I'm thinking I might need to use a SQL Prepared Statements

[SuperDude_123]

As a continuation to my last post, I'm finding that when I use the LIKE '*' command, I don't always get consistent results. From looking in to it, I have a feeling that it's where I am concatenating my results for WHERE, and one thought was that I might need to use some prepared statement ahead of time, and feed it in to my WHERE later. Looking online, I haven't had much luck with this, so I was wondering if there was a way to solve my problem, or if there's a way to modify my WHERE line to get me consistent results.

SELECT Database.Description, Database.[Part Number], Database.Manufacturer, Database.Symbol, Database.[Job Number], Database.File, Database.DirectoryFROM [Database] INNER JOIN (SELECT Database.[Job Number] FROM [Database] GROUP BY Database.[Job Number] HAVING (((Sum([Symbol]=[Enter First Symbol] Or [Symbol]=[Enter Second Symbol]))=-2)))  AS qry1 ON Database.[Job Number] = qry1.[Job Number]
WHERE (((Database.Symbol) LIKE '*' + [Enter First Symbol] + '*')) OR (((Database.Symbol) LIKE '*' + [Enter Second Symbol] + '*'))
ORDER BY Database.File;

[June7]

First, use & instead of + for string concatenation.

Second, show example data.

[Micron]

Maybe show results too - especially incorrect ones. I could be way off here, but this is how I solve this part of your expression, assuming the inputs are 3 and -5 respectively. Shouldn't matter what they are, I see the result of the equals operation (=) being either True or False in any case.

HAVING (((Sum([Symbol]=[Enter First Symbol] Or [Symbol]=[Enter Second Symbol]))=-2)))

HAVING (((Sum([Symbol]= 3 Or [Symbol]=-5))=-2)))

HAVING (((Sum(True Or False))=-2)))

HAVING Sum(-1 Or 0))=-2 *this seems unlikely, not to mention I don't think I'd use Sum with OR
Maybe I'm taking the whole thing out of context. Will have to revisit the posted link. Very nice idea to include that in your post SD.

[SuperDude_123]

So here's some example data and how things are generally structured:

Description	Part Number	Manufacturer	Symbol	Order Number	File	Directory
Temp Sensor	H1234	Honeywell	TS1	11001	order_details.xls	C:\orders\hill_billy_Bob_inc
Speed Drive	VFD112	Siemens	VFD1	11001	order_details.xls	C:\orders\hill_billy_Bob_inc
120V Relay	S1234	Siemens	RE1	11001	order_details.xls	C:\orders\hill_billy_Bob_inc
Temp Sensor	H1234	Honeywell	TS1	11002	New_Order.xls	C:\orders\Michael_Jackson_inc
Temp Sensor	H3344	Honeywell	TS2	11002	New_Order.xls	C:\orders\Michael_Jackson_inc
Speed Drive	VFD112	Siemens	VFD	11002	New_Order.xls	C:\orders\Michael_Jackson_inc
120V Relay	S1234	Siemens	RE	11002	New_Order.xls	C:\orders\Michael_Jackson_inc

The type of errors that my code throws is that if I search on the symbols "VFD" and "TS1", I should get the following results:

Temp Sensor	H1234	Honeywell	TS1	11001	order_details.xls	C:\orders\hill_billy_Bob_inc
Speed Drive	VFD112	Siemens	VFD1	11001	order_details.xls	C:\orders\hill_billy_Bob_inc
Temp Sensor	H1234	Honeywell	TS1	11002	New_Order.xls	C:\orders\Michael_Jackson_inc
Speed Drive	VFD112	Siemens	VFD	11002	New_Order.xls	C:\orders\Michael_Jackson_inc

But I will get something like this:

Temp Sensor	H1234	Honeywell	TS1	11001	order_details.xls	C:\orders\hill_billy_Bob_inc
Speed Drive	VFD112	Siemens	VFD1	11001	order_details.xls	C:\orders\hill_billy_Bob_inc
Temp Sensor	H1234	Honeywell	TS1	11002	New_Order.xls	C:\orders\Michael_Jackson_inc

I haven't ran all possible situations yet, but this is the type of error pattern that I am getting, where it will puke on some exact matches, but works on other exact matches.

I did try modifying the code from "+" to "&", and it didn't do anything different.

[isladogs]

If you searched for VFD1, you won't have got the record which has VFD

[SuperDude_123]

By using the LIKE '*', I've managed to get partial data, and that's what I am after. I'm not trying to get exact matches, but partial matches as well. So if I search for TS, I want to see all TS's (TS1, TS2, ....), and not just "TS".

[isladogs]

So did you search for *VFD* or *VFD1*?
If the latter, then as already stated you wouldn't get the VFD record which you said was missing in your output
If the former, you should indeed have got both records

I'm asking as you said you searched for TS1 rather than *TS*