Re: Efficient walk/iterate along a string

Perl is a tool that handles complicated string manipulations well, but it's not designed to handle individual characters near so well. For that you need something closer to the silicon like C. However, on a case by case basis it may be that you don't need to handle individual characters and instead can use index or a regex to do much of the heavy lifting. The alternative may be to invest the time in writing an XS module that provides an iterator to efficiently traverse a string returning a character at a time and providing the current index on demand.

True laziness is hard work

Comment on Re: Efficient walk/iterate along a string

Replies are listed 'Best First'.
Re^2: Efficient walk/iterate along a string by DrHyde (Prior) on Nov 23, 2010 at 10:22 UTC
Please do this - I've occasionally had to solve similar problems, and thought "this is where perl needs pointers", but never had the tuits to actually implement it.	[reply]
Re^3: Efficient walk/iterate along a string by Anonymous Monk on Nov 23, 2010 at 10:49 UTC
I throw links at you :) String::Index - Perl XS module for strpbrk()/index() hybrids Pointer - Object Oriented Memory Pointers in Pure Perl PeekPoke - Perl extension for reading and writing memory Pack/Unpack Tutorial (aka How the System Stores Data) Algorithm Showdown: Fuzzy Matching Regexes are slow (or, why I advocate String::Index) To read Char-by-Char from a file	[reply]
Re^3: Efficient walk/iterate along a string by GrandFather (Saint) on Nov 23, 2010 at 19:35 UTC
I suspect that if there was a real need for this and an XS solution does actually provide substantial speed gains over substr or the reverse and chop trick then it'd have already been done. Most times a smarter algorithm gives much more bang for the buck than tinkering with an implementation to make the old algorithm a little faster. True laziness is hard work	[reply]