How to read a line-by-line text diff

Short answer

A line-by-line diff finds a common sequence between two texts and marks what disappears or appears around it. It does not understand the meaning of code or prose by itself.

-timeout=30
+timeout=45
 retries=3

The first line was removed, the second added, and the third remains equal. A replacement is commonly displayed as one deletion followed by one addition.

Reading the result

1. Find the unchanged lines that provide context.
2. Review removals before their related additions.
3. Check whether the change is content, whitespace, or a line ending.
4. Evaluate its effect in the source format.

The text comparator runs locally and sends neither document to a server.

Changes that are hard to see

key=value and key=value are different lines: the latter has a trailing space. Tabs, letter case, Unicode normalization, and LF versus CRLF can also differ. Automatic normalization can hide a real issue, so it should be a deliberate choice.

A diff does not replace other checks

• For JSON, validate and format first if you want a readable structural layout.
• A fingerprint can confirm equality of exact bytes or normalized text, but it cannot show where a change occurred.
• For code, run tests too: a small textual change can alter behavior.
• Line comparison is not appropriate for binary files.

Large comparisons

Diff algorithms consume memory and time according to line count and the shape of the changes. Split very large inputs or use specialized tools when you need move detection, word-level comparison, syntax awareness, or version history. A simple line diff is especially useful for configuration, command output, and text fragments that fit human review.

Primary source

The POSIX diff specification defines file comparison and its standardized output formats.