DSXchange

A full outer run sequentially sounds slow but would work. If these are large files, you should be able to add an extra column by renaming the column in file2 to value2 (for example) so that when merged with a regular inner join you get (id, desc, value, value2) in your output, preserving both values...

Any chance this is the classic floating point problem where the display shows you 45 but you have in one system 45.00000000000000001 and in the other system 44.999999999999999999999999 or some similar hidden microscopic difference in values? Convert the values to raw hex and compare those in a peek ...

It locates it. As far as IGC is concerned the type is the old type. Something makes it update -- but I do not know what. I started with about 35 jobs that needed a change. On Friday morning my query in IGC found all 35, on Monday morning it found 15 (but all 15 were fixed already from Friday) and to...

I had a number of jobs that needed a parameter changed from one type to another and nothing else. I am trying to use IGC queries to ensure I got them all but IGC is not updating so most of what it is showing me are jobs I already fixed. It has been over the weekend since I fixed some of these jobs. ...

But if your company or yourself pay & attend their class on qualitystage you will get access to a virtual machine with it installed that you can use for a few days and can see it in action as you do the class problems & play with it beyond that a bit on your own. That is probably the closest...