was a stupid segfault). Jackie was nice to work with.

No Comments

So, I ended up with a 125/150 on the test, which is acceptable enough to me. Overall I thought the test was fine, though I couldn’t exactly remember what CRC stood for (I thought that “cards” was part of the name, not that it was CRC cards). The test was perfect, as far as the length of the actual programming questions was concerned. As far as the actual contents of the week, function defaults looks to be incredibly useful. Enums are pretty similar to the Java version, which I never really found proper uses for to begin with.

As far as pointers and their const versions go.. I still don’t see much difference from references in the const pointer case. However, typedef is something I feel I’m going to be using far more often (though I didn’t throughout Project 3).

Project 3 ended up without a partner, and somehow unaccepted by UVA. It runs with correct output on something like 96 of the 99 student test cases posted to the Google Group and then correctly on several test cases that I had made as well, and after weighing the differences for turning in a solution that works on large numbers of sample output vs turning in a correct solution a day late, I’m throwing in the towel. I predicted it’d take about 3 hours–to get to the point I’m currently at (~96/99, 6/6 of my cases) it’s taken about 8.

Most of the additional time was figuring out workarounds for various things–avoiding a blank line in C++, static array sizes being required in structs, etc. On the plus side, I am now far more comfortable with crazy istream things than I was before.

No Comments

Instead of putting this entry later in the week, post-test, I choose to instead put it into a pre-test optimist mindset post. So here it goes.

I actually think I’m prepared for the test. Global, class, and instance variables in C++ aren’t too difficult to understand compared to other languages that I know. Some of the weird handling of static variables with generic types and built-in types is different, but nothing too complicated to remember.

I realize the last entry was a little bare other than the project information–The limits of variables and how C++ handles the reading in was really the only thing that tripped me up, but I think I’ve got it all down now.

No Comments

I partnered with Michael Burnham on this one, and it took us pretty close to what I expected to be 2 hours.

Okay so it took 3, nothing very problematic with it other than having written a method backwards at first. Oops. Accepted by UVA, runtime of 0.024s and rank of 73.

Class for the week helped cover a few more things about C++ that I’m not entirely used to yet, though I still made one or two stupid mistakes on the quizzes for the week.

No Comments