Hacker News Reader

Comment by 16s | Hacker News Reader

Reversing a string in Python is not obvious nor consistent:

    str[::-1]

While it is in C++:

    std::reverse(str.begin(), str.end());

And Ruby is even more obvious and consistent:

    str.reverse()

Most people who rag on C++ have never seriously used it. It's not that bad. Really, it isn't.

View on HN

That's one point against python, and half a point for C++ [1]. By my accounting that leaves the score still hugely in python's favor when it comes to obviousness and consistency.

Python vs Ruby is another subject entirely. Largely I feel like it's mostly a narcissism of small differences situation, and have only been doing more Python than Ruby lately because more people around me are Pythonistas than Rubyists at the moment.

1] ideally you wouldn't have to tell the reverse function the starting and ending points for the string unless you only wanted the reverse of a substring

begin() and end() return iterators. Python has iterators too, so this concept shouldn't be alien to Python programmers. reverse() is a generic algorithm that operates on any pair of bidirectional iterators.

Actually, in C++, you ideally wouldn't reverse the string at all... you'd just call str.rbegin() and str.rend() to grab the reverse iterators, and pass them straight to your next algorithm.

The two approaches are just as consistent as one another, Python just uses a terser syntax (which, in my opinion, isn't as obvious).

In C++14, if all things go according to plan with Concepts Lite, you'll be able to write a short 3-5 line, reusable, version of reverse() that takes your string (or container) as one argument, deduces at compile time, during overload resolution, that begin() and end() return bidirectional iterators, and then does the right thing. Failing that, C++14 may introduce Ranges. So C++ is only getting terser.

reverse iterate a string? i have never done that

> Failing that, C++14 may introduce Ranges.

It would have been nice to have some sort of simple range syntax added to C or C++ long ago. In the true spirit of C, don't even make it safe, just make it bloody well work.

FirstArray[0..3] = SecondArray[4..7];

case 2..10:

for(int i : 0..ArrayLength)

for(int i : ArrayLength..0)

(hah that'd prove horrible if ArrayLength ended up being a negative number, obviously some proper syntax would need to be determined :) )

I am always annoyed that such simple things are ignored in the language. Sure they don't enable any "cool new abilities", but they make using the language a lot more friendly (especially in comparison to having a ton of case fall through statements!)

The problem is your proposed [] syntax implies random access. Many algorithms and data structures don't require or support random access.

Well, there's always reversed(str) for Python (I realise this returns a reversed object and not a string object, which is something I dislike immensely).

I've always looked at [::-1] as being a bit of a hack, even though we use it all the time for both strings and lists.

How often do you need to reverse a string, or even a list?

that is little verbose, you can save a whole 7 characters.

  reverse(begin(str), end(str));

also you will need #include <algorithm>

Python 3 also has `reversed(str)`.

Reply on news.ycombinator.com