Comments on: Word Counting in C++: Extracting words from camelCase symbols

By: Jonathan Boccara

Jonathan Boccara — Thu, 08 Nov 2018 12:17:00 +0000

In reply to Colin Pitrat. Great, thanks for the suggestion! I will try it in the code.

By: Colin Pitrat

Colin Pitrat — Sun, 04 Nov 2018 19:27:00 +0000

“it duplicates the code performing the search of the beginning of a word, and I’d be glad to hear your suggestions to improve it”

Well, you can simply use an infinite loop & break:

auto beginWord = begin(code);
for(;;)
{
beginWord = std::find_if_not(endWord, end(code), isDelimiter);
if(beginWord == end(code)) break;
auto endWord = std::find_if(std::next(beginWord), end(code), [](char c){ return isDelimiter(c) || isupper(c); });
words.emplace_back(beginWord, endWord);
}

By: Jonathan Boccara

Jonathan Boccara — Tue, 23 Oct 2018 20:36:00 +0000

In reply to flowing. I see your point. Then in the example of XCoordinate and YCoordinate, how many time do you think we should count the word Coordinate? 0 or 2?

By: flowing

flowing — Sat, 20 Oct 2018 08:58:00 +0000

The camel case splitting would split ALL_UPPERCASE_VARIABLES into single letters and bigrams with underscores. At least, it would have to check if the preceding character is lowercase to work properly.

By: jft

jft — Wed, 17 Oct 2018 08:46:00 +0000

“suggestions to improve it, and potentially to simplify it by using STL algorithms”

Well ‘simplification’ is in the eye of the beholder, of course, But this can be done as a 1 liner

std::vector getCamelCaseWordsFromCode(std::string const& code)
{
auto words = std::vector {};

for (auto beginWord = begin(code), endWord = beginWord; (beginWord = std::find_if_not(endWord, end(code), isDelimiter)) != end(code); words.emplace_back(beginWord, endWord = std::find_if(std::next(beginWord), end(code), [](char c) { return isDelimiter(c) || isupper(c); })));

return words;
}