Fixed Width Record Designer
Choose the byte width and alignment of each field to create a line-delimited fixed-width file or read a local file with the same schema. A character crossing a byte boundary causes a clear error.
Key features
- Set a 1–256-byte width, name, left/right alignment and ASCII pad per field
- Emit and parse up to 200 records with UTF-8 or a pinned CP949 character map
- Reject Korean multibyte splits, field overflow, malformed bytes and unrepresentable characters
- Inspect each field's byte start/end and the first 96 bytes in hexadecimal
- Download a fixed-width .dat file and reusable JSON field schema
- Process local files in browser memory without a network request
How to use
- Edit the sample schema and choose UTF-8 or CP949.
- Set each field's name, byte width, alignment and one-character ASCII pad.
- Enter multiple rows of values and generate the fixed-width file.
- Inspect byte offsets and the preview, then download the .dat file.
- To read an existing file, keep the same schema and select it locally; review the parsed table.
Use cases
- Draft a legacy line-delimited bank or logistics exchange file
- Test whether a Korean CP949 field fits a contracted byte width
- Explain exact byte offsets when exchanging a record schema with a partner
- Find a UTF-8 Korean character split at a field boundary in an existing file
Frequently asked questions
How is byte width different from character count?
One Korean character normally occupies three UTF-8 bytes or two supported CP949 bytes. Widths and offsets here use the actual encoded byte count, not the number of characters.
Are long values truncated to fit?
No. The tool never cuts through a character or silently drops content. It reports the row and field when the encoded value exceeds its width and blocks output.
How are records separated?
An LF byte follows each emitted record. The parser accepts LF or CRLF and also accepts a final record without a newline. Embedded line breaks and a UTF-8 BOM are not supported.
What if the value itself contains the pad character?
Parsing removes that pad character at the selected edge, so identical data characters at that edge cannot be distinguished from padding. Inspect original file bytes when that matters. Downloading an imported file retains its original bytes.
Which CP949 characters are supported?
The tool reuses the pinned WHATWG EUC-KR character index from the existing encoding workbench. Characters outside that mapping, including emoji, and malformed bytes are rejected instead of replaced.
Is my selected file uploaded?
No. Selected file bytes are read only in the browser tab. The tool does not fetch external files or contact a remote service.
Privacy
Field schemas and record values are processed in browser memory. A selected local file is not uploaded; downloaded files contain the values you entered.
Comments & questions