{{ summaryTitle }} {{ summaryValue }} {{ summaryLine }} {{ badge.label }} {{ badge.value }}
Review the parsed table and warnings before using an output in another system.
{{ sourceStatus }}
Auto detect defaults conservatively to tab on a tie; pin the separator when the source contract is known.
Examples include :: or ^; line breaks cannot be used as a field delimiter.
Pin the choice when the first source row could reasonably be labels or data.
Choose the format required by the next workflow; the parsed table and profile remain available for review.
Double quote is the interoperable default for spreadsheet-style delimited text.
This control is inactive when Quote character is None.
{{ workflowFeedback }}
Use 0 for all rows or 10–500 for a shorter on-screen review.
rows
Leave off when spaces are meaningful data.
{{ params.trim_fields ? 'On' : 'Off' }}
Leave off to preserve blank records in the converted output.
{{ params.skip_empty_rows ? 'On' : 'Off' }}
CSV, TSV, Markdown, and HTML continue to preserve text cells.
{{ params.detect_types ? 'On' : 'Off' }}
Valid date-only and offset-bearing date-time values remain in explicit ISO form.
{{ params.detect_dates ? 'On' : 'Off' }}
Leave off when an intentionally empty string differs from a missing value.
{{ params.empty_as_null ? 'On' : 'Off' }}
Leave blank for automatic labels. Duplicate and unsafe names are normalized with stable suffixes.
Review generated SQL against the destination dialect and schema before execution.
Use two names separated by a comma, such as records, record.
{{ values.converted_text || '' }}
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.
JSON Converter Convert JSON or JSON Lines into tables and handoff text with record-path selection, flattening controls and warnings for sparse or risky data.
Extensible Markup Language (XML) Converter Turn XML into structured JSON or record-based tables with local parsing, record-path selection, whitespace controls, and schema 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.

Tab-separated values (TSV) is a plain-text table format in which tabs mark field boundaries and line breaks mark records. It is convenient for copying data between terminals, editors, spreadsheets, and databases because commas may appear inside ordinary text without needing CSV-style quoting.

Plain text does not carry a schema. The first row might be column names or real data, a blank field might mean an empty string or a missing value, and a number-looking identifier such as 001 may need to remain text. Converting TSV safely therefore requires decisions about structure as well as the target syntax.

  • Delimiter choice controls where fields split. A wrong separator can collapse a whole record into one field or create too many columns.
  • Quote policy controls whether delimiters and line breaks can appear inside a field.
  • Header policy determines whether the first row names columns or belongs in the data.
  • Type policy determines whether booleans, numbers, nulls, and dates remain text or become typed values in structured output.

TSV has no single quoting convention as universal as many users expect. A producer may use double quotes, single quotes, backslash escaping, repeated quotes, or no quoting at all. Pinning the known policy is safer than relying on automatic detection when fields contain tabs, apostrophes, or embedded line breaks.

A successful parse only proves that the chosen rules produced a table. It does not prove that columns carry the intended meaning. Row-width warnings, header review, type profiles, and a read-back test in the destination are the checks that prevent a syntactically valid but shifted dataset.

How to Use This Tool:

Match the parser to the source contract, then choose the representation required downstream.

  1. Paste the source or browse for a local text file. Use Auto detect for an unknown separator, or pin tab, comma, semicolon, pipe, or a custom delimiter.
  2. Set Quote character and Quote escaping to match the producer. If the source ends inside a quoted field, correct these choices or repair the source.
  3. Choose Header row. Pin the first row as headers when labels could be mistaken for ordinary text, or treat every row as data when generated column names are safer.
  4. Choose Output format. Enable type and date detection only when JSON, XML, or SQL should carry interpreted values rather than the original text.
  5. Review Parsed rows ledger and Column profile ledger. Resolve irregular row counts and unexpected dominant types before copying the converted text.

Interpreting Results:

The resolved delimiter and header decision explain how the source became a grid. When auto detection reports Tab, Comma, Semicolon, or Pipe, inspect several rows before accepting it. A tie resolves conservatively to Tab, so a one-column or irregular source should be pinned manually.

  • Irregular rows have fewer fields than the widest source row and are padded with empty cells. Check whether the short row is genuinely incomplete or contains an unrecognized quoted delimiter.
  • Dominant type is a column profile, not a schema guarantee. A mostly numeric column may still contain meaningful text exceptions.
  • Unique counts raw text values. It helps spot constant columns and duplicates but does not normalize case or whitespace unless trimming is enabled.
  • Generated SQL is a starting statement. Review identifiers, value types, destination dialect, schema constraints, and transaction handling before execution.

Technical Details:

Delimited parsing scans the source one character at a time. A delimiter ends a field only when parsing is outside a quoted region, and a line break ends a row under the same condition. Repeated-quote mode turns a doubled active quote into one literal quote; backslash mode treats a backslash immediately before the active quote as an escape.

Transformation Core:

Stage Exact rule
Resolve separator Score tab, comma, semicolon, and pipe against the first 20 rows; reward a consistent multi-column width and penalize irregular widths.
Parse rows Honor the selected quote and escape policy, optionally trim field edges, and optionally remove fully empty rows.
Rectangularize Use the widest row as column count and pad shorter rows with empty strings while recording the irregular-row count.
Resolve headers Use or detect the first row, otherwise generate names. Replace unsafe characters with underscores, prefix leading digits, and suffix duplicates.
Interpret values Optionally convert booleans, numbers without ambiguous leading zeros, null-like words, valid ISO dates, and empty cells.
Serialize Write the same parsed rows as CSV, normalized TSV, JSON records, JSON arrays, JSON Lines, Markdown, HTML, XML, or SQL INSERT text.

Automatic header detection requires a non-empty, unique, label-like first row. It is accepted when a familiar label such as id, name, date, or score appears, or when later rows contain typed values while the first row does not. Pin the decision whenever a textual first record could satisfy the same heuristic.

Source text Typed result when enabled Protection against over-conversion
true, false, yes, no Boolean Case is ignored.
null or undefined Null Conversion occurs only when type detection is on.
Finite decimal or scientific notation Number A value such as 001 remains text so leading zeros survive.
Valid YYYY-MM-DD or offset-bearing ISO date-time Date-labeled text The value remains explicit ISO text; no timezone conversion is performed.
Empty field Empty string or null Empty cells as null controls the distinction.

CSV and normalized TSV quote cells containing the active delimiter, a double quote, or a line break, and double embedded double quotes. Markdown and HTML preserve raw text cells with appropriate escaping. JSON records and JSON Lines use the resolved headers as keys. SQL emits quoted identifiers, doubles apostrophes in text, and writes typed numbers, booleans, and nulls without text quotes.

The source limit is 2 MiB. A conversion supports 1 through 500 data rows and 1 through 40 columns. Custom delimiters contain one to four non-line-break characters. The preview limit is either 0 for all rows or a whole number from 10 through 500. The completeness chart shows at most the 12 columns with the most empty cells, while the column ledger retains all columns.

Safety and Privacy Notes:

Source text and files are parsed in the browser and are not sent to a server-side converter. Remove unnecessary secrets or personal records before loading a dataset, especially when the converted text will later be pasted into another service.

Unlike the general table converter, this TSV route does not add a spreadsheet formula guard. Cells beginning with =, +, -, or @ should be reviewed before CSV or TSV output is opened in a spreadsheet.

Worked Examples:

Typed JSON records

Rows headed id, active, and score keep 001 as text, convert true to a Boolean, convert 5 to a number, and can represent an empty score as null. The parsed row ledger still exposes the original text for review.

Two-character separator

A source such as A::B can use :: as a custom delimiter. With no header row, it becomes one two-column data record with generated column names and no lost characters.

References: