Spreadsheet work is mostly plumbing: data arrives as a CSV export from some system, lives in an .xlsx someone else started, and has to go back out in a format the next person can open. RixSheets is built around that reality. Under the Univer grid sits a custom Rust xlsx engine running as a sidecar, and its whole job is reading and writing .xlsx, .xls and .csv faithfully — formulas, formats, styling and sheets included — so files move in and out of RixSheets without drama.
Opening files
Three routes in, all the same result:
- File, Open and pick the file — .xlsx, .xls or .csv.
- Drag the file from Explorer or Finder onto the Rixoffice window.
- Recent files on the Home screen for everything you have touched lately.
For .xlsx and .xls, the engine reconstructs the workbook: multiple sheets in their tabs, formulas ready to recalculate, number formats and fonts in place. For .csv, RixSheets asks about delimiter — comma, semicolon, tab — and encoding before splitting the text into columns, which heads off the classic all-data-in-one-column failure with European semicolon exports.
Encoding is the second half of that conversation. Exports from older systems often arrive in legacy character sets, and a wrong guess turns every accented name into mojibake before any cleanup can start. If the preview shows garbled characters, stop there and switch the encoding rather than fixing names by hand — the difference between a two-second correction and an afternoon of repairing a customer list. Dates, too, deserve a glance on first open: text-formatted dates from some systems sort and calculate as nothing until you convert them, which is a find-and-replace away but easiest before the workbook accumulates formulas that trust them.
What survives the round trip
The Rust engine is built for fidelity on the features that matter:
| Feature | Round-trip behavior |
|---|---|
| Formulas | Read, recalculate locally, written back with the workbook |
| Number formats | Currency, percent, date, custom codes preserved |
| Styling | Fonts, borders, fills and alignment survive |
| Multiple sheets | Tab structure and names kept intact |
| Charts and pivots | Recognized structures come through with the file |
The practical test: open a real Excel workbook, check the totals, save it, and open it again in Excel. That is the bar the engine is aimed at, and it is why Rixoffice vs Excel compatibility questions usually end at the file level rather than the feature level.
Because the engine runs locally as a sidecar, none of this parsing involves an upload. A workbook containing payroll, contracts or anything commercially sensitive is read from disk, calculated in memory and written back to disk — the same machine in, the same machine out. Teams that switched from browser spreadsheets for compliance reasons tend to care about exactly this: the file never had to leave to be editable.
Cleaning an import
CSV exports from other systems arrive messy — that is not a RixSheets problem, it is a systems problem. The first-ten-minutes routine:
- Scan the headers — rename blanks and duplicates immediately.
- Find and replace the recurring typos and trailing spaces; see sort, filter and dedupe.
- Remove duplicate rows with the duplicate removal tool.
- Sort and filter to eyeball each category column for variants.
- Apply data validation to the columns humans will keep filling — see data validation.
- Save as .xlsx — CSV has no formulas, validation or styling, so the working copy should be a workbook.
Keep the original .csv untouched as the receipt; work in the .xlsx.
The routine takes ten minutes on a typical export and less every time after, because the tools are the same five buttons in the same order. Teams that adopt it stop treating imports as events — a fresh dump from the ERP stops being a project and becomes a worksheet that happens to have arrived by CSV.
Round-trip etiquette
Working across Excel and RixSheets smoothly is mostly etiquette, and it is worth spelling out:
- Keep formulas as formulas. Paste values only when the recipient genuinely needs frozen numbers; a live workbook is more useful and more honest.
- Leave the sheet structure alone in shared files — renaming tabs or reordering columns silently breaks everyone else's references and pivots.
- Note anything unusual: a non-standard delimiter in a CSV export, an encoding quirk, a column typed as text on purpose. One line of cell notes saves the next person the rediscovery.
- Save in the format the recipient expects unless you have agreed otherwise; surprise format changes are how compatibility problems start, not how they end.
None of this is RixSheets-specific — it is how workbooks stay trustworthy between any two tools — but the engine makes it painless on this side of the exchange.
Saving and exporting out
Ctrl+S saves in place, keeping the format the file arrived in. File, Save As offers the choices:
- .xlsx for the working copy and anything going back to Excel users.
- .xls only when an ancient system demands it.
- .csv when the destination is another system that eats flat text — one sheet at a time, values only, no formulas or formats by design.
For fixed-layout sharing, skip CSV entirely and export PDF through the print pipeline; the workflow is in print and page layout.
What to explore next
- Rixoffice XLSX spreadsheets — the format itself, in depth.
- Rixoffice CSV support — delimiter and encoding details.
- Sort, filter and dedupe — the cleanup routine for fresh imports.
- Print and page layout — PDF export for fixed-layout sharing.