Hi, two questions:

One of my newer custom sources has two lines of data as "headers" in the csv file.  In my custom code, I invoke base subscription reader, etc, but not sure if it will throw warnings or errors when attempting to read a line that doesn't parse correctly.  Is there a way to control how many lines are skipped in reading custom files?

Also, is there a way to invoke custom data so that it redownloads a new file every day in live mode, but only download the file once in backtesting mode (so that during backtesting the custom data source doesn't get pinged millions of times within the span of five minutes)?

Thanks!

-Stephen