Importing last week of November Impact files

[Carl]

When I downloaded weekly Impact files, I got two less lines of data than when I imported HTR data.

First time this has ever happened to me, Impact total and HTR total have always matched before.

I even did it twice, still got same result. Error message was the one to the effect of "two lines could not be imported due to mismatch of type" thing.

Anybody else have this problem? Is there an easy way to see what two lines I do have in HTR that I do not have in Impact via a query so that maybe I could "patch" them in Notepad before they came in?

[Victor]

Carl,
I think I had the same problem. My database is current through 11/23. I have not tried adding the last week of November yet, but when I imported for daily plays on 11/25, I got a similar error message. My assumption was that there was a problem with the 11/25 data files. I'll wait for possible solutions before I update my database.
Victor

[Rick]

Since someone was nice enough to narrow down the day, I was able to narrow down the races where the problem is.

It is HOL, 25 Nov 01, Races 5 and 7.

If I can figure out what the problem is I will post it.

Maybe someone else can figure it out.

[Rick]

The problem is that there was two twelves listed in each race.

After checking the charts there was a 12 and 12F in each race.

I think HTR only looks at the first two characters.

[MikeDee]

If you have your key fields set you will get this error becasue the key field will not allow duplicates. To fix it got to your table and remove the key so that there are no key fields and then import the data.

Once it is imported go into the table and fix the duplicate to 12f and then reset your key fields

[Carl]

Rick,

Since I haven't seen this 12/12A before in 18 months of importing data, is it something that HDW has changed in their download?

Do you think it will be a future problem?

[Rick]

This is the first time I have noticed it.

I have the primary key in ALL_MSA using post position as the fourth part. In ALL_IMPC it uses the program # as the fourth part.

It looks like this could occur in any race that has a field bet. And these will normally be the highest class of race.

If it happens so rarely we are better off to just delete the races instead of going to some of the work arounds like adding horses name to primary key. Even that is not perfect. You would occasionally run into horses in the same race that have the first part of their names the same.

[MikeDee]

I think it is extremly rare where there is a entry with 2 digit program numbers that result in a 3 character prg no. Since 99% of the entires are entires of either no1 or no2. I think just deleting the race or fixing the offending race manualy in access is the way to go.

[Carl]

We've had large fields before, and somehow Impact didn't get "mixed up".

Something must be different. If I weren't so lazy, I would find it.

Agree otherwise, if it is a one time thing, easier just to delete the offending races.

[MikeDee]

It's not the large fields but the entry on a No. 12 post position. This makes the prg no 3 characters. Normally the prg no is only 2 characters even with an entry like 1A or 2B. A 12F prg no is vary rare. Any program looking for a 2 character prg no is going to screw this up and you end up with duplicate no 12's not a 12 and a 12F

[Carl]

I don't think "field" horses are all that rare, but maybe they are.

Most tracks nowdays go to at least 14 betting interests.

I assume this was coupled as a betting interest only as a "12 and 12F"? Otherwise I would be full amazed, in forty years of betting I have always seen the first entry as the 1/1A, then the 2/2B etc.....