Upload multiple CSV files at once and convert each one to JSON. Download results individually or grab them all as a ZIP.
// Converted results will appear here
The first row is treated as the header and each subsequent row becomes a JSON object, keyed by those headers. Quoted values containing commas are handled correctly, so a field like "Smith, John" won't get split into two columns. Files are processed locally in chunks — nothing is uploaded to a server, and the parser yields control back to the browser periodically so large files don't freeze the tab.
Each file is capped at 8 MB. This isn't arbitrary — browser-based parsing and rendering genuinely can't scale to unlimited file sizes without risking an unresponsive tab, so files over the limit are rejected up front rather than attempted. For larger datasets, split the file into smaller chunks first (most spreadsheet tools can do this via "save as" with a row range, or a quick script).
CSV files exported from Excel are often saved in a legacy encoding — EUC-KR/CP949 for Korean, Shift-JIS for Japanese — rather than UTF-8, or sometimes as UTF-16 ("Unicode Text" export). This tool checks for a byte-order mark and sniffs for UTF-16 automatically, then falls back through UTF-8 and EUC-KR, so most Excel exports should decode correctly without any manual step on your end. See this post for a deeper walkthrough of why the same file opens fine in one program and turns to gibberish in another.
This almost always means a field contains an unescaped comma or a stray quote that breaks the quoting rules mid-row. Open the source file in a text editor and check the row in question — a single unmatched " can throw off every column after it for the rest of that row. This post covers a related issue: even a perfectly-parsed CSV loses type information once it becomes JSON, since every CSV value starts life as a string.
This is expected, not a bug. JSON repeats every field name in every object (CSV only states it once, in the header row), so JSON output is reliably 2–4x the size of the source CSV for typical data. For files over 1,000 rows, this tool skips pretty-printing automatically to keep that overhead from doubling again.
Not currently — this parser expects standard comma-separated values. Semicolon-delimited files (common in European Excel locales) will parse as a single column per row rather than splitting correctly.
An empty cell becomes an empty string ("") in the resulting JSON, not null and not an omitted key — every row's object has the same set of keys, even where a value is blank.
Yes — select or drag multiple CSV files at once, and they're each converted and listed separately, with an option to download everything together as a single ZIP.