Able to use the row id number as unique identifier when import a file

[alexpavel]

It would be really useful if the row id number ("Line #") would be available as additional column when import a csv file or from any other external source.

When import files in Anaplan, usually the first step is to import the file into a numbered list where the file columns are transformed in numbered list properties.  To be able to load the file usually the “Choose properties below to build the unique identifier for the imported items” option is used where are chosen the combination of columns that identify uniquely a row in the file. 

Sometimes identifiying the "primary key" of a file is not always possible. 

This way it would be possible to use the row id (row number "Line #") of the file as a property to be used to be able to choose as unique identifier. 

This row id it seems that is already available in the "back-bone" of Anaplan becasue when rejection file is created there is the information of the row number in the "Line #" column. 

It would be very useful if the row number from the file would be available to be chosen as an option for “Choose properties below to build the unique identifier for the imported items”

Of course there is the workaround to ask the provider of the file to create this additional column, but also, sometimes in a project this could become really difficult...

Thanks

Alex

[ben_speight]

If the source data is denormalized - that is, with no set of properties that define a composite key - then maintaining the data becomes impossible as you cannot identify each item. Incorporating the row number of a source file still leaves a maintenance headache - you'd have to ensure a subsequent import source file contained the items in exactly the same order, provide an empty row in lieu of any removed items, and require any new items to be appended after the existing ones.

[alexpavel]

Hi Ben, 

Yes, I agree with you. The denormalized data cannot be "re-used" in the model using the rowid as "matching key" between 2 uploads of the file.

It's impossible and far too risky to rely on the same order of the rows in a file to load it twice in the Anaplan model. 

In these cases the numbered list where the file is loaded should be deleted before another import would be done. 

 Immagine this scenario:

- file has no primary key

- the scope is to import data by aggregation of some numeric column, but the aggregation key need in Anaplan to be manipulated (i.e. it is a result of a concatenation of 2 columns or add a sufix/prefix, etc.)

- or the scope is to filter some data before importing in the final Anaplan modules/lists

In this cases, having the possibility of the "rowid" of the file as primary key it would be very useful follwing these steps:

- delete existing elements in the numbered list

- load data in a numbered list using rowid as "primary key" and as properties the other columns of the file

- create a module based on this numbered list

- manupulate the in the module applying the required modifications (like concatenation of columns, filter data, etc...)

- create action to import from numbered list module into final modules/lists,  

What do you think ?

Ciao

Alex