I am reworking and expanding a somewhat complex database schema that has a small number of tables and queries but they are closely related. The only problem I had with it was that in one of the tables the 2 fields that were relating to another table were using the field name of the record and not the ID of the record.
I changed the referring fields data type from text to number and entered some data. The queries and the reports work fine with 1 exception:
There is one report that uses both referring fields. One of the fields is ok but the other one shows symbols instead of numbers. ( The IDs in my sample entries were 14 and 20 and the symbols shown were a double barred music note /alt code 14/ and the symbol for an end of a paragraph /alt code 20/ ) Investigating further I have found that if I make a query that contains the query source for the report both fields display fine, but if I add another table to that query the second field once again shows symbols instead of numbers.
I have found a workaround of this by converting those fields back to text and the id fields in the other tables to text as well. This text key will probably haunt me later on, so I’d like to make it right before it is too late.
This is all access 2010 btw. The source file was already in 2010 (couldn’t open in 2007 even)
Sounds like a corruption issue for sure. I would try adding a new column and run an update query to populate it with the values from the old column (maybe use cint(indexfield)), then delete the old column.
It might also be a good idea to decompile the database. This often helps resolve corruption issues.