Question

FeatureReader Excel keeps old Column Names

  • 27 January 2022
  • 3 replies
  • 2 views

Hello i have Columnname A and B.

I read the Excelfile with a FeatureReader. Get Columns A and B.

Now i close FME. Rename Columname A to C in the Excelfile.

Open FME. Read the Excelfile. Get Columns A and B instead of C and B.

Why does FME cant see my renamed Column C and still sees Column A wich doesnt exist anymore?


3 replies

Userlevel 4
Badge +25

Those kinds of changes will require you to redo the FeatureReader, there doesn't seem to be a way to update it (like you would have with a regular reader). Also any subsequent transformers that rely on the column that you have renamed will need to be updated.

Those kinds of changes will require you to redo the FeatureReader, there doesn't seem to be a way to update it (like you would have with a regular reader). Also any subsequent transformers that rely on the column that you have renamed will need to be updated.

If i have 2 Excel Files in a directory.

Excel File 1 has the following columns:

A B

Excel File 2 is the same Excel but the renamed column A so:

C B

 

If i have a directory and filepathnames reader and read both path_windows and then i use a testfilter to only get Excel File 1.

Then i use the FeatureReader with the path_windows Attribute as Source.

As "to reading Featuretypes" i set it to a whitespace. So in the cache i get "no shema" but the attributes are hidden. A and B. so i could expose them.

If i change my testfilter to get Excel File 2 and read it with the FeatureReader then i get "no shema" again and now there are C and B hidden.

So is there a way to make all hidden columns visible without specifically naming the columns that needs to get visible?

Userlevel 4
Badge +25

If i have 2 Excel Files in a directory.

Excel File 1 has the following columns:

A B

Excel File 2 is the same Excel but the renamed column A so:

C B

 

If i have a directory and filepathnames reader and read both path_windows and then i use a testfilter to only get Excel File 1.

Then i use the FeatureReader with the path_windows Attribute as Source.

As "to reading Featuretypes" i set it to a whitespace. So in the cache i get "no shema" but the attributes are hidden. A and B. so i could expose them.

If i change my testfilter to get Excel File 2 and read it with the FeatureReader then i get "no shema" again and now there are C and B hidden.

So is there a way to make all hidden columns visible without specifically naming the columns that needs to get visible?

You'd want to look into a dynamic workflow but you still would have to know what the attributes are called if you want to do something with them.

There's a section in the 2019 Advanced training material about Dynamic translations that could be helpful.

Reply