Meziantou.Framework

Introduction

edit

Meziantou.Framework.TextDiff computes the differences between two texts. It splits both texts into chunks (lines, words, characters, or your own unit), matches the chunks they have in common, and tells you, chunk by chunk, what was kept, deleted and inserted.

What you can do with it

  • Compare two texts line by line, word by word or character by character, or with your own chunks, such as sentences. See Chunking.
  • Ignore what does not matter: casing, the whitespace around each chunk, and \r\n versus \n line terminators. See Comparison options.
  • Refine the changes: find the changed lines, then the changed words inside them, then the changed characters inside those words. See Hierarchical diffs.
  • Choose the algorithm that suits your texts: Myers, Patience, Histogram or Hunt-Szymanski. See Algorithms.
  • Rebuild either text from the result, even when an option made two different chunks compare equal. See Reading the result.
  • Show the result your way: the result is data, so you write it as a unified diff, as HTML or as anything else. See Displaying a diff.

How it works

A diff is a list of entries. Each entry is one chunk of the texts with an operation:

OperationMeaning
EqualThe chunk is present in both texts.
DeleteThe chunk is only present in the old text.
InsertThe chunk is only present in the new text.

Read the entries in order: the Equal and Delete entries give the old text, the Equal and Insert entries give the new text. A change is a Delete followed by an Insert, so a moved block of text appears as a deletion at its old place and an insertion at its new place.

Namespace

Every type is in the Meziantou.Framework namespace: TextDiff, TextDiffOptions, TextChunker, TextDiffResult, TextDiffEntry and the hierarchy types.

Next steps