[Data Merge] MacRoman special characters corrupted after updating data source in InDesign 21.5.1
Product version: Adobe InDesign 21.5.1
Operating system: macOS 26.5.2
Last working version: Adobe InDesign 21.4.1
Description:
Since updating to InDesign 21.5/21.5.1, Data Merge incorrectly decodes special characters in legacy MacRoman tab-delimited TXT data sources.
The same data sources and InDesign documents work correctly in InDesign 21.4.1. In InDesign 21.5.1, special characters such as German umlauts, ß and accented characters become corrupted when “Update Data Source” is used.
Selecting the source with these import settings does not resolve the problem:
- Delimiter: Tab
- Encoding: ASCII
- Platform: Macintosh
Re-selecting the data source with the same options also does not reliably restore the characters.
Steps to reproduce:
- Create a tab-delimited TXT data source encoded as Western Mac OS Roman/MacRoman.
- Include values containing characters such as ä, ö, ü, Ä, Ö, Ü, ß and é.
- Connect the file through the Data Merge panel.
- Modify the data source without changing its encoding.
- Choose “Update Data Source” in InDesign 21.5.1.
- Check the merged-data preview.
Expected result:
The updated data is read using the original MacRoman import settings, and all special characters remain unchanged.
Actual result:
After updating the data source, special characters are decoded incorrectly and appear as unrelated characters.
Regression:
The identical documents and data sources work correctly in InDesign 21.4.1 on the same Mac. Rolling back from 21.5.1 to 21.4.1 resolves the problem immediately.
This may be related to the new UTF-8 and multiline Data Merge implementation introduced in InDesign 21.5. Please ensure that updating a data source continues to respect legacy encoding and platform settings.