ignore duplicates or just show one of n-duplicates?

General discussion related to "Everything".
Post Reply
randfee
Posts: 19
Joined: Fri Jul 27, 2018 5:38 pm

ignore duplicates or just show one of n-duplicates?

Post by randfee » Sun Sep 23, 2018 10:58 pm

Hi,

I am wondering if there is a way to ignore duplicate files (by file name and size, or selectable either one)?
Also it would be useful to just show one file for each number of dupes.
advanced search feature.
something like
  • !dupe: --> doesn't show any file that exists twice or more often by name
  • !sizedupe:--> doesn't show any file that exists twice or more often by size
  • dupedupe:one --> shows only one of the duplicates
  • sizedupe:one --> shows only one of the duplicates by size
is there any way to create a search result like that?

cheers

therube
Posts: 2680
Joined: Thu Sep 03, 2009 6:48 pm

Re: ignore duplicates or just show one of n-duplicates?

Post by therube » Mon Sep 24, 2018 12:10 pm

(Without looking too closely) the !'s look to be working already.

zeus04
Posts: 13
Joined: Wed Oct 17, 2018 3:45 pm

Re: ignore duplicates or just show one of n-duplicates?

Post by zeus04 » Wed Oct 17, 2018 3:56 pm

randfee wrote: [*]dupedupe:one --> shows only one of the duplicates
[*]sizedupe:one --> shows only one of the duplicates by size[/list]
is there any way to create a search result like that?

cheers
Hello!

This would be a tremendous improvement making Everything the best solution for visual duplication management. A whole genre of programs dealing with duplicate-management couldn't compete with Everything with some extra options like randfee suggested.

I suggest to also have options like;
sizedupe:oneNEWEST
sizedupe:one[path]

What do you think about that?

void
Site Admin
Posts: 5654
Joined: Fri Oct 16, 2009 11:31 pm

Re: ignore duplicates or just show one of n-duplicates?

Post by void » Thu Oct 18, 2018 6:33 am

With dupe:one and sizedupe:one how should Everything pick the single file to show? Would the first file found for the current sort be enough?

zeus04
Posts: 13
Joined: Wed Oct 17, 2018 3:45 pm

Re: ignore duplicates or just show one of n-duplicates?

Post by zeus04 » Sun Oct 21, 2018 11:08 pm

void wrote:With dupe:one and sizedupe:one how should Everything pick the single file to show? Would the first file found for the current sort be enough?
I was also a little curious about that, hence my suggestion for the option to choose by path or by time, like:

Code: Select all

"somepic.jpg sizedupe:1newest" or "somepic.jpg sizedupe:1newestmodified" 
"somepic.jpg sizedupe:1biggest"
"somepic.jpg sizedupe:1d:\pics"
"somepic.jpg sizedupe:1newest"
Am i missing something basic here? Haven't really thought thru what kind of database/index/filesystem -problems this might cause. Im aware that this is not a competent feature-spec, but the purest user-perspective on what would be nice.


----
I use Everything for ~30 TB of data and several million files, its one of the most impressive softwares i seen. Using it more and more over (other) databases for finding documents and files. When nothing else plays along, Everything saves the day again and again!

void
Site Admin
Posts: 5654
Joined: Fri Oct 16, 2009 11:31 pm

Re: ignore duplicates or just show one of n-duplicates?

Post by void » Mon Oct 22, 2018 7:31 am

"somepic.jpg sizedupe:1newest" or "somepic.jpg sizedupe:1newestmodified"
specifying the limits in the search could work as that would allow you to keep the results sorted by size.
I've added this to my TODO list.

Thanks for the feedback.

DerekZiemba
Posts: 2
Joined: Thu Sep 27, 2018 4:46 pm

Re: ignore duplicates or just show one of n-duplicates?

Post by DerekZiemba » Thu Sep 17, 2020 7:32 pm

Any update on this?

I'd like to have what's outlined below. I suspect some of this would take quite a bit of work so I broke it down into stages where releases could occur - or you decide it's not worth going further. I'd be happier with this software than I already am with just Stage 1
  1. Stage 1. Basic greater than, less than, and range support
    • N or more: `dupe:2+`
    • N or less: `dupe:2-`
    • N to M: `dupe:2...4` (ellipses instead of '2-4' as I suspect it'd be easier to implement. When it sees '4' it won't need to backtrack and revise the original interpretation of 2 or less)
  2. Stage 2. (unless stage 3 is implemented). Option to specify sorting thats different from the current default (alphabetical descending)
    • By Date Created: `dupe:2+oldest` and `dupe:2+newest`
    • By Date Modified: `dupe:2+oldestmod` and `dupe:2+newestmod`
    • By Size: `dupe:2+largest` and `dupe:2+smallest`
  3. Stage 3. Group by Name Column ascending or descending (depending on name column header clicked state)
    • Clicking sort by one of the other columns only sorts within each name group.
    • Dupe/grouping column dynamically inserted into view if `dupe:` query.
    • Groups are collapsable/expandable by clicking the arrow in the "Dupe" or grouping column.
    • Clicking sort by one of the other columns only sorts within each name group (makes Stage 2 unnecessary)
    • When a group is collpased:
      • Size column show the sum of all dupes
      • Path column of collapsed group shows common parent, in the mockup that would be `A:\applied\rater\...`
      • Dates could maybe show the range from oldest to newest?
        ie: "3/4/2020 - 9/19/2020" (and time if column width is wider than some minimum size)

    Mockup below. I suspect this would take a lot of work and probably won't be implemented. But a man can dream...
    MSPaint-Voidtools-Mockup.png
    MSPaint-Voidtools-Mockup.png (185.96 KiB) Viewed 53 times

Post Reply