When creating a new database via Menu | File | New | Database ==> (*)Connect to an existing Database [Text], (basically, to link to an entire folder full of CSV data files), how to skip the first row of data from being accessed?
The Database Wizard interface for linking to CVS files allows for adjusting some “parsing”-settings, like filename extensions & row format, but nothing for skipping rows.
The CSV file format & content is set by a 3rd party manufacturer, basically outside of my control, and every new CSV data file coming in from the manufacturer will have their format. They are adding meta data into the 1st row, which does not correspond to the column headers in the 2nd row. That in turn confuses the file linking, as it only links to the first column of the CSV files, because it’s looking for column headers in the 1st row, while they are actually in the 2nd row.
I understand that one could write scripts to modify format & content of each and every CSV before linking, but that’s just another layer of complexity to be added, which I’d like to avoid, especially since it seems to exist already in LO:
- The LO Calc interface for importing CSV files has a field allowing for selecting: From Row [ xx ]
- It’s also possible with HSQL statements like: SET TABLE mytable SOURCE "myfile.csv;ignore_first=true;
This [Connect to an existing Database (Text)] function is very awesome for quick process setup for CSV processing, so I hope I’ll be able to figure this out.