parsercsv: Minor refactor to prefer standard algorithms and datatypes #10871
Conversation
| for (int row = 1; row < tokens.size(); ++row) { | ||
| if (locationColumnIndex < tokens[row].size()) { | ||
| locations.append(tokens[row][locationColumnIndex]); |
There was a problem hiding this comment.
would you mind making this range-based for as well?
tcoyvwac
force-pushed
the
fix/refactor/prefer-standard-library-datatypes-parsercsv
branch
2 times, most recently
from
09ccbe6
to
5429ad4
Compare
src/library/parsercsv.cpp
Outdated
| std::for_each(std::next(std::begin(tokens)), | ||
| std::end(tokens), | ||
| [&](const auto& token) { | ||
| if (locationColumnIndex < token.size()) { | ||
| locations.append(token[static_cast<int>( | ||
| *locationColumnIndex)]); | ||
| } | ||
| }); |
There was a problem hiding this comment.
Mhmm. the code has just gotten more complex. I didn't anticipate that. My fault. Would you mind reverting the change again. Sorry.
There was a problem hiding this comment.
Yeah 😿 , the Mixxx project does not use the range-v3 library yet (hint-hint! 😉)... Then the code would be as simple as:
for(const auto& token: ranges::views::tail(tokens)) {
if (locationColumnIndex < token.size()) {
locations.append(token[static_cast<int>(
*locationColumnIndex)]);
}
}and many more nice things... 🥲 (range-v3 stuff will arrive into the standard in C++23)
I agree with you jointly that a range-based approach should be the default writing-style / default way to go.
Will revert back to the previous raw-for-loop form.
There was a problem hiding this comment.
Yeah... but expect to wait a bit for that... It was already a pain to switch to C++20...
You're welcome to add the ranges-v3 library as a sort of polyfill to mixxx if you'd like. Do you have any good resources to recommend to learn about ranges? I don't know any mixxx maintainer that knows how to use them (me included).
There was a problem hiding this comment.
Interesting question @Swiftb0y, humm, it really depends on one's C++ knowledge.
The tl;dr form is:
- Understanding Standard Library algorithms and why they are important (
C++03minimum).
- 1.1 Wikipedia - Generic programming - Stepanov–Musser and other generic programming paradigms
- 1.2 Bjarne Stroustrup - What is "generic programming" and what's so great about it?
- Understanding the concept & reasoning behind boost::range (Basically
range-v2).
(Why is thisrange-v2? From the History section of boost::range:)
This version introduced Range Adaptors and Range Algorithms. This version 2 is the result of a merge of all of the RangeEx features into Boost.Range.
- 2.1 boost::range - Introduction
- 2.2 boost::range - Range Adapter - Introduction & Motivation
- 2.3 boost::range - Range Algorithm - Introduction & Motivation
- 2.4 Bonus Algorithms Resource: - Christian Aichinger - Boost Range For Humans
- Understanding about
range-v3
Some bonus stuff:
- Sean Parent - GoingNative 2013 - C++ Seasoning
"No raw loops" and "That's a Rotate! 😄"
(Sadly,) This is a quick writeup, so thats all for now! You can always playtest with the concepts about ranges on godbolt.org.
Hope this helps and guides on your ranges-concept journey, ye? 😸
But beware, this programming-paradigm path is quite addicting and could lead you to some "heavier stuff"....! 🙃
There was a problem hiding this comment.
Thanks a bunch. I'll definitely check those links out when I find the time.
tcoyvwac
force-pushed
the
fix/refactor/prefer-standard-library-datatypes-parsercsv
branch
from
5429ad4
to
b563ed9
Compare
tcoyvwac
force-pushed
the
fix/refactor/prefer-standard-library-datatypes-parsercsv
branch
from
b563ed9
to
3ae9de6
Compare
|
Thanks @Swiftb0y for the... swift PR review and merge! Very grateful ye? 😸 🎉 |
tcoyvwac
deleted the
fix/refactor/prefer-standard-library-datatypes-parsercsv
branch
As titled. 🎉