← All modules

A drawer of string chores — thousands separators, whitespace normalising, lists written out in English, comment stripping, substring counts and two kinds of line wrapping — each sub behind its own export tag.

Version

4.0.2 zef:tbrowder

Depends

File::Temp, Font::AFM, AlgorithmsIT

License

Artistic-2.0

Its own test suite

18 files, green

Checked

2026-09-14 against Raku++ 3.28.0 and Rakudo 2026.08

Where it lives

raku.land · source

Install it #

$ rakupp install Text::Utils

zef install Text::Utils writes the same store; either installer leaves the module usable by both engines.

What it is for #

The small string jobs that every program ends up doing once, badly, in a line of regex: putting commas into a number, collapsing runs of whitespace, turning three items into "a, b, and c", cutting a # comment off a config line, counting how often a substring occurs. This distribution collects a dozen of them with names, and seventeen other distributions on raku.land reach for it rather than rewriting them.

It is one author's drawer rather than a designed API, and the units under it show that — Text::Utils::Subs holds the helpers the main unit is built from, and the seventeen dependents mostly want commify or strip-comment.

The chores #

use Text::Utils :ALL;

say commify(1234567);
say commify(1234567.891, :decimals(2));
say normalize-string("  a   b\t\tc  \n d ");
say normalize-string("a\t\tb  c", :c<s>).raku;
say list2text(<apples pears plums>);
say list2text(<apples pears plums>, :!optional-comma);
say count-substrs('banana', 'an');
say strip-comment('x = 1 # set x', :normalize).raku;
say strip-comment('x = 1 # set x', :save-comment).raku;

1,234,567
1,234,567.89
a b c d
"a b c"
apples, pears, and plums
apples, pears and plums
2
"x = 1"
("x = 1 ", "# set x")

normalize-string trims and collapses whitespace by default and keeps tabs and newlines as they are; :c<s> collapses every run, tabs included, to a single space. strip-comment hands back the line with the comment gone and its trailing space still there unless :normalize is on; with :save-comment you get both halves. list2text writes the Oxford comma unless told not to.

Wrapping comes in two flavours, and the names are close enough to mix up. wrap-paragraph counts characters, which is what a terminal or a text file wants:

use Text::Utils :wrap-paragraph;

my $t = 'The quick brown fox jumps over the lazy dog and keeps running far away from the farm';
.say for wrap-paragraph($t, :max-line-length(32));

The quick brown fox jumps over
the lazy dog and keeps running
far away from the farm

wrap-text is not the same thing with a shorter name: its :width is in typographic points, measured against a font from Font::AFM — it exists for laying out text on a page, and asked to wrap a sentence at 30 it dies with Word 'brown' has length 30.36, too long for max line length of 30 points. For anything that ends up in a terminal, wrap-paragraph is the one.

The one thing to know #

use Text::Utils; imports nothing. Every sub sits behind its own export tag — is export(:commify), is export(:strip-comment) and so on — so a plain use compiles and leaves you with Undeclared routine: commify at the first call. Either name the subs you want, use Text::Utils :commify, :list2text;, or take the lot with :ALL as the first example does. The tags are the sub names, which makes the selective form easy to write and also easy to forget one of.