Unwanted extra decimals when ingesting XLSX with text

When ingesting an .xlsx file into Dataverse, if a row contains text in a column, the other rows in that column have their numeric values altered with unwanted extra decimals.

Example:

The value 4.31343 becomes 4.3134300000000003.

👉 This behavior is not observed with a CSV file, where numeric values are preserved correctly.

An example of this behavior is available in the following dataset:
https://demo.dataverse.org/dataset.xhtml?persistentId=doi:10.70122/FK2/VKQOK1

Steps to reproduce:

1. Import an Excel file containing both numbers and text in the same column.
2. Check the ingested data.
3. Observe that some numbers have unwanted additional decimals.

Expected result:
Numeric values should be preserved exactly, without extra decimals.

Are you thinking about creating a pull request for this issue?
Our organization does not currently plan to address this issue.

Start by reproducing the XLSX import with the dataset linked in the issue, using a column containing both text and numbers, and compare the result with CSV ingestion. Trace the XLSX ingestion path to identify where numeric values gain extra decimals; done means values such as 4.31343 remain unchanged after import.