Topic: Using multiple files for a datastore
Using multiple files for a datastore
Hi folks,
brand new forum member, just downloaded the Data Cleaner and started playing around with it and love it so far.
Though, a few wonderings on my side. I've tried adding multiple csv files as one data store and found out that this is not quite possible. I've managed to add several csv files and then creating a "Composite datastore", which works just fine, but is there another way or am I missing something? If not that - is there a feature to filter the datastores? (eg. *.csv, db*, etc.)
Another problem I ran into is column naming. From what I see the DataCleaner takes the first row as header, but what happens if, like in my case, there are no column headers? Taking the first row is not really neat and I can not find a way to rename columns? Any suggestions in that scenario?
brand new forum member, just downloaded the Data Cleaner and started playing around with it and love it so far.
Though, a few wonderings on my side. I've tried adding multiple csv files as one data store and found out that this is not quite possible. I've managed to add several csv files and then creating a "Composite datastore", which works just fine, but is there another way or am I missing something? If not that - is there a feature to filter the datastores? (eg. *.csv, db*, etc.)
Another problem I ran into is column naming. From what I see the DataCleaner takes the first row as header, but what happens if, like in my case, there are no column headers? Taking the first row is not really neat and I can not find a way to rename columns? Any suggestions in that scenario?
Hi there,
The composite datastore is for working with multiple datastores yes. What do you mean by "filtering" of datastores? What is the use case?
The header issue is a known one. There is work being done to allow a "no header" option where you columns will just get names like "Column A", "Column B" etc.
The composite datastore is for working with multiple datastores yes. What do you mean by "filtering" of datastores? What is the use case?
The header issue is a known one. There is work being done to allow a "no header" option where you columns will just get names like "Column A", "Column B" etc.
Hi Kasper,
thanks for the answer.
The thing is, that just for one project I need to use about 10-12 csv files, imagine what a mess it would be if I have more than 5 projects and have to create datastores for all of them. I`d either have to register all the input files, create the composite data store and then remove the input files (not sure if that would even work) or scroll like crazy to sort everything out.
It would be much cooler if we can have a filter on the top and can just say "Hello, Mr. DataCleaner, can You filter out all the datastores that contain 'projectX" in its name? Thank you very much!"
thanks for the answer.
The thing is, that just for one project I need to use about 10-12 csv files, imagine what a mess it would be if I have more than 5 projects and have to create datastores for all of them. I`d either have to register all the input files, create the composite data store and then remove the input files (not sure if that would even work) or scroll like crazy to sort everything out.
It would be much cooler if we can have a filter on the top and can just say "Hello, Mr. DataCleaner, can You filter out all the datastores that contain 'projectX" in its name? Thank you very much!"
Actually I like that idea ... A text field that will work as an on-the-fly filtering mechanism.
I've added a feature request in the bugtracker for it: http://eobjects.org/trac/ticket/595.
I've added a feature request in the bugtracker for it: http://eobjects.org/trac/ticket/595.
Log in by clicking the login link at the top of the screen
Go back to forum.
