Fixed Width to CSV Converter
Split aligned fixed-width records into clean CSV. Auto-detect column breaks from whitespace, or provide exact widths, ranges, and names for repeatable imports.
About this tool
Fixed-width text is common in mainframe exports, bank files, old reporting systems, COBOL-style extracts, and command output where columns are aligned by position instead of separated by commas. This converter turns those records into CSV without uploading anything.
Leave the column spec blank to auto-detect boundaries from whitespace that lines up on every row. For production imports, provide an explicit spec so the same layout is used every time: widths such as 10,4,*, one-based ranges such as 1-10,11-14,15-30, named widths such as name:10,age:4,city:*, or pipe-separated position,length,name entries such as 1,10,name|11,4,age|15,*,city.
Worked example
Input:
name age city
Ada 36 London
Bo 7 Oslo
With auto-detection and the default first-row header mode, the result is:
name,age,city
Ada,36,London
Bo,7,Oslo
For a repeatable import, use the spec name:10,age:4,city:* and set the header mode to “Use names from the column spec”. That reads characters 1-10 as name, 11-14 as age, and the rest of each line as city.
Limits and edge cases
- Positions and widths are counted in Unicode characters, not UTF-8 bytes.
- The rightmost
*column reads to the end of each line, so longer notes are not clipped. - Short rows are padded with empty CSV fields.
- Auto-detection treats a character position as a separator only when every row has whitespace there, so explicit specs are safer for irregular files.
- A run converts up to 50,000 data lines and 512 columns.
quote=neveris intentionally lossy if a field contains the selected delimiter; use minimal quoting for normal CSV imports.
FAQ
When should I use an explicit column spec instead of auto-detect?
Use an explicit spec when the file layout is known or must be repeatable. Auto-detect is convenient for quick report output, but a real fixed-width feed should usually be parsed with widths or ranges such as 10,4,* or 1-10,11-14,15-30.
Are column positions zero-based or one-based?
Specs use one-based positions because that is how most fixed-width layout documents describe columns. For example, 1-10 means the first through tenth characters. Width specs such as 10,4,* start at character 1 and advance automatically.
Can I create TSV or semicolon-separated output?
Yes. Set the delimiter to tab, semicolon, pipe, space, colon, or any single character. The output still uses the same CSV quoting rules, so fields containing the delimiter are protected when quote=minimal or quote=all is selected.
Why did the first input line disappear from the data?
The default header mode treats the first line as column names. If your input has no header row, choose “Generate col1, col2, …” or “No header row”. If your spec has names such as name:10,age:4, choose “Use names from the column spec”.
Developer & Automation Access
Run it from the terminal
Same engine as this page, headless — via the gizza CLI:
gizza tool fixed-width-to-csv "name age city
Ada 36 London
Bo 7 Oslo"New to the CLI? Get gizza →
Open it by URL
Pre-fill and auto-run this tool with query parameters — the names match the API/CLI:
https://gizza.ai/tools/fixed-width-to-csv/?text=name%20%20%20%20%20%20age%20city%0AAda%20%20%20%20%20%20%20%2036%20London%0ABo%20%20%20%20%20%20%20%20%20%207%20Oslo&spec=name%3A10%2Cage%3A4%2Ccity%3A%2A&header=first-row&trim=true&delimiter=comma"e=minimal&newline=lf&skip_lines=0&comment=%23&skip_blank=true&bom=trueMachine-readable descriptor: tool.json — title + parameters JSON Schema for agents.