Why splitting on commas is not enough
The CSV format is defined by IETF RFC 4180.
It allows a field to contain the separator, as long as the field is wrapped in quotes. A row reading
"Smith, John",john@acme.com has two fields, not three. It also allows quotes inside quoted
fields by doubling them, and even line breaks inside a field.
Any tool that splits each line on commas gets all three cases wrong, and the failure is quiet: you get a table that looks plausible but has columns shifted on exactly the rows containing commas — usually the rows with names and addresses. This parser implements the actual format, so those rows survive.
How to extract columns from a CSV
- Load your data. Drop a .csv file onto the box, or paste the rows directly. Nothing is uploaded either way.
- Confirm the separator. It is detected automatically by counting candidates outside quoted fields, so semicolon files from European Excel and tab-separated files both work. Override it if the guess is wrong.
- Tick the columns you want. Every column is listed by its header. Tick and untick to see the preview update immediately.
- Export. Download the chosen columns as CSV, copy them as tab-separated values to paste into a spreadsheet, or export a single column as a plain list.
What comes out
- Only the columns you tick, in the original column order.
- Properly quoted CSV, so values containing commas, quotes or line breaks are re-quoted correctly on the way out.
- Tab-separated output on the clipboard, which pastes into Excel or Google Sheets as real columns rather than one long string.
- A single column as a plain list — one value per line, which is the format most import tools and scripts want.
- Optional deduplication across the selected columns, so picking just the email column and removing duplicates gives you a clean address list in one step.
What it accepts
Comma, semicolon, tab and pipe separated files all work, detected automatically. A UTF-8 byte order mark — which Excel adds when saving as CSV — is stripped rather than becoming part of the first header. Pasted text has no size limit beyond your device; dropped files are read up to 400,000 characters, which covers several thousand typical rows.
Excel workbooks are a different format entirely and are detected and reported rather than
mis-parsed — use the Excel data extractor for a
.xlsx.
Why the file never leaves your device
Parsing runs as JavaScript in your browser. Nothing you paste or drop is transmitted.
CSV is the format organisations use to move their most sensitive lists around — customers, employees, subscribers, transactions. An exported CSV is very often the entire contents of a table. Uploading one to a website to keep three columns of it is a disclosure that local parsing simply avoids.
What it does not do
This tool selects and exports columns. It deliberately does not:
- Filter rows by value. There is no "where city = London". Export the columns and filter in a spreadsheet, or use the regex extractor on the result.
- Reorder or rename columns. Output follows the original order and headers.
- Merge or join files. One file at a time.
- Fix genuinely malformed rows. A row with an unclosed quote will swallow following lines, because that is what the format says it means. The row count makes this obvious.
Rows with a different number of fields from the header are kept as they are, with missing cells left blank, rather than being silently discarded.
Who extracts columns from CSVs
- Marketing and sales — pulling just the email column out of a full contact export.
- Developers — reducing a wide export to the two or three fields an import needs.
- Analysts — trimming a file before loading it into a tool that struggles with width.
- Privacy work — producing a copy that deliberately excludes personal columns before sharing it.
- Anyone without Excel — doing a column selection in a browser on any device.
For tips on keeping data intact when moving between CSV and Excel formats, read how to convert Excel to CSV without losing data.
CSV compared with the spreadsheet it came from
If your data is still an .xlsx, work from that instead: the
Excel data extractor reads every sheet and knows about shared strings
and cached formula results, none of which survive a CSV export. Convert to CSV only when something
downstream requires it.
For structured data that is not tabular, the JSON field extractor pulls one field out of every record. For a table on a web page, the HTML table extractor turns it into CSV directly.
CSV format standards and parsing edge cases
RFC 4180 defines the base CSV rules: fields are separated by commas, records end with CRLF, and fields containing commas, double-quotes or line breaks are enclosed in double-quotes with any literal double-quote doubled. Not all CSV files comply: Excel uses the system list separator (a semicolon in many European locales) and many export tools omit enclosing quotes for fields they consider "safe."
Three edge cases: (a) a UTF-8 BOM at the start of the file is stripped before parsing, because Excel writes one and parsers that do not strip it push the BOM into the first column header, corrupting the name; (b) a multiline field — one that contains a literal newline inside double-quotes — is valid per RFC 4180 and is handled correctly, with the newline preserved inside the cell value; (c) a field containing only whitespace is preserved as-is rather than treated as blank, because a single-space value in a database export is intentional and trimming it silently would corrupt the data.
Frequently asked questions
How do I extract one column from a CSV?
Load the file, untick every column except the one you want, then use Download one column as a list for one value per line, or Download CSV to keep the CSV format.
Does it handle commas inside quoted fields?
Yes. That is the main reason this tool exists — the parser implements RFC 4180, including doubled quotes and line breaks inside quoted fields.
Is my file uploaded to a server?
No. Parsing happens entirely in your browser. Nothing you paste or drop is transmitted.
It detected the wrong separator — can I change it?
Yes. Set it explicitly in the Separator dropdown: comma, semicolon, tab or pipe.
Why does my first header have strange characters?
That is a byte order mark from Excel. It is stripped automatically, so if you still see one the file may be in an encoding other than UTF-8.
Can I filter rows, not just columns?
Not here. Export the columns first, then filter in a spreadsheet.
Can I open an Excel file directly?
No — .xlsx is a different format. Use the Excel data extractor, which reads every sheet and exports CSV.
Is there a row limit?
Pasted text is limited only by your device. Dropped files are read up to 400,000 characters, which is several thousand typical rows.