Open a CSV of millions of rows on a Mac.

Where the usual tools stop

• Excel holds at most 1,048,576 rows on a sheet. A longer file opens with the rest cut off, and saving it loses them for good.
• Text editors can open large files but know nothing about columns — no sorting, no filtering by column, no sums.
• Many CSV editors read the whole file before showing anything, and hold every value as a separate object in memory. A few hundred megabytes is then several gigabytes.

What Intact CSV Editor does

It reads enough rows to fill the window before the window appears, and streams the rest in behind it while you scroll, edit and search. A 325 MB file of 5.2 million rows is on screen in well under a second and fully read about eight seconds later. Values are held as UTF-8 bytes rather than as separate strings, so those five million rows take about 186 MB for the table instead of 816 MB.

Measured on a release build on the Mac the app is developed on.

While the file loads, anything that needs the whole table — saving, exporting, sorting, SQL — waits for the last row, so a half-read file is never written out truncated.

Working with a big file

1. Look first. Scroll, search and filter; filtering and search run off the main thread, so the window stays smooth.
2. Count it. Table ▸ Statistics counts every column: filled, empty, distinct, minimum, maximum, median.
3. Ask it. The SQL console answers grouped questions in a fraction of a second once the table is loaded.
4. Cut it down. Extract the filtered rows to a new file, or File ▸ Split into Several Files for pieces of a fixed size.

When the file is bigger than memory

Intact CSV Editor holds the table in memory, compactly. For files of many gigabytes — tens of millions of rows, or more than your Mac's memory — a tool that reads the file without loading it is the better choice: a memory-mapped editor such as Caxton, a viewer built on DuckDB such as Tad, or command-line tools such as qsv and csvkit. We compare them honestly on our comparison pages.