OpenEuphoria

Re: ODBC

• Posted by "Carl W." <euphoria at cyreksoft.yorks.com> Sep 29, 2003
• 571 views

George Walters wrote:

> I'm setting up to try and use Matt's ODBC and connect to MySQL for a
> database. In reviewing what I need to do I've found a point of confusion
> in how I've been handeling dates and how MySQL handle's dates. I've been
> using a "sub sequence" in my record when using EUDB. as below.
>
> {a,b,c,{2003,09,29},e,f.....}
>
> this is quite convenient for dates to use the DateTime library (from
> user contributions) of functions I found in the archives. The problem is
> that MySQL uses yyy-mm-dd format. Is anyone doing this or has a
> suggestion on what happens when this is sent to MySQL. Hopefully there
 > is a solution or a better way to solve this issue.

Passing a Date as a sub-sequence to MySQL probably will likely pass the 
database three junk characters - not what you want.

If your Date is in variable 'd' then you'll want to do:
{a,b,c,sprintf("%04d-%02d-%02d",d),e,f.....}

Reading back into a date is a bit more tricky and you'll need to use 
Euphoria's get.e to parse the string back into a Date:

include get.e
sequence s, temp
Date d

s    = ??? -- get the stringified date from the db here
temp = value(s[1..4]) & value(s[6..7]) & value(s[9..10])
d    = temp[2] & temp[4] & temp[6]

There's no error checking in the above code. Look up value() and 
GET_SUCCESS / GET_FAIL in the Euphoria docs if you need to check that 
things are being parsed correctly. The status codes in the above snippet 
would be in temp[1], temp[3] and temp[5] respectively.

HTH,
Carl

-- 
[ Carl R White == aka () = The Domain of Cyrek = ]
[ Cyrek the Illogical /\ www.cyreksoft.yorks.com ]