Back Home.

[swestrup]

Back home now. Second interview appeared to go okay (although it was harder to tell this time how I was doing). The C++ test was fairly straightforward although I was a bit surprised to discover that all the questions were in French. Still, it seems a fair challenge as I'm bound to be handed specs in French from time to time if I get the job, and they did let me answer in English (otherwise I would have been doomed).

There were only two questions where I think I may have flubbed things. I didn't think vector<>.erase() even existed because, as an operation, its an inherently bad idea and so I've never used it. Secondly, part of one question wanted me to draw something in UML and I simply don't know it. So, provided I didn't make loads of other mistakes and not notice, I think I did quite well.

Now we just have to wait and see what the potential employer thinks.

Comments

[sfllaw.livejournal.com]

vector<>::erase() exists for symmetry with other containers. That way, you can use a generic algorithm on it. The authors just assume you know what you're doing in terms of runtime.

[swestrup]

I'd buy that explanation if it also had push_back, push_front, pop_back and pop_front, but it doesn't.

[sfllaw.livejournal.com]

It does have push_back and pop_back!

[swestrup]

Man, the reference I was using doesn't list them!

Okay, that does make more sense then, if they're all in there...

[swestrup]

Oops, I had a braino. I both mistyped and misread. Of course it has push_back and pop_back (I must be getting tired). Thats the one place where modifying a vector makes sense.

I should only have listed push_front and pop_front above. My thinking was that if they weren't there, then it didn't make much sense to assume erase was either.

But, of course, it turns out erase does exist. Oh well.

In any case, its clear that my brain shut down some time ago and I never noticed. I think I should go to bed now.

[designergirl.livejournal.com]

*crossing everything for you* Heheh! Good luck! :)