Line Counter

Line counts matter wherever a system counts rows rather than words: CSV imports, subtitle files, code blocks, address fields, and any form that caps input at a fixed number of lines. In those contexts a line is the unit of failure — one extra row shifts an entire import, and one missing row silently drops a record. Word count tells you nothing useful here.

The recurring trap is line endings. Windows terminates lines with a carriage return and a line feed, Unix and macOS use a line feed alone, and some legacy exports still use a bare carriage return. Text that moves between systems can end up with mixed terminators, which is how the same file produces different line counts in different tools and why a file that validated on one machine gets rejected on another. Normalising endings before counting removes that whole class of confusion.

Blank lines are the other ambiguity. Whether a trailing newline at the end of a file counts as an extra line is a genuine disagreement between tools, and whether blank separator lines should count depends entirely on what you are measuring — they are meaningful in a subtitle file and noise in a word list. Decide which convention you need before comparing a count against a limit, because being off by one is the most common way a submission gets bounced.