This repository contains a template function that being initially developed for easy CSV like content parsing, through generic programming is now able to achieve much more than that. The code solves much of the limitations I've found in other common algorithms:
- Embed the concept of
lines, by giving a way to choose EOL sequence. - Embed the concept of
fieldsby giving a way to choose a fields separator. EOLand the fields are collections (i.e. not a singlechar).- Can skip lines from the beginning (by allowing being used iteratively).
- Can limit the number of lines to be read (by allowing partial parsing).
- Allows multiple unintended usages (i.e. splitting strings, tokenizing...)
The overall complexity in the worst case should be of order N, where N
is the size of the input.
Just use the parse_tuple.hpp file wherever you need it. Some template
features used by the code requires a C++ 17 capable compiler; older
compiler may generate some warning, unwanted behaviour, or not compile
at all. In those cases be sure to run at least the basic functional tests
given.
In the code are given some basic functional tests which demonstrate some use cases and check expected behaviour is confirmed for some edge cases. The tests make use of the Catch2 test framework.
Contribution are welcome in any form they came. I hope this content has been and will be useful to you, and let's this be for other newcomers.
Basically the software is provided as is, without any warranty. You can
use and distribute it freely or commercially provided you don't remove
the license notice and the attributions. See License.md.