Perl — a learning library

Perl for text processing, measured. One idea per page. Every claim on every page is backed by a program that runs, whose output a tool pastes into the page and CI checks again, on Ubuntu and on macOS — so nothing here is quoted from perldoc and hoped for.

Perl 1.0 came out in December 1987 as a better way to pull reports out of text files, and it is still the tool with the shortest distance between "this file is wrong" and "this file is fixed": a one-liner. Its Unicode support arrived later, in layers — usable in 5.8 (2002), consistent only from 5.12 (2010) when asked for, with fc in 5.16 (2012) — and every layer kept the old behaviour as the default. Most of what surprises people about Perl and text follows from that: the correct answer is almost always available, and almost never the one you get by typing the least.

Start here

00 — Start here — what this library assumes, how a page works, and the five pages to read first.

The chapters

	Chapter	What it covers
01	One-liners	-n, -p, -l, -a, -F, -i and -0 as the lines of code they are — and the loop they build
02	Unicode text	use utf8, "Wide character", decoding at the edges, what length counts, and fc
03	Regex	the Unicode bug, /g in two contexts, the counts s/// and tr/// return, and split
04	Records and fields	the \r that chomp leaves, CSV, sorting, and numbers that arrive as text
05	Resources	the perldoc pages behind each chapter, and the sibling libraries

Running the examples

You need perl 5.36 or later and nothing else — no CPAN modules, no build tool. On a Mac that means Homebrew's perl: /usr/bin/perl is 5.34, one release too old for use v5.36. Every example is a single file, run from its own folder:

cd 03_Regex/split_has_sharp_edges/examples
perl split_pl.pl

To run all of them and check every recorded output:

python3 tools/run_examples.py --check

Written against perl 5.42, which CI pins on both machines. Where a fact depends on the version, the page says which version.

Sibling libraries

Same house style, same answer-key contract. Each page here links the page in these that teaches the same idea in another language:

• Encodings ↗ — bits, bytes, code points and UTF-8 from the bottom up. Read it first if "code point" is a new word; this library assumes it.
• Python ↗ — str and bytes: the same boundary, drawn as two types that will not mix.
• Java text ↗ — a language that copied Perl's split and chose UTF-16.
• Ruby text ↗ — the language that kept Perl's -n, -p, -a, -F, -i, split and pack, and labels every string with an encoding where Perl keeps one flag.
• C ↗ and Rust ↗ — text with no encoding in the type at all, and text with the encoding enforced by the compiler.

The one rule

No page hand-types what a program prints. A lesson marks the spot and the runner fills it from a real run — so an example that changes behaviour on a new perl breaks the build instead of quietly making a page wrong. See CONTRIBUTING.md.