Folks I need a concept/idea to start out a project. A software tool my company uses produces a "backup" file in a very unique form. It generates multiple CSV formatted files inside a single source datafile. Each CSV is separated by the original table name within brackets and a blank line. Then the column headers followed by the data lines. This one file has about 100 tables defined in the single source file.
I'm looking for a way to read the source file and break it down into one file physical for each table so I can process them with CSV readers. An example of the source file looks something like this:
Hello,
I attached my solution (workflow.svg):
test.csv - input data
ACCOUNTS.csv - loop output (iteration 0)
CONTACTS.csv - loop output (iteration 1)
Workflow loops through all parts of input file and creates separate files for each ([header] + data), description:
node 126 (126.png) - imports all data into one column (no blank rows)
node 127 (127.png) - creates column (output file name) only with header data, regexp = "^\\[(.*)\\]$" matches only string which starts with "[" and ends with "]".
node 128 - fills empty rows in new column (output file name) with data from previous row (string=previous value)
node 130 - group by new column (output file name)
node 134 (134.png) - creates output file path
node 131 - filters only data column
my remarks:
node 2 - don't replace data column, create new one
node 3 - choose 'previous value' option
node 4 - include only new column (group by)
node 7 - choose flow variable (outputfilupath) as your output path