Compare line endings and non-breaking spaces in a UTF-8 file
Admin ·
Use the original and marked outputs to check whether a text field changed during import. A line break can look the same on screen while using different code points in the file.
Find the three line-ending styles in one sample
Select Load example. The first row contains a TAB between Item and Qty, followed by CRLF. The next two rows end in LF and CR respectively. A separate Unicode LINE SEPARATOR occurs before the Hebrew word.
In the summary each of CRLF, LF and CR has a count of one. CRLF groups two code points into one occurrence. For a file of your own, use Open UTF-8 text file before comparing endings. Editing the text in the browser field can convert carriage returns to line feeds.
| Break | Code points | Marker |
|---|---|---|
| Windows-style pair | U+000D U+000A | CRLF |
| Line feed | U+000A | LF |
| Carriage return | U+000D | CR |
| Unicode line separator | U+2028 | LS |
Unicode 17.0 property list including White_SpaceWHATWG textarea API value and newline normalization
Compare the spaces around the quantities
The gap between Tea and 2 has one NBSP at U+00A0. The gap after Coffee has one NNBSP at U+202F. The regular space before 3 is U+0020. Keep Non-breaking spaces selected and turn off Regular spaces to narrow the preview and report.
Choose Download marked text to check the added labels, then Download original for the exact source. If you need to preserve words and quantities together in typeset text, keep the non-breaking spaces. If a machine import expects an ordinary space, compare its input requirements before editing.
Unicode 17.0 property list including White_SpaceUnicode line breaking and non-breaking spaces
Use the download that matches the job
Use Download original to compare file bytes after an import. The tool retains a UTF-8 byte-order mark when present. Use Download JSON report for searchable character labels, counts and source positions after category filtering.
Download marked text for a human review. Its bracketed labels show where selected characters occurred. Text beside the labels can still contain unselected marks, so use the JSON code points when a mark needs a precise identification.
Unicode 17.0 property list including White_SpaceUnicode Character Database and derived properties
Common questions
Why is a normal space counted separately from a non-breaking space?
They have different Unicode code points and line-breaking behavior. The tool leaves the two unchanged and reports them with separate labels and counts.
Why is the selected count lower than the detected count?
The detected count covers every supported mark. The selected count follows the category filters used in the report.
Sources
Put it into practice.
- Whitespace Visualizer
- Zero-Width Character Detector
- Text cleaner & line sorter
- Word & character counter
Related guides
- Find Zero-Width Spaces and Keep Emoji Joiners
- How to Remove Duplicate Lines and Sort Text
- Export Word Counts and Vocabulary Statistics
- Trace Why Git Still Ignores a Kept File
- Keep an Exception in a Next.js Repository
- Check Prerelease Order and npm Range Matches
- Check GSM-7 and Unicode SMS Segment Boundaries
- Convert Line Endings and Limit Blank Lines
- Identify a Legacy Receipt’s Text Encoding
- Split a Radio Brief into Sentences
- Keep Replies While Cleaning an Email Thread
- Fix PDF Wraps and Keep Compound Hyphens
- Reflow Hard-Wrapped Paragraphs