I have been tasked with connecting to an FTP site, retrieving a csv file, parsing that file into multiple Excel worksheets within one workbook and then FTPing the Excel spreadsheet back to another FTP site.
I have all the component pieces working, in 6 different workbenches, the download of the original csv file, the parsing and writing of the data for each Excel sheet and creation of the four individual worksheets and the upload to the destination FTP site. However, I have been advised by the local FME Admin that the process should be housed in one FME workbench. And in order to achieve that, instead of downloading a physical csv file, I should be downloading the data as an attribute so that all of the subsequent processes could be housed within one FME workbench and the attribute data shared by each subsequent process, leaving no artifacts at the conclusion of processing.
I have created one FME workbench and copied all of the processes into it and made all the adjustments to User Parameters and such, however, I am not sure how to switch from a physical file to an attribute for the initial downloaded file. I modified the FTPCaller to download to an Attribute but I am unclear where to go from here. The FME Admin mentioned an Attribute Exploder. But that doesn’t seem to be what is needed. I can see the data from the file in the file contents of the resulting download. I am unclear how to access the data and continue on with the next process with that data.
TLDR; I am trying to access the data and all the fields as an Attribute, rather than a physical file that has been downloaded.
Is there an example somewhere of what I am trying to do that I can review?