{{ summaryTitle }}
{{ summaryValue }}
{{ summaryLine }}
Source{{ sourceBadge }} Target{{ targetLabel }} Review{{ warningCount ? warningCount + ' note' + (warningCount === 1 ? '' : 's') : 'Clear' }}
Choose Auto detect for mixed paste and file work, or pin an exact parser.
Paste or load one CSV, TSV, JSON, JSONL, Markdown, TXT, or HTML table source.
{{ sourceMeta }}
{{ fileStatus }}
Examples: ^ or ::.
Choose the document needed by the next spreadsheet, API, database, or documentation step.
Examples: ^ or ::.
Headers become JSON keys, Markdown labels, HTML headings, and preview columns.
Turn this option on when output header row is required.
{{ params.include_headers ? 'Include headers' : 'Rows only where supported' }}
Turn this option on when trim cell edges is required.
{{ params.trim_cells ? 'Trim parsed cells' : 'Preserve edge whitespace' }}
Turn this option on when spreadsheet formula guard is required.
{{ params.formula_guard ? 'Guard formula-like cells' : 'Keep exact leading characters' }}
Preserve keeps the detected labels unchanged.
Minimal quoting is the neutral interoperability default.
Leave empty for ordinary blank cells, or enter a token such as NULL.
Use 10 to 1,000 rows; the converted document keeps every row.
rows
{{ convertedDocument }}
#{{ header }}Copy
{{ row.index }}{{ cell.value }}
CheckValueEvidenceCopy
{{ row.label }}{{ row.value }}{{ row.detail }}
Comma-Separated Values (CSV) Converter Convert CSV-like text into structured formats, review row alignment and column completeness, and keep sensitive source data in your browser.
Tab-Separated Values (TSV) Converter Convert TSV and delimiter-separated text into typed table formats with local processing plus clear header, row-width and column-profile checks.
JSON to CSV Converter Turn JSON or JSON Lines into spreadsheet-ready CSV or TSV with controlled record selection, nested-value shaping and formula-risk warnings.
XLSX to CSV Converter Convert one Excel worksheet to CSV in your browser, choose formula and hidden-data handling, and review warnings before download.

A table is more than rows separated by punctuation. Its meaning also depends on which row contains column names, how quoted delimiters are handled, whether blank and missing cells are distinct, and what each destination expects. A clean-looking paste can still shift values into the wrong columns or turn a label into executable spreadsheet input.

Conversion is safest when parsing and serialization are treated as separate jobs. First, the source becomes a rectangular grid with stable column names. Then that grid is written as delimited text, Markdown, JSON, JSON Lines, or HTML. Reviewing the intermediate rows catches alignment mistakes before they become a polished but incorrect document.

Source family Structural concern
CSV, TSV, and other delimited text Delimiters inside quoted cells, inconsistent row widths, and ambiguous first rows.
JSON and JSON Lines Records may expose different keys; nested values must be represented without losing their text.
Markdown and HTML tables Separator rows, escaped characters, header cells, and markup must become plain cell values.

Headers deserve special attention because downstream systems often bind data by name. Blank labels need replacements, duplicates need stable suffixes, and names may need conversion to snake_case or camelCase. Automatic header detection is a convenience, not proof; a first row made of ordinary words can plausibly be either labels or data.

No universal table conversion can infer schema, locale, or business meaning from text alone. Decimal commas, identifiers with leading zeros, line breaks inside cells, and formulas require an explicit decision. A row preview and column-coverage check are therefore part of the conversion, not optional decoration.

How to Use This Tool:

Choose how the source should be parsed before choosing the document needed by the next system.

  1. Paste the table or load one local text file. Leave Source format on Auto detect for mixed work, or pin CSV, TSV, Markdown, JSON, JSON Lines, HTML, or a custom delimiter when the source contract is known.
  2. Set Header handling. Use the first row when it contains labels, generate column names when every row is data, or inspect the automatic choice in the audit.
  3. Choose Target format and the header naming policy. For custom delimited text, enter a one- to four-character delimiter.
  4. Keep Spreadsheet formula guard on when output may be opened in a spreadsheet. Turn off trimming only when spaces at cell edges are meaningful data.
  5. Review Row preview, Column coverage, and the conversion warnings before using the converted document. Fix a wrong parser or header choice at the source rather than editing shifted output by hand.

Interpreting Results:

Start with row and column counts. They should match the source after the chosen header row is removed from the data. Missing cells are positions absent from a short row, while blank cells are present but empty; both can look empty in the final document, but they reveal different source problems.

  • A renamed-header warning means duplicate or unsafe labels were made unique. Check downstream field mappings before handoff.
  • A guarded-cell count means formula-like text was prefixed with an apostrophe. The visible value is preserved for spreadsheet review, but the serialized text has changed deliberately.
  • An unclosed-quote warning means the parser reached the end while still inside a quoted field. Correct the source instead of trusting the resulting rows.
  • When quoting is disabled, a delimiter or line break inside a cell can make later imports split the record incorrectly.

Technical Details:

The conversion model normalizes line endings, parses the selected syntax into rows, decides how headers are obtained, pads short rows to the widest row, and then serializes one canonical grid. Changing the target format does not change that parsed grid.

Transformation Core:

Stage Rule Material result
Detect Prefer recognizable HTML, JSON Lines, JSON, and Markdown structures before scoring comma, tab, semicolon, and pipe delimiters. A resolved source format and a review note.
Parse Honor doubled quotes in delimited text, ignore a Markdown separator row, decode text from the first HTML table, and collect JSON record keys in encounter order. Rows of text cells before header handling.
Normalize Promote or generate headers, make names unique, pad short rows, optionally trim cell edges, and substitute the selected missing-cell token. One rectangular grid with stable column names.
Protect When enabled, prefix cells beginning with =, +, -, @, tab, or carriage return with an apostrophe. Formula-like text is less likely to execute when opened in a spreadsheet.
Serialize Apply the target's quoting, escaping, header, and markup rules without reparsing the data. The converted document plus preview and audit evidence.

JSON object input may be a single object, an array of objects, or an object containing an array under data, items, results, records, or rows. Values that are themselves objects or arrays are kept as JSON text inside a cell. JSON object output and JSON Lines always use object keys even when the output-header-row option is off.

Target Serialization rule
Delimited text Minimal quoting surrounds cells containing the delimiter, a quote, a line break, or leading or trailing whitespace; embedded double quotes are doubled.
Markdown A header and separator row are always required. Pipes are escaped and cell line breaks become <br>.
JSON records or JSON Lines Each row becomes an object keyed by the normalized headers; cell values remain text.
JSON row matrix Rows remain arrays, with an optional header array at the beginning.
HTML Headers and cells are HTML-escaped before being placed in table markup.

Input is limited to 1,000,000 characters, and a browsed file must be no larger than 1 MiB. Custom delimiters contain one to four characters. The row preview accepts 10 through 1,000 rows; limiting the preview does not truncate the converted document.

Privacy Notes:

The source is read and converted in the browser. It is not sent to a server-side conversion service. Local processing does not make a dataset harmless, so remove secrets and personal data that are unnecessary for the task.

Formula guarding reduces a common spreadsheet risk but is not a complete content-security policy. The receiving spreadsheet, database, documentation system, or API should still validate imported values according to its own rules.

References: