Created attachment 203501 [details] To be imported data Download the attached csv file and the "DataProviderTarget.ods" file to your PC. Open the ods file. Menu > Data > Data Provider. In that dialog do these: * in section `Source` select the database range "myImport". * Click on the `Browse` button and select the downloaded csv file. * in section `Transformations` select entry `Split Column` and click on `Add` button. You get additional fields. * In field `Separator string` insert a pipe symbol, that is | . In field `Column index` enter 1. Click the `Apply` button below the Preview. * in section `Transformations` click on `Add` again to insert the next `Split Column` transformation. * Enter the pipe symbol in the field `Separator string` and value 2 in `Column index` field. Click the `Apply` button below the Preview. The preview looks goood. Click on OK button. You should have got a list with 100 popular male and female English names. Make sure you save to extended ods format. Save the document to a new name. When you examine the saved file, you will find a <calcext:data-mappings> element in the content.xml, that contains all the settings you have made in the dialog. Reopen the saved file. Error: The "Split column" transformations are not applied and thus the entire record is in the first column.
Created attachment 203502 [details] The target document for Dataprovider import
The import looks strange. The export writes a structure <calcext:data-mappings> <calcext:data-mapping ...> <calcext:data-transforms> <calcext:.../> <Calcext:.../> </calcext:data-transforms> </calcext:data-mapping> <calcext:data-mappings> That is, the export writes the transformations as child element of a single data-mapping. But the import in ScXMLMappingsContext::createFastChildContext() has ScXMLTransformationsContext on same level as ScXMLMappingContext.
I can confirm with Version: 26.2.0.0.alpha0+ (X86_64) / LibreOffice Community Build ID: 2595f031fa93c1eb89fb4dce6f337de9be813e15 CPU threads: 4; OS: Linux 6.8; UI render: default; VCL: gtk3 Locale: cs-CZ (cs_CZ.UTF-8); UI: en-US Calc: threaded First column contains all the data (for example "1|James|Mary"), second contains second column (James) and third contains third column (Mary).
I think, I've found the reason: https://gerrit.libreoffice.org/c/core/+/192929. The patch has no unit test, thus I have not submitted it.
Regina Henschel committed a patch related to this issue. It has been pushed to "master": https://git.libreoffice.org/core/commit/48cf7f1123637be847140ef9acc49e8f5f056677 tdf#169019 write char not number, read col from 0 It will be available in 26.2.0. The patch should be included in the daily builds available at https://dev-builds.libreoffice.org/daily/ in the next 24-48 hours. More information about daily builds can be found at: https://wiki.documentfoundation.org/Testing_Daily_Builds Affected users are encouraged to test the fix and report feedback.
Unit tests will be made with further patches.
Regina Henschel committed a patch related to this issue. It has been pushed to "libreoffice-25-8": https://git.libreoffice.org/core/commit/4ff6d294eac41288bce19b377c0a8857b8cb4992 tdf#169019 write char not number, read col from 0 It will be available in 25.8.4. The patch should be included in the daily builds available at https://dev-builds.libreoffice.org/daily/ in the next 24-48 hours. More information about daily builds can be found at: https://wiki.documentfoundation.org/Testing_Daily_Builds Affected users are encouraged to test the fix and report feedback.