A very strange issue occurred this week that I've never seen before. I have a table with a primary key ID. I have another table with the same ID in a QVD file, and I wanted to bring some of the fields from the QVD into my table. So I did a simple left join from the QVD onto the data table, and it created hundreds of thousands of duplicate records. I thought a left join didn't create any rows, only appended columns?
In one case, it created 11M records and caused our publisher to crash. I can't for the life of me understand why a left join from a QVD would cause QV to blow up like this. I've been developing this way for years, and this is a very simple query. Does anyone have any ideas?
I bet the records are not perfect duplicates. Don't they differ in at least one field?
- If perfect duplicates, could the duplicates already exist in your table B?
- Obviously check you actually do join with key fields, which in QV (stating the obvious) should have identical field names (no typos, UpPer/Lower case differences etc)
- If table B (qvd) holds multiple values per key field joining, it will naturally create more rows in table A than originally.
Hope above helps. If not, please state info about your two tables
