[RFE] UNIQ (aka Hide Dupelicates)

[therube]

[RFE] UNIQ (aka Hide Duplicates)

Just as 'Find ... Duplicates' is very helpful, its opposite, UNIQ, would be too.

you've got a list of files...
sorted by... doesn't matter, name
in the list you've got multiple drives (drive letters)
some of the drives are backups &/or dups of various others
so with that, you're apt to have dups...

you've got a list of files...
you don't necessarily want to exclude an entires drive (or directory tree)
- as the drives may not necessarily be mirrors, only that they /may/ happen
- to have dups

given that, say you're sorted by Name

c:/red train.mp3
c:/red train 2.mp3
z:/red train.mp3
z:/red train Z.mp3

- 'Find UNIQ names'

c:/red train.mp3
c:/red train 2.mp3
z:/red train Z.mp3

IOW, you're left with a set of uniq file names

after that, you might then do a 'Find Size Duplicates' based on Size
& that may point out that train2 & trainZ are the same size
which might make you investigate... given similar name & size
& you might find you've got an actual dup there, with just diff name...

or you could drag that set of uniq's into, GridPlayer
to visually look for dups...

or

sizedupe:
then, UNIQ name
might point out dupe: with differing name
- the UNIQ part would cut out, very quickly, a lot of "noise"

or

all kinds of situations where this would be very helpful.



(Earlier thread, No Dupes in Search Results.)


(I guess I also have to see what distinct: does.
Though right-click a column, like you do with Find...Duplicates, would be much easier.
And at the least, perhaps a synonym uniq: for distinct: [which is rather hard to say & spell & type ].

Also a right-click a column method would not "dirty up" History.)

[void]

The following search should do what you want:

distinct: sort:name


I will consider a column right click context menu item to find distinct / unique items.
Thanks for the suggestion.

[therube]

so...

i guess what i RFE'd for & what i (called it) are actually different
what i want - more so, is "distinct:"
with my original thought that "UNIQ" would leave "1 distinct" copy
of Property (thereby making it "unique"), but in order for it to
do that, it has to be "distinct"

& UNIQ, in a uniq sense, would be something not existing elsewhere,
heh.


& /that/ UNIQ would be good for finding files that are, say not backed up,
elsewhere, aka unique.


So... My H: drive "dropped out", yesterday. Poof, & it was gone.
immediate thought was, "do i have everything on it backed up, elsewhere"?
[or should that be, "do i have Everything on it backed up, elsewhere? ]
(rebooted, & it was back, thankfully. S.M.A.R.T. said OK, CHKDSK said OK,
& with that... <who cares about having backups > & my thoughts turned
elsewhere .)


so then came my need for "my UNIQ", aka, distinct: (which i didn't realize
what it did, nor was what i needed).

so on one hand, distinct: meets that need
(not as cleanly as 'Find...UNIQ', but it's workable...)

& on the other hand, there is the need for (the "other") UNIQ (1 = 1), to
appease my "do i have a backup of" concerns (after all, one must care
about backups ).

so, if you find distinct: sets of Properties, you are left with UNIQ,
but by the same token UNIQ <> distinct:, so in that respect, having
uniq: be a synonym wouldn't be politically correct.

BUT, if uniq: were to mean distinct: & uniq1: were to mean 1=1, then
you could have two unique's; 1=1 & 1=many

(& both easy to say & type & spell .) heh.


---


so i've got a search:
> cats blue
& i know i have dups in there (same files different drives)
so i then change the search:
> cats blue sort:name: distinct:
- awkward, hard to remember, hard to type... is it sort:name: (yes) or name:sort (no) <- it matters [& spellin too]

so i've got a search:
> cats blue
UNIQ #using UNIQ here, only cause its easy to remember & type - even if not politically correct ['Find Distinct Names']
now i've got:
> cats blue (but UNIQ)
then, i do a 'Find Size Duplicates'
& with that i find actual duplicates

by doing a UNIQ, first, that prunes the search results, removing dupes [Name],
such that i then don't have to (see or) run a hash check (or other operations) on duplicated files
- only on (size dup'd) UNIQ files (which is what i'm looking for, at this point)

& once i know just which are dups,
i can double-click the (status bar) UNIQ symbol to (once again) see my [dupe Size filtered] (cats blue) search
i can double-click the (status bar) DUPE symbol to (once again) see my [fully unfiltered] (cats blue) search
& all this happens without mudding up the search itself & the search History
(so search Bar & History show "cats blue" rather then "cats blue sort:name: distinct:")

[void]

Thanks for your post therube,

I have borrowed the distinct terminology from SQL.


I'm not sure what terminology to use here for NOT DUPLICATED.
unique might not be the best name as unique and distinct are mostly the same in SQL (UNIQUE != NOT DUPLICATED).
NOT DUPLICATED might be the best for now.



Everything currently has a NOT DUPLICATED option, except it is not exposed to the search or UI!
I've put on my TODO list to add a notduplicated: search function.
A having count(*) = 1 search (or equivalent, eg: havingcount:1) might also be useful.


For now, to find not duplicated results, please consider right clicking the result list column header and clicking Find xxx duplicates.

• Select all the files and press Ctrl + Shift + C
  (this will copy all the filenames to the clipboard from the current selection)
• Add the following to your search !filelist1:
• Hold down Ctrl and left click the !filelist1: text in the search box.
  (this will open the file list editor)
• Press Ctrl + V in the file list editor.
  (this will paste the filenames in the clipboard to the file list editor)
• Click OK.

I am reluctant to add a Find Uniques or Find NOT DUPLICATED context menu item as the column menu is getting too bloated.
I have on my TODO list to support menu customization, so maybe one day this will be an item you can add to your column context menu.



SQL Distinct
SQL query for non duplicate records

[therube]

I'm thinking a Filter should work decently...

Search: distinct: sort:name:
Sort: Size
Keyboard shortcut: Alt+U

(I'll have to see how that works out in practice...)


(You wouldn't believe how many times I've typed; name:sort: rather then sort:name: & then once I've gotten that straightened out, I come to find I forgot the colon on distinct: .

Also, a Filter does not "muddy up" the search string or history.
So both show as;

ext:pdf

rather then

ext:pdf sort:name: distinct:

.)


I'll note that distinct: sets the DUPE status bar notification, where maybe it should be deDUPE?
('Find ... Duplicates' also sets that same DUPE notification.)

If you use a Filter, the status bar shows the filter Name, & that filter can be cancelled by double-clicking the filter name.
If you use distinct:, that also sets the DUPE notification, & in this case, that (distinct:) cannot be cancelled by double-clicking DUPE.

So... if you
ext:pdf sort:name: distinct:
(or ext:pdf {Alt+U}
then, 'Find Size Duplicates'
that should point out different named files of the same size, i.e., potentially duplicates.

Code: Select all

C:\000\TMP\SEA\screamer\diff-pdf\old.pdf
C:\My Documents\IRS\_Forms\2012 MD State Employer Withholding_Guide-OLD.pdf

OK, neat.
distinct: is a very nice feature .

[raccoon]

What is RFE? Is it some companion program or release candidate? Been seeing a lot of these.

[therube]

Request For Enhancement
(In this case, the Enhancement was already there, I just didn't know it.)

[void]

Thank you for your feedback therube,

Everything 1.5.0.1286a make some changes to find duplicates:

duplicated: is now an alias for find-dupes:
unique: is now an alias for distinct:
added support for !duplicated: (find files and folders where the property value for the current sort is not duplicated)
added support for !distinct: (find duplicated files and folders excluding the first duplicate -not sure this one is useful)


DUPE is shown in the status bar when searching for duplicates.
DISTINCT is shown in the status bar when searching for distinct results.

DISTINCT in Everything means find unique files and folders including the first duplicate.


Note: you can use the case: search modifier with duplicate: functions to match case.

[Coldblackice]

This is a fantastic feature! Everything is truly living up to its name, making this utility technologically "lethal". Thanks for all the hard work, void!