Delete Duplicate can delete ALL instances!

[johnmeyer]

I have used Ace Utilities for years. I use many of the features, but especially the delete duplicate function.

However, there are several flaws that really should be corrected.

The most important flaw is that if you use the "Check all folders" or the similar options which also selects sub-folders, AND if you do this on more than one folder, then Ace Utilities will -- depending on how your duplicates are distributed -- sometimes check ALL of the duplicates in a particular group. This is a major flaw because when deleting thousands of duplicates, as I am now doing right now, there is no way to ensure that you will not completely delete a particular file unless you use the "check all folders" function only once. This means that I have to delete and then re-scan after each and every use of of this feature.

This is not good.

The obvious solution -- although the developers may have a better solution -- is to provide an alert asking you if you want to completely delete all instances of particular files. If you say "NO, I don't want to completely remove all instances," Ace should then give you a new window showing all groups in which all duplicates have been seelcted. You can then de-select those files which shouldn't be deleted.

Alternatively, you could provide an alert prior to the selection(s) going into effect, telling you that the action you just took will result in one or more files being completely deleted from your disk. This implies an ability to undo the last operation, something that would be useful for other reasons as well.

The key point is this: Ace should never delete all files in a group without first alerting the user.

Another feature request is to provide a running total of both the number and the size of all duplicates currently selected. This is not a big deal, but would be helpful.

The next issue has to do with groups of duplicates which are scattered across many folders deep in a directory tree, for instance:

e:\files\tree1\folder1
e:\files\tree1\folder2
e:\files\tree1\folder3

e:\files\tree2\folder1
e:\files\tree2\folder2
e:\files\tree2\folder3

Suppose, just to make it simple, that tree1 and tree2 are actually identical. Further, suppose that instead of just three folders all at just one level below the "tree" level, there are dozens, and they are at all sorts of additional levels below "tree." The "check mark all within this file's folder and it's subfolders" really isn't very useful because you have to find an instance of a duplicate in each and every subfolder before you can get them all. This is really not practical. To get around this, I often intentionally put a duplicate file way up at the "tree1" and "tree2" level, and make that file huge so that it becomes the first file in the duplicate list. I then have to execute only one "check mark all ..." command and I'm done. It also avoids the problem I described above.

Thus, my feature request would be to have a way to specify a starting point for the check marking that is further up the directory tree from the actual tree level shown in the current duplicate group. For instance, in the simple example above, I'd like to right click on e:\files\tree1\folder2, but be able to specify that I'd like to check mark all files in the e:\files\tree1 and below. That would then get folder 1, folder 2, folder 3, etc.

Thank you for a great product!

[XtremeMaC]

i dunno if it applies to you but u can first choose "check one from each" then say "invert selection"
i've been using this method. it ensures that all except 1 is selected for deletion.

[Jabberwockybreath]

How does one access the "check one from each" option for selection?

[XtremeMaC]

when you see the results right click anywhere in the result window and you should be seeing