ODS to CSV — what OpenDocument holds that CSV has no word for
A mailing tool that only accepts CSV is usually fed by a spreadsheet kept in LibreOffice. The two formats are not two spellings of one thing: OpenDocument is a compressed container with styles, several sheets and formula history, and CSV is a flat text table. Extraction is therefore a subtraction.
Cell formatting, colours, column widths and merged ranges are dropped; only stored values survive.
Processed on your device. Files aren’t sent anywhere.
What ODS holds that CSV has no word for
The container keeps more than cells. Each of these has no counterpart in a delimited file, so it is dropped rather than approximated.
| OpenDocument feature | In the CSV | Why |
|---|---|---|
| Cell styles and fonts | Nothing | a text file has no presentation |
| Conditional formatting | Nothing | there is no rule language in CSV |
| Multiple sheets | Rows only with All sheets | no sheet boundary exists |
| Formula | The cached value | the engine never evaluates expressions |
| Date cell | The displayed text | CSV has no date type |
Choosing which sheets become rows
First sheet writes only the sheet that appears first in the file. All sheets appends every sheet into the same file, one after another, with nothing marking the boundary.
- First sheet suits an importer that expects exactly one header row.
- All sheets suits extraction, but you clean the boundaries afterwards.
- A one-sheet document gives the same output either way.
Formulas in an OpenDocument file
OpenDocument stores the formula and the value last computed. This converter reads the stored value and discards the formula text. It does not recalculate, and it does not follow anything outside the file.
A conversion that recalculated formulas would be a spreadsheet. This one is a reader.
Why the engine never evaluates cell expressions
Dates and cell types in the text output
A date cell leaves as the text shown on screen. Numbers leave as characters too, because cells are read as the formatted text of the sheet rather than as their internal type. An empty cell leaves as an empty field, so columns stay aligned inside a row, while trailing blanks are trimmed from the end of each row.
A date that looks obvious to its author may be ambiguous to the receiving application. The text 03/04 is a good reason to inspect a known record rather than guess the locale. CSV has no date-type declaration. Keep the original OpenDocument workbook for reference and configure the receiving importer deliberately; changing the filename cannot settle a day/month interpretation.
Checking the extracted file
Check the file in a text editor before the mailing tool reads it.
- Convert the .ods file and download the CSV.
- Open it in a text editor so nothing is retyped.
- Count the rows and compare them with the sheet.
- Look for a second header row if you chose All sheets.
- Import only after the counts match.