Re: How to read batches of SQL from a file in Perl

You can also make the iterator a little simpler by changing the record separator:

local $/="\ngo\n";
while(<IN>) {
    chomp;
    # now the entire statement is in $_...
    ... execute the SQL statement, with error checking, etc.
}
[download]

The disadvantage is that $/ can't be a regex, so you can't handle upper/lower case, for example.

Michael

Comment on Re: How to read batches of SQL from a file in Perl Download Code

Replies are listed 'Best First'.
Re^2: How to read batches of SQL from a file in Perl by KurtSchwind (Chaplain) on Oct 25, 2007 at 02:05 UTC
Wouldn't it be nicer and more flexible to allow people to specify the seperator? I like the idea of essentially slurping the files in and specifying the break as a NEWLINEgoNEWLINE as you did, but I think that a semi-colon is more common. I realize that the stated assumption was that the statements end with a 'go', but it would be such a minor change to make it so much more flexible.	[reply]
Re^3: How to read batches of SQL from a file in Perl by erix (Prior) on Oct 25, 2007 at 14:32 UTC
A similar module is SQL::Script, which seems like a good idea (and it has the separator :)	[reply]
Re^3: How to read batches of SQL from a file in Perl by mpeppler (Vicar) on Oct 26, 2007 at 14:02 UTC
The default for Sybase isql scripts is to end with "go", so it makes sense to do it this way. Other separators can of course be used - it all depends on what your purpose is. Michael	[reply]