While trying to use mysql through sqldb i found some problems trying to apply the updates using TSqlQuery.ApplyUpdates:

Database info:
TableName = mysql_bench
Table structure = CREATE TABLE `mysql_bench` (`AInt` int(11) default NULL,`AFloat` float(13,5) default NULL,`AStr` varchar(100) default NULL)TYPE=MyISAM;

1 - FTableName was storing the ; caracter if it's near to the tablename in sql query. For example Select * from ATable; -> FTablename = 'ATable;'

I fixed it with this

sqldb.pp:699 -> Move(PS^,FTableName[1],(P-PS-1));

but its buggy in the case there's no ; in the end

2- After fixing the table name i got the following:

Sql executed through TSQLQuery.ApplyRecUpdate = insert into mysql_bench (AInt,AFloat,AStr) values
(0,10,123456789,'aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa aaaaaaaa');

Gives this error (also when used with a mysql management tool)

An unhandled exception occurred at $0040F0DD :
EDatabaseError : : Error executing query: You have an error in your SQL syntax.
Check the manual that corresponds to your MySQL server version for the right s
yntax to use near '' at line 1

3- I noticed that could be the Float format so i added the following line in TSqlConnection.GetAsSQLText

sqldb.pp:299 -> ftFloat    : Str(Double(Field.AsFloat),Result);

Now the sql is
insert into mysql_bench (AInt,AFloat,AStr) values (0,1.012345678900000E+001,'aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa aaaaaaaaaaaaaaaaaaa');

But still gives the same error. Although the same sql can be executed with a mysql management tool.

I hope this info can help resolve this bug

Luiz

I'm rewriting this whole parser. It was more a proof-of-concept. It'll get soon better.

I've never tested it with float's. And i'm pretty sure they won't work. I'll have a
Met vriendelijke groeten,

Joost van der Sluis
CNOC Informatiesystemen en Netwerken
http://www.cnoc.nl