tab separated export / import

[curious99]

Hi all,

I just finished reviewing many posts on that subject and I still need input from more experienced users than I am.

I have a large database ( 200 mb, 6000 fields, 600 layouts, hundreed of scripts, 17 years in the making) that I do not want to rewrite. I know there is corruption in there somewhere but I have decided I could live with it ( back in the 1989 - 1993 period I was working with a Mac that had a ROM problem that took years to identify and caused many crashes).

I am wondering if I could, without using the recover function which I understand after reading all these posts on recovery is a last resort solution, improve the integrity of my data.

One problem I have is that when I want to import from tab separated, the import screen "data in" does not match every field in the "fields in" side for all the 6000 or so fields. If I change from "last order" to "creation order", I am not able to correctly match every fields and so on for the other choices available ( field names, field types) I am not trying "custom order" because, as you probably understand, 6000 custom order is too many.

What would you suggest I do? Am I missing on something basic that could make this process of importing easier?

[BobWeaver]

Years ago, before anyone warned me not to use the recover command, I would recover files whenever the slightest thing appeared amiss. The usual symptom was erroneous results returned from a Find operation, indicating a corrupted index. Never ever had a problem with recover. Nowadays, I avoid it if possible, but if you are sure there is something wrong with your file it may not be a bad idea to recover it. If it makes things worse, you can always go back to the original one.

Here's another idea. Save a clone of your file. Then import the records directly from the recovered file into the clone. That should rebuild the index and may fix a number of problems.

[curious99]

Thanks Bob,

I agree, I never had much problems with recover back then. I'll try what you suggest. Got any idea though about the import not matching?

Thanks for your input, it is very much appreaciated

:-)

Luc

[BobWeaver]

Last import order will not necessarily match the last export order, and if you do it by creation order, you may have problems with calculated fields and container fields shifting things out of order. But, if you import from a recovered FM file rather than a tab delimited file, then you should be able to do it by matching field names.

[curious99]

Thanks Bob.

Since Filemaker inc. does suggest on their support site to import from tab separated in case of a real bad corruption, I wish they would have an easy way to match fields and data when the situation occurs. I understand that we should backup every night (which I do since implementing FM server 3 years ago) but with the millions of users out there, I am certain that this is a must when you have a database with hundreds or thousands of fields. Hope they are aware of this situation. If not, is there an efficient way to let them know? Is FM 7 better in that regard?

Thanks again for your input, it has helped me a lot as I thought I was really out of track with this tab separated import feature. Data integrity being the base of databases, this has stressed me for years.

:-)

Luc