Hi hdb,
these are very interesting ideas, but I'm not really convinced by this one:
If the first row has a number, it is not likely to be a header.
The header could consist in years, month numbers, quarters, test IDs, etc., all appearing to be numerical.
@ nysus: in general, a very strong principle is "know your data." Dou you know anything about the data you're going to deal with, or is this just a general purpose tool where you can't know in advance anything about the type of your data?
| [reply] |
| [reply] |
