Oh man, does this bring back bad memories. I had this exact problem on 
my 256 field with umpteen formatted memo fields table.

In the end it turned out that there was some invisible control 
characters or something in one memo field of a few records. To fix the 
problem I deleted everything in the record's memo field and retyped (not 
pasted).

You could copy the table and then blank the memo field for every record 
to see what happens.


On 08/20/2019 08:46 AM, Steven Green wrote:
> so this was a big wabbit hole, still trying to figure it out completely
> 
> first off, indexes
> 
> INDEX MAINTAINED, single field, still makes an X0* file.. multi-field 
> and/or CASE INSENSITIVE, it makes an XG* file.. I guess we're all 
> supposed to still know that :-)
> 
> second, changing the index type didn't fix this problem.. all three 
> reproducible GP errors point to values past the end of the index values 
> of the linked table, which is what made me notice the old indexes, but 
> the bottom line on that linked table is still "invalid formatted memo 
> header".. make that go away, everything else is fine
> 
> so today's experiment is changing the two F fields to M fields.. we'll 
> see.. but in another related scenario in that app, losing the F will 
> lose some critical BOLD paragraph headers.. so we're still trying to ID 
> whatever it is in the F fields that allows the data in the field but 
> causes the table failure
> 
> -- 
> 
> Steven Green
> Myrtle Beach, South Carolina, USA
> 
> http://www.OasisTradingPost.com
> 
> Collectibles and Memorabilia
> Vintage Lego Sets and Parts
> - and Paradox support, too
> "Steven Green"  wrote in message 
> news:5d4ad422$1@pnews.thedbcommunity.com...
> 
> wow.. haven't seen that expression in a long time
> 
> tableframe in a form.. get the stack overflow when using Record / GoTo /
> Last (Ctrl-F12).. don't get it interactively on the raw table..
> 
> same table I reported here the other day, with the unexpected filter
> behavior.. table has been rebuilt and packed regularly, always.. I'm gonna
> do some testing, but I don't have a clue
> 
> -- 
> 
> Steven Green
> Myrtle Beach, South Carolina, USA
> 
> http://www.OasisTradingPost.com
> 
> Collectibles and Memorabilia
> Vintage Lego Sets and Parts
> - and Paradox support, too