Newline Tests

Free text files with LF (Unix), CRLF (Windows), CR (classic Mac) and mixed line endings, for testing conversion and cross-platform compatibility.

A text file using LF (Unix)

lf.txt / 704 B

A text file using CRLF (Windows)

crlf.txt / 714 B

A text file using CR (classic Mac)

cr.txt / 704 B

A text file using Mixed newline conventions

mixed.txt / 708 B

Why line-ending tests matter

Line endings differ across OSes: Unix/Linux/macOS use LF (n), Windows uses CRLF (rn), and Classic Mac OS used CR (r).

Use these to verify line-ending handling in version control (Git), text-processing tools, and programming languages.

📖 Where people get stuck

The same content written with LF (Unix), CRLF (Windows), CR (classic Mac) and a deliberate mix. Line endings change behaviour while staying invisible, which puts them among the harder text-processing bugs to trace. You cannot tell by looking, so trying the real thing is the short path.

Case	What happens	What to do
git diff shows every line as changed	Change nothing at all and every line still differs if the endings went from LF to CRLF. An editor setting or the operating system does it silently.	Put `* text=auto` in `.gitattributes` and pin the files that care — `*.sh text eol=lf` for shell scripts and config.
A shell script with CRLF will not run	The trailing is handed to the shell as part of the command. `bash: $' ': command not found` and `syntax error near unexpected token` are the usual symptoms.	Run `file script.sh`; `with CRLF line terminators` confirms it. Strip them with `tr -d ' ' < in > out`. Safer still, convert on the server rather than copying straight from Windows.
A stray survives at the end of each string	Reading a CRLF file line by line, many languages treat only as the separator and leave the in the value. Comparisons stop matching and numeric parsing fails.	Trim as soon as you read. `rtrim($line, " ")` in PHP, `line.rstrip()` in Python. When the stray bytes are mid-file rather than at the ends, normalise with the line ending converter.

Mixed files are the awkward case. LF through the first half and CRLF through the second is a perfectly ordinary result of two people editing on different machines. The file command reports only the first form it finds, so counting with grep -c $' ' is the reliable way to spot a mix.