Select a CSV file from your device. The browser reads it locally; it is not uploaded to WordPress or SnassiDevs. CSV files up to the current 25 MB limit are supported.

Drag a .csv file onto the upload area. Dropping a file behaves like choosing it manually, and the file remains local to your browser.

Opens your browser’s file picker so you can select a CSV. Loading another file replaces the current CSV state.

Reveals a text area for raw CSV copied from another application. Select Load CSV after pasting; the data stays local and must be loaded before it can be cleaned.

Before loading data, the sample table demonstrates the expected tabular structure. After loading, the preview shows parsed data and is capped at the first 100 rows for performance. Cleaning, copying, and downloading still use the complete dataset—not only visible rows.

Removes leading and trailing whitespace from every CSV cell value. For example, John Smith becomes John Smith. Column headers are not transformed.

Removes rows where every cell is empty or whitespace-only. A partially filled row such as John,,Copenhagen is retained.

Enables duplicate removal. The exact match depends on Duplicate matching; the first matching row is preserved and later matches are removed.

Collapses repeated inline whitespace within cell values, including spaces and tabs. It helps normalize inconsistent imported data while preserving meaningful cell content.

Entire row compares the applicable row values. Selected column uses one chosen column as the duplicate key. In both modes the first match is kept; selected-column matching is exact after the chosen transformations.

Becomes available when duplicate matching is set to Selected column. Choose a field such as Email, ID, URL, or SKU. Rows with an empty value in that column are retained rather than treated as duplicates of other empty values.

No change, lowercase, UPPERCASE, and Title Case apply to data cells. CSV column headers remain unchanged.

Searches all data cells for literal, case-sensitive text and replaces every match. Regex is not used, and headers remain unchanged. For example, finding Denmark and replacing with DK changes those cell values throughout the dataset.

Columns are detected from the CSV header row. Checked columns appear in the cleaned preview and export; unchecked columns are omitted while the original source data remains intact. At least one column must remain selected. A selected duplicate-matching column can still be used even if it is excluded from export.

Applies the selected settings to the originally loaded CSV, not to a previous cleaned result. It updates statistics, creates the cleaned preview, and prepares the complete output for copying or downloading—all locally in the browser.

Shows the number of data rows remaining after cleanup. The CSV header row is not counted.

Shows how many rows were removed by duplicate detection. The number depends on whether matching uses the entire row or a selected column.

Shows how many fully empty rows were removed. Rows with at least one non-empty cell are not counted.

Represents changed surviving rows plus rows removed during cleanup. A row is counted once even if several cells changed; this is not a complete audit log.

Shows the cleaned result with up to the first 100 rows. Copy and Download use the complete cleaned CSV, not just the visible preview. Values are inserted as text so cell content cannot execute as HTML or JavaScript.

Copies the complete cleaned CSV as CSV-formatted text, including headers, all cleaned rows, and only selected output columns.

Generates a cleaned CSV locally with proper CSV escaping, the detected delimiter where possible, and UTF-8 compatibility. A file named customers.csv normally downloads as customers-cleaned.csv; pasted data uses cleaned-data.csv.

Unloads the current CSV, clears pasted data and results, resets statistics and cleanup controls, restores the initial upload state, and affects only this CSV Cleaner instance.

No. CSV files are read and processed locally in your browser. The file does not need to be uploaded to SnassiDevs in order to clean or export it.