Hello,
How can i load and parse additional csv files in the Workflow so i can extend or sort the datastream ?
I have some data on which i have to apply some sorting that comes from a third-party software that is run manually at another company; So we send them a file with data for sorting, and they return us the same file with a new column with SORT ORDER and on some projects with a BARCODE fields in the csv.
Thanks for any ideas!
With Connect V1.5, you can create a process that will store the extra fields in the Data Repository. Then, you can load up your original data file and perform lookups in the Repository (I presume you have a uniquely identifiable field for each record) to retrieve the extra fields associated with each specific record.
Hi Phil,
Can the same be used to extract data? If a client gives me a list of account numbers and I need to extract those records with the same account numbers from a master database?
Regards,
S
Of course! However, be aware that the Data Repository is not a full fledged database, therefore you can’t expect the same kind of performance and functionality if you stuff it with thousands and thousands of entries.
Usually requests like this would only be for a few thousand records not in the tens or hundreds of thousands. I checked the Workflow user guide and although it gives you some info about the data repository I don’t see any more detailed info. I see I can manually add groups and keys but how does one populate it automatically? Is there something somewhere where I can get more info about how to use it for extracting certain records from a database? Perhaps a how to that I have not seen?
This will make my day if I can get this going since I was thinking scripting this sort of thing.
Regards,
S
The Push To Repository task is used to store data in the repository. It will also create the structure if it doesn’t exist yet. To retrieve information from the Repository, you use the Lookup function in any of the data-enabled fields in any plugin. Just right click on the field and use the Get Repository Location option.
There is a full API for the repository that allows you to retrieve lists of items, create/delete groups and keys, but we are waiting just a bit longer before releasing it be cause in its early stages it was bound to change slightly and we didn’t want to run into potential backward compatibility issues for users that would have used it early on.
Struggling a bit. I I try to add the Push to Repository plugin to a process there is nothing to select so when clicking the ok button I get a error stating that I must select a group. After adding a group to the repository manually and then trying to add the plugin again I get a error stating that I must specify at least one Value. The plugin doesn’t give you a chance 
Perhaps I should leave it alone till the API is released.
Regards,
S
The structure of the Group/Keys must already have been defined manually through the Data Repository Manager.
The “Create Group and Keys if they don’t exist” option is only there when you are sending your config to another Workflow Server on which the manual operation may not have been performed.
By default, a new KeySet will be added, using the values you specify in the GUI. But you can also elect to update an existing KeySet, in which case you can select which Keys to update and you have to specify a condition to identify which KeySets to update (e.g. “ID=1234”, or “NumberOfPages>10”)
Hello,
The condition has no influence on which rows get updated I believe
I have the condition only update record id = 1.
But it still finds 2 rows and updates both.
What am I doing wrong?
Have you tried specifying another condition, just to see how the operation reacts?
You should create a separate Group for testing, manually add a few Keysets and then run your process in step by step mode to see if you can get any condition to work (which you should!)