III. Text Processing
In the first part of The Textsmith Primer, we discovered the enduring qualities of plain text.
In the second part, we learned how markup languages give that text structure, transforming simple characters into documents, websites, books, diagrams, and many other forms of digital expression.
A natural question now follows.
Once information has been written as structured text, what can we do with it?
The answer is text processing.
Few ideas have shaped modern computing more profoundly than the realization that text is not merely something to read. It can be searched, transformed, analysed, organized, combined, and automated.
This insight lies at the heart of the Unix tradition.
Rather than relying upon a single application that attempts to solve every problem, Unix offers a workshop of small, focused tools. Each performs one task exceptionally well. Together, they allow the textsmith to build workflows that would be difficult—or impossible—to achieve within a single monolithic program.
Throughout this part, we explore not only the tools themselves but the habits of mind they encourage.
Searching teaches us to discover information.
Regular expressions teach us to recognize patterns.
The stream editor teaches us to transform text.
awk teaches us to compute with information.
Sorting and comparison reveal hidden relationships.
The shell teaches us to compose simple ideas into larger workflows.
Structured data formats demonstrate how ordinary text can become portable databases.
Desktop search expands our view from individual files to entire libraries of knowledge.
Finally, automation shows how all these ideas come together into repeatable, reliable practices.
The goal of this part is therefore larger than learning a collection of Unix commands.
It is to learn how a textsmith thinks.
Every tool answers a question.
Every question develops a habit.
Every habit becomes part of a craft.
In This Part
17 The Unix Toolbox: Small Programs, Big Results Introduces the Unix philosophy in practice, showing how independent tools cooperate through plain text.
18 Searching the Text Universe: grep, ripgrep, and Friends Explores the art of discovering information quickly within large collections of text.
19 The Art of Pattern Matching: Regular Expressions Introduces the language of patterns that allows computers to recognize forms rather than merely matching words.
20 Transforming Text: sed and the Art of the Stream Shows how text flowing through a pipeline can be transformed automatically without manual editing.
21 awk: When Text Becomes Data Explores how text can be analysed as structured information through records, fields, and computation.
22 Ordering the Text Universe: sort, uniq, cut, paste, join, and comm Demonstrates how organization reveals relationships hidden within collections of information.
23 The Shell: Where Tools Become Workflows Examines how pipes, redirection, and scripts transform individual commands into complete applications.
24 Text Databases and Structured Data Introduces CSV, TSV, JSON, YAML, SQLite, and the idea that many databases can remain portable, human-readable, and remarkably simple.
25 Searching Knowledge: From find to Full-Text Search Expands searching from individual files to complete libraries through indexing and desktop search.
26 Automation and the Philosophy of Text Brings together the entire Unix toolbox, showing how thoughtfully composed workflows become lasting expressions of craftsmanship.
Entering the Forge
If the previous part taught us how plain text gains structure, this part teaches us how structured text gains power.
Here we enter the forge of the textsmith.
Ideas are searched.
Patterns are recognised.
Documents are transformed.
Knowledge is extracted.
Workflows are composed.
By the end of this part, the Unix toolbox will appear not as a collection of unrelated commands, but as a coherent philosophy of working with information.
Welcome to the forge.