Comparison options

By default, two chunks are equal when they are the same string. Three options of TextDiffOptions relax the comparison:

The options combine, and they apply to hierarchical diffs too. They change what compares equal, not the text of the entries: an Equal entry keeps the chunk of each text in OldText and NewText, so you can still show or rebuild the exact texts.

Casing

IgnoreCase compares chunks with an ordinal, case-insensitive comparison: the result is the same whatever the current culture.

Whitespace

IgnoreWhitespace trims each chunk before comparing it. Only the edges are trimmed: the whitespace inside a chunk still counts.

var options = new TextDiffOptions { IgnoreWhitespace = true };

// The whitespace at the edges of a line is ignored...
Assert.False(TextDiff.ComputeDiff("    indented\n", "indented  \n", options).HasDifferences);

// ...but not the whitespace inside it
Assert.True(TextDiff.ComputeDiff("a  b\n", "a b\n", options).HasDifferences);

To ignore the whitespace inside the lines too, compare words. TextChunker.Words puts each run of whitespace in its own chunk, which trimming reduces to an empty chunk, so Hello world and hello world compare equal:

var options = new TextDiffOptions
{
    Chunker = TextChunker.Words,        // Lines (default), Words, Characters, or custom chunker
    IgnoreCase = true,
    IgnoreWhitespace = true,
    IgnoreEndOfLine = true,
};

var result = TextDiff.ComputeDiff("Hello   world\r\n", "hello world\n", options);
Assert.False(result.HasDifferences);

Line terminators

IgnoreEndOfLine replaces every line terminator with \n before the texts are split into chunks. A file saved with Windows line terminators then compares equal to the same file saved with Unix ones:

var options = new TextDiffOptions { IgnoreEndOfLine = true };

var result = TextDiff.ComputeDiff("one\r\ntwo\r\n", "one\ntwo\n", options);

Assert.False(result.HasDifferences);

// The entries hold the normalized text, not the original "\r\n"
Assert.Equal(
    [
        new TextDiffEntry(TextDiffOperation.Equal, "one\n"),
        new TextDiffEntry(TextDiffOperation.Equal, "two\n"),
        new TextDiffEntry(TextDiffOperation.Equal, ""),
    ],
    result.Entries);

Since the texts are normalized before they are split, the entries hold \n rather than the original terminators. Unlike the other options, the entries no longer rebuild the original texts exactly, only their normalized versions.