Excel writer unnecessarily changing names of Named Ranges

Hi ​@btl,

Thank you for posting your question on the FME Community!

Sorry to hear you’re running into issues with the Excel Writer and named ranges. Unfortunately, I wasn’t able to recreate that warning message on my end using the information provided so far.

Would you be able to share a small sample workspace that reproduces the warning? That would help us better understand the setup and understand what’s triggering the behavior.

Hi ​@crystalwang , thanks for taking the time to look at this for me.  i attach one of my workspaces

• In the attached, the source is Excel, but my production version reads from a linux based postgres db (someone suggested Linux may have hidden line return characters causing my issue, but I get the same behaviour reading from a spreadsheet; and anyhow, the names of the Named Ranges are hard coded into the workspace not read from the db).
• In my production version, the Named Ranges already exist in my target spreadsheet, but I get the same behaviour writing to a completely blank spreadsheet.
• My overall workflow consists of several workspaces, similar to this, which are run from a separate workspace using FMEFlowJobSubmitters - but I get the same behaviour running the workspaces separately.

​@crystalwang - have you had any luck reproducing my error.  I am finding that it only seems to occur with ranges whose name starts with something similar to my naming convention:

So the follow produce the error:

• A_1_Named_Range
• A_1_1_Named_Range
• AA_1_Named_Range
• AA_1_1_Named_Range

But the following do not:

• AAA_1_Named_Range
• AAA_1_1_Named_Range
• AAAA_1_Named_Range
• AAAA_1_1_Named_Range
• AA_A_Named_Range

You’ve only shared the fmw file not the source data. If you want to include the test spreadsheets you should save as a template file and upload that

Here’s a file I was testing with, that includes an FME template and a spreadsheet with existing ranges that I might write to.

This seems to be a bug introduced in 2025.2 (I’ve corrected my original post that says I was using 2025.1) I have tested in 2025.1 and don’t experience the same issue.

It looks like the warning is new, but it’s actually alerting you to something that previously would have gone unnoticed until you try and open the excel file. So it’s not a bug but an enhancement

Named ranges cannot have a name that could be interpreted by Excel as a valid range address in either A1 or R1C1 notation e.g. C, C2, AA44, R12C13, etc are invalid

If you create a simple file with a named range e.g. C1_Test in 2025.1 if you open the excel file you will find you get a “we found a problem message”. If you then try and recover you’ll get a message that the Named range has been removed.

I'm not sure that's right - the named ranges that cause this issue always have an underscore between the letter and the number (A_1)  so they don't follow the standard A1 form of ranges.

They are also valid in Excel, I initially create them in excel and then import my writer definitions from the output file.

It doesn’t seem like quite the whole story, as you say it looks like if you enter them directly in excel names like AAA_1_Named_Range are accepted.

But if i try and write that in FME 2025.1, I get no error in FME, but an error on opening the excel file

Doing the same in FME 2025.2, I get a warning in FME, the name gets changed and the excel file will open but with the amended named range.

2024.1 has no error and creates the file with the named range correctly. 

So perhaps a bug after all, just that the 2025.1 behaviour isn’t necessarily more desirable than the 2025.2 behaviour...