i have the following SQL query that works fine (returns multiple rows) when i run it from Oracle's SQL developer tool:
select * from [schema].[table] where resource_unique_name in ('abc123') and (slice_date >= '01-Nov-10') and (slice_date < '01-Dec-10')
but somehow this code below returns 0 records (no errors just, 0 records):
OracleDataReader reader = new OracleDataReader();
const string CONNECTION_STRING = [myConnetionString];
connection.ConnectionString = CONNECTION_STRING;
connection.Open();
OracleCommand command = connection.CreateCommand();
command.CommandText = [same SQL as above];
OracleDataReader reader = command.ExecuteReader();
int rowCount = 0;
while (reader.Read())
{
rowCount++;
开发者_StackOverflow社区 }
Console.Write(rowCount);
the connection open's fine because if i hard code a different SQL query i do get results.
is there any reason a query would work fine in SQL query tool but not from my C# code ??
UPDATED NOTE: in my code, when i remove the:
and (slice_date < '01-Dec-10')
it seems to work fine ( i get data back). Any clue whats wrong with the above criteria ??
I will answer the last part in your update.
'01-Dec-10'
is a string. It is not a date. It has nothing to do with dates. Not until it gets converted into a date. There are (to my knowledge) three ways of doing that.
- Use TO_DATE()
- Specify a date literal as
date '2010-12-01'
, which is always iso date YYYY-MM-DD - Rely on implicit date conversion using the NLS settings (which you are doing)
If your NLS settings is 'DD-MON-RR'
, the string '01-Dec-10' would get converted into 2010-12-01 (the 1st of december, 2010). If, on the other hand, your settings are 'RR-MON-DD'
, it would be converted into '2001-12-10'
(the 10:th of december, 2001).
I don't know your settings, and since you don't either, you should do the right thing, and that is to always explicitly use TO_DATE() or date literals.
Use ANSI date literals rather than just strings for the date values:
select * from [schema].[table]
where resource_unique_name in ('abc123')
and slice_date >= date '2010-11-01'
and slice_date < date '2010-12-01'
精彩评论