Michael Paesold wrote:
> Here is a first version of the patch. It works well for me so far, so
> I am asking for comments now.
Before everyone forgets about my patch in the discussion about the later
patch by Jure Koren...
I have already spent some hours on doing research, coding, creating the
test case. So I am willing to improve my work, of course. It just much
easier with input from the community. I would really welcome any
suggestions or criticism.
While at it, don't you think that we should handle SQL comments in
parseQuery? For example when loading a schema from a file using JDBC,
there might be comments in it that will break the current code. Does the
JDBC spec say anything about that matter?
One other thing that I wanted to try is to not parse the whole query in
a single loop, but instead split up the parsing a bit more, in the style of:
case '"': scan for next '"';
case '\'': scan for matching '\'';
case '$': if dollar-quote, scan for end of dollar-quote
It would simplify the other branches of the switch statement and we
could get rid of the if (!in... && !in...) stuff, then.
Do you think that would be an improvement?
Best Regards
Michael Paesold