Well, that is very interesting. The reason I posed the question was that I
have a large number of instances where TransferText is to be used and, for
practical reasons, it would have been much quicker to get a definitive answer
before planning major code changes. The alternative was to write a test
fragment and ‘suck it and see’, but that was obviously going to take some
time. After waiting a while, I took the latter approach in parallel – belt
and braces just in case no one responded – but, rather disturbingly, my
initial results appear to contradict your answers Obviously I will need to
sort this out before I amend the original code. (It’s in almost continuous
use.)
In my test fragment, TransferText does create the target table and, if the
latter already exists, it replaces it. The first characteristic is very
welcome but the second not so. I need to assemble incoming data, contained
in a series of files, into one large table. It would appear that I have to
input the source documents individually into a series of tables (potentially
up to 75) and then write a corresponding number of append queries to
concatenate these tables into a single master. The snag is that, although I
can use the TransferText method in a For…Next loop to import the source files
into a set of tables, e.g. T1, T2, T3…T75, I can’t immediately see a way of
doing something similar with the following append query. Instead of a single
query in a loop, it appears that I will need 75 virtually identical append
queries to do the job. That’s messy and I feel sure there must be an easier,
or at least a more elegant, solution. If you have any suggestions I would be
most grateful.