1299, !DUPE is taking CASE into consideration

[therube]

1299


!DUPE is taking CASE into consideration
- should it?

naomi <> Naomi

(in one respect it's good that it's not,
cause it tells me i need to change case of the other...)


Thinking about this more, I actually like it, "cause it tells me i need to change case of the other..." ("matching" filename).

[raccoon]

I noted dupe taking case into consideration too, at least for Regex-Match-0, but I thought I'd investigate more before reporting. The odd thing to me was the inverse (unique) would also match some of the same results, but not others. I hadn't pegged the exact behaviors. Where fileA has a different CaSe from fileB.

[void]

Yes, the case search modifier can be applied to dupe:, distinct: and unique:

Normally dupe:, distinct: and unique: will be case insensitive.



For example, to match case with dupe:
case:dupe:



dupe:

[raccoon]

When I right-click on the Name column header, and I right-click on Find Name Duplicates and I left click on the last item Find Unique (not duplicated), it does not operate CaSe InSeNsItIvE under any condition Only filters out exact case matching names. This is different from using the command versions. They are all inconsistent except for DUPE.

I will compare them below. I have Match Case disabled.

Everything64_case-files.png (27.48 KiB) Viewed 3045 times

dupe: and !unique: and unique: and !dupe: commands versus menu selections:

Everything64_dupe-shots.png (91.74 KiB) Viewed 3045 times

Here are the files in my test.

case.zip

(2.21 KiB) Downloaded 118 times

[raccoon]

Request: Consider changing the Statusbar indicator DISTINCT to UNIQUE to match the words that are used in the Find Duplicates submenu, and for word familiarity in the English lexicon.

[void]

Thank you for the bug report raccoon,

The following are matching case from the UI:
Find Duplicates (Except First One)
Find Unique (Including First Duplicate)
Find Unique (Not Duplicated)


They should be ignoring case.
I will fix this for the next alpha update.


The "Unique" definition in Everything will change in the next alpha update.
Currently "unique" and "distinct" are the same. (I originally borrowed this from SQL, but it doesn't apply to Everything)

"Unique" in the next alpha update will mean Not Duplicated.
"Distinct" in Everything means Unique (Including First Duplicate).

Thanks for the suggestion.

[raccoon]

The "Unique" definition in Everything will change in the next alpha update.
Currently "unique" and "distinct" are the same. (I originally borrowed this from SQL, but it doesn't apply to Everything)

"Unique" in the next alpha update will mean Not Duplicated.
"Distinct" in Everything means Unique (Including First Duplicate).

Hmm, I'm still a bit confused. Especially hurts when you explain "Distinct" in Everything means Unique.

New UNIQUE will mean Old !DUPE ?
New DISTINCT will remain the same?

So the menu will go...
from DUPE -> !DISTINCT -> DISTINCT -> !DUPE
to DUPE -> !DISTINCT -> DISTINCT -> UNIQUE

Or is there a special behavior that will distinguish UNIQUE from !DUPE?

[void]

ha, it's tricky to explain without using the word unique or duplicated..

New UNIQUE will mean Old !DUPE
New DISTINCT will remain the same.


DUPE = Find Duplicates (Including First One)
!DISTINCT = Find Duplicates (Except First One)
DISTINCT = Find Unique (Including First Duplicate)
UNIQUE = Find Unique (Not Duplicated)

[NotNull]

Lets see if I understand:

UNIQUE = There is only 1 of this kind; show it
DUPE = There are more than 1 of this kind. Show them all
DISTINCT = There are 1 or more of this kind. Show only 1 result

!DUPE = There is only 1 of this kind; show it (same as UNIQUE)
!DISTINCT = There are 1 or more of this kind. Show them all, minus 1
!UNIQUE = same as DUPE

?

[void]

Yes, this is a good explanation.

!DISTINCT is the odd one out here as it's not very useful.

[raccoon]

!DISTINCT is the odd one out here as it's not very useful.

I expect if someone is careful and conscientious about their sort ordering, use of !DISTINCT will permit for easy Select-All -> Delete.