Subject | Re: [ib-support] Select syntax |
---|---|
Author | Doug Chamberlin |
Post date | 2001-12-22T03:55:55Z |
At 12/21/2001 09:59 PM (Friday), Claudio Valderrama C. wrote:
What we are seeing here is typical of parsers which are generated via
software. Since the parser can tell the delimiters are separate tokens it
does not require the spaces. No matter that most people would consider this
syntax wrong. I don't know anyone who, on a SQL syntax test, would consider
"SELECT*" correct or who would intentionally write it that way!
While on the subject of the parser, I would sure love to see the
identification of tokens in error messages done in terms which are more
useful to the user. The "line X, character Y" counting is very difficult to
understand!
>I'm going to paraphrase Doug: unfortunately, our DSQL parser sucks. :-)Thanks, Claudio! I'm so glad I did not have to repeat myself. VBG!
What we are seeing here is typical of parsers which are generated via
software. Since the parser can tell the delimiters are separate tokens it
does not require the spaces. No matter that most people would consider this
syntax wrong. I don't know anyone who, on a SQL syntax test, would consider
"SELECT*" correct or who would intentionally write it that way!
While on the subject of the parser, I would sure love to see the
identification of tokens in error messages done in terms which are more
useful to the user. The "line X, character Y" counting is very difficult to
understand!