Migration of Document Attachments from APPS10 to IFS Cloud

Hi Deshan,

Thank you for your quick and informative response. We appreciate the guidance on the available migration options.

As you pointed out, the Cloud File Storage Migration Tool is primarily designed to facilitate the migration of documents from a database repository to File Storage. However, our challenge is a bit different and significantly more complex.

We are tasked with migrating over 1.6 million files from a shared repository in our current APPS10 environment to IFS Cloud. To perform the Tech Doc mentions about a text file, and we need to add the details of each document to the text file, these files are stored in a shared file repo, and our primary concern is how to accurately and efficiently gather all the necessary metadata for each document. This metadata includes document number, class, revision, file name and path

A significant portion of our documents are named using result keys, which complicates matters further, as these filenames do not contain the metadata required for the migration. 

Given the scale of our migration and the complexity of the file structure (which includes multiple subdirectories), The process of extracting this information seems to be the most critical and challenging part of our migration effort.

Could you please advise on any tools, scripts, or methodologies that might help streamline this process? Additionally, if there are any best practices or considerations we should keep in mind when dealing with such a large-scale file migration, we would greatly appreciate your insights.

Thank you once again for your support and guidance as we navigate this complex migration.

Best regards,  
Azhar

Hi @ZOKAAMEER 

Can you elaborate a bit more on the file naming on shared repo? What do you mean by result key? Can you give an example of each variation you have?

In IFS Document Management, we name the files saved in such repositories in a unique key combination, this is called as the repository file name
format: DOC_CLASS - DOC_NO - DOC_SHEET - DOC_TYPE. docx

Does the above match your file naming? Or is there a different kind of name handling?

Thanks @deshan.ekanayake 

A small correction: the default naming scheme for the repository file name is DOC_CLASS-DOC_NO-DOC_SHEET-DOC_REV-FILE_NO. EXTENSION

That is, there are no spaces between the keys, document type (ORIGINAL or VIEW, in most cases) is not part of it, and the file number is there as well.

When in doubt though, check the file_name column in edm_file_tab. That's what we use when accessing files on FTP, Shared and File Storage repositories.

@ZOKAAMEER

Creating that text file is not hard, it can be done by any SQL tool and all the information you need is there in the table edm_file_tab.

Yes @Mathias Dahl @deshan.ekanayake 

From the below query I was able to get values in a correct way 
and i was able to a sample file migration too thanks for the replies 

SELECT 
    REPLACE(doc_class, ',', '\,') || ',' || 
    REPLACE(doc_no, ',', '\,') || ',' || 
    REPLACE(doc_sheet, ',', '\,') || ',' || 
    REPLACE(doc_rev, ',', '\,') || ',' || 
    REPLACE(doc_type, ',', '\,') || ',' || 
    REPLACE(file_no, ',', '\,') || ',' || 
    REPLACE(file_name, ',', '\,') || ',' || 
    REPLACE(user_file_name, ',', '\,') || ',' || 
    REPLACE(path, ',', '\,') AS combined_values
FROM 
    edm_file_tab

@ZOKAAMEER 

Thanks for replying back, and sharing that SQL.