Thanks Jeff and Roy

>>How about using EXPORT TABLE to get two CSV files and use BeyondCompare on
>>them?

This is actually not a bad idea at all!

>>Difficulties with blobby fields.

It would be pretty easy to generate BLOB free tables as a starting point. I don't use BLOBs much, and perhaps I could do a SIZE on each BLOB so differences would still be tracked.

... before I look at doing it in code I will look at using Beyond Compare.

Thanks Jeff and Roy

I am feeling a bit stupid now (not an unusual situation).

I have always used Beyond Compare on .pas files.

Reading the manual (I saw the existence of "data compare" options.

These are bloody brilliant, and do everything I had hoped to achieve through my own programming.

1. Generate 2 datasets you want to compare.
2. Use EDB Export statement to create comma-CSV.
3. Open them in BC.

... you get a cell-by-cell comparison of both datasets, highlighting any differences at row or cell level.

I would encourage anyone using EDB to experiment with the tool. There is a free version and the pro version is $30.

I think we are reaching a point where all the code you could ever imagine writing has already been written (better) by someone else. Does that make me feel sad?

Adam

>I would encourage anyone using EDB to experiment with the tool. There is a free version and the pro version is $30.

I use ExamDiff, not as good but does what I want. Just had a look at their site and the standard edition is $30, pro is $50

>I think we are reaching a point where all the code you could ever imagine writing has already been written (better) by someone else. Does that make me feel sad?
>
Only when they want you to pay for it

Looking at the Beyond Compare website it looks as though you can produce a file with just the differences which should mean you'd end up with a csv file you could then import to sync the tables.

Neat, and as long as you don't have to do it to often quite manageable

Roy Lambert.

>>Looking at the Beyond Compare website it looks as though you can produce a file with just the differences which >>should mean you'd end up with a csv file you could then import to sync the tables.

really good point Roy, I could definitely automate re-importation of corrected rows using this technique.

>>Neat, and as long as you don't have to do it to often quite manageable

It is really a "disaster recovery" situation, where a user has messed up their database & I want to be able to cross check values with a good version. Once a year or less, hopefully.

Adam,

There is also KDiff3 from Sourceforge which is open source (free).
http://kdiff3.sourceforge.net/

I just downloaded it but haven't had time to test it yet. There is ample online documentation and has 5 stars.

Barry

On 4/12/2014 2:05 AM, Adam Brett wrote:
> I would encourage anyone using EDB to experiment with the tool. There is a free version and the pro version is $30.

I highly recommend BC as we;; - owned it for what feels like forever.
It's updated regularly but major versions are rare (v3 was released
around 2008 and only now are they doing v4) so cheap to own also.
Not to mention it's written in delphi (might be fpc/lazarus now for
linux/mac support).

Raul