asked on

Microsoft Query Functions

I have a pivot table based on an Access file of trucks weighing in and out of a facility. The data includes the time the truck weighs in, and then (after it is loaded or unloaded) the time it weighs out. I want to determine the elapsed time using MS Query and then return the data to Excel.

Here is a little bit of the data:

GTime      TTime      ElapsedTime
07:42      07:14      28
14:33      15:11      39
19:52      20:33      41
17:25      18:29      64
09:09      09:53      45
12:20      11:53      28

Does MS Query have built in date/time formats that I can use to create a calculated column like this:

(GTime - TTime) = Elapsed

or something like

timevalue(GTime) - timevalue(TTime) = Elapsed

I got it to work when I did this:

abs(left(GTime,2)*60+right(Gtime,2) - left(TTime,2)*60+right(TTime,2))

I am looking for information about time/date functions that are existing in MS Query. I have spent hours on Google trying to find The Definitive Guide to MS Query, but there does not appear to be any good reference material on it. Anyone know where such a book exists?

(One of you experts would make a ton of dough by writing a book like that.)

TommySzalapski

MS Query is based mostly on SQL and there are so many books out on that that I certainly couldn't write one worth anything.

The query should support datediff and n means minutes so abs(datediff("n", GTime, TTime)) should get you what you want. More datediff stuff here http://www.techonthenet.com/access/functions/date/datediff.php

If that doesn't do it for you, read on.
If your times are stored as strings, then there really isn't a much better way to do it than you have. If they are stored as numbers then they are stored in days so abs(GTime - TTime)*24*60 will give you the minutes.

GRayL

There does not appear to be any correlation between the start and end times and your two GTime and TTime fields. I would have expected one to be always greater than the other. Do you ever have a situation where a before time is just before midnight and the loaded time is some time after midnight?

ASKER CERTIFIED SOLUTION

GRayL

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

jkasavan

ASKER

In the Access data source file, the Gtime and Ttime fields are text.

So this works:

abs((left(gtime,2)*60+right(gtime,2))-(left(ttime,2)*60+right(ttime,2)))

and so does this

abs(1440*(timevalue(gtime)-timevalue(ttime)))

TommySzalapski

The cleanest solution would then probably be
abs(datediff("n", timevalue(GTime), timevalue(TTime)))

jkasavan

ASKER

Cdate(timevalue(gtime)-timevalue(ttime)) gives:

1899-12-30 00:29:00
1899-12-30 00:37:00
1899-12-30 00:33:00
1899-12-30 00:34:00
1899-12-30 00:44:00

abs(datediff("n", GTime, TTime)) gives:

"Too few parameters, Expected 1."

jkasavan

ASKER

abs(datediff("n", timevalue(GTime), timevalue(TTime)))

also gives

"Too few parameters, Expected 1."

TommySzalapski

Must be abs(datediff(minute, GTime, TTime))
or abs(datediff(minute, timevalue(GTime), timevalue(TTime)))

GRayL

I'm running A2003 and do not get the 1899 date. However you can get rid of it by wrapping with timeValue()

tb = time()
ta = time()+.1
? ta
14:54:12
? tb
12:29:28
?ta-tb
0.100196759259259
? timevalue(cdate(ta-tb))
02:24:44

jkasavan

ASKER

I appreciate the assistance. Thanks very much.

jkasavan

ASKER

Tommy - these

Must be abs(datediff(minute, GTime, TTime))
or abs(datediff(minute, timevalue(GTime), timevalue(TTime)))

both still yield

"Too few parameters. Expected 1."

GRayL

Thanks, but why the B? I explained away the 1899-2-30.