CSV Repair
Share tool
Only this public address is included. Your files, inputs and results stay out of the link.
Fix a CSV import that shows broken accents, uneven columns or everything in one column. Review the detected settings and table before exporting.
Preparing your tool…
A visual example
Separate columns without splitting a decimal number
A comma can belong to a number. In this file, the semicolon separates the columns.
- Sample CSV
id;amount
0012;-12,50
Inspect the source characters
The row contains two fields: ID 0012 and amount −12.50, written with a decimal comma.
Confirm the two different rules Output value Type 0012 Text −12.5 Number Confirm the two different rules
Choose semicolon as the delimiter. Keep the ID as text; convert the amount only with decimal-comma interpretation confirmed.
The column delimiter and the decimal separator are different settings.
Illustrative example. CSV itself does not store spreadsheet cell types. To protect leading zeros, use explicit text cells in XLSX or import that CSV column as text in your spreadsheet.
Read the practical guide and worked example
How to use it
- Choose the original CSV or use the example file.
- Inspect encoding, separators and warnings, then apply explicit corrections or edit cells.
- Choose text or numeric columns, review export options and confirm the preview.
- Create and download the verified CSV or XLSX.
0012 stays text; selecting the amount column as numeric converts -12,50 to the number -12.5 with a comma decimal setting.
Details & limits
CSV up to 5 MiB, 10,000 rows, 256 columns and 100,000 cells. No recovery of already lost digits.
Can it recover already lost leading zeros?
No. It preserves characters present in your source. CSV cannot enforce spreadsheet column types; import identifiers as text or use explicit XLSX text cells. Formula protection intentionally adds an apostrophe in CSV, while XLSX text cells preserve the original characters.
Separate an import problem from missing source data
CSV carries characters and separators, not a spreadsheet's full type system. Before repairing a file, identify whether the problem is decoding, column boundaries or the receiving application's interpretation. Work from the original export so an earlier incorrect import does not become your new source.
Everything appears in one spreadsheet column
Consider the record 0017;-7,50;"Rue, Nord". With a semicolon separator it has three cells; the comma in the quoted note belongs to the note. Selecting a comma just because the file is called CSV would be the wrong fix. Confirm the separator against several records, including a note containing punctuation or a line break.
Keep product IDs while making amounts usable
Treat 0017 as text and choose only the amount column for number conversion. With a decimal comma, -7,50 becomes the numeric value -7.5. XLSX can store the identifier explicitly as text and the amount as a number. If you need CSV instead, control the destination's import types too; the file cannot force that application to retain leading zeros.
Check your result
- Inspect accented names and several columns after choosing the encoding; readable ASCII alone does not establish the right choice.
- Compare row widths, multiline notes, empty cells and long identifiers before confirming repairs or an export.
- Review formula protection: a CSV value such as =1+1 receives an apostrophe when protection is enabled; this is an intentional content change.
If something looks wrong
A row has fewer cells
Padding can add empty trailing cells; it cannot identify a missing value or move shifted data back into the correct column. Check the source record and quoted fields before accepting padding.
An ID already reads 17 instead of 0017
Return to the original export or source system. This tool cannot infer how many zeros were lost. Choose text handling before importing the replacement file.