Slice array with negative indexes's get the unlogic result's

[LutAle]

Prerequisites

• Write a descriptive title.
• Make sure you are able to repro it on the latest released version
• Search the existing issues.
• Refer to the FAQ.
• Refer to Differences between Windows PowerShell 5.1 and PowerShell.

Steps to reproduce

I'm build grab docker cli string output with slicing waste header and processing remain string as need.
I is assumed that index -1 point last element in array, that is convenience.
But with reverse index i get items in reverse order, with garbage string header the it need drop.
Assumed that Array[0..Count] and slice operator [1..-1] produce sequence item's in direct odrder's [1], [2], [3], ... [Count] , but it in real case has output that slice with reverce item order [1], ... [0], [Count]
This operation rotate item sequence in buckward dirrection

Expected behavior

PS:\user>docker ps
CONTAINER ID   IMAGE              COMMAND     CREATED        STATUS        PORTS     NAMES
230d25385033   python:3.12-slim   "bash"      32 hours ago   Up 32 hours             pedantic_herschel
3f4b28c315cd   neovim             "sh"        37 hours ago   Up 37 hours             xenodochial_murdock
8b88e3fa3e4a   alpine             "/bin/sh"   47 hours ago   Up 47 hours             zen_meninsky

PS:\user>(docker ps)[1..-1]
230d25385033   python:3.12-slim   "bash"      32 hours ago   Up 32 hours             pedantic_herschel
3f4b28c315cd   neovim             "sh"        37 hours ago   Up 37 hours             xenodochial_murdock
8b88e3fa3e4a   alpine             "/bin/sh"   47 hours ago   Up 47 hours             zen_meninsky

For buil command like: (docker ps)[1..-1]  |  % {docker stop $_.Split(' ')[0]}
I get:
PS:\user>(docker ps)[1..-1] | % {$_.Split(' ')[0]}
230d25385033
CONTAINER
8b88e3fa3e4a

Actual behavior

PS:\user>(docker ps)[1..-1]
230d25385033   python:3.12-slim   "bash"      33 hours ago   Up 33 hours             pedantic_herschel
CONTAINER ID   IMAGE              COMMAND     CREATED        STATUS        PORTS     NAMES
8b88e3fa3e4a   alpine             "/bin/sh"   2 days ago     Up 2 days               zen_meninsky

Error details

No response

Environment data

Name                           Value
----                           -----
PSVersion                      7.4.0-preview.6
PSEdition                      Core
GitCommitId                    7.4.0-preview.6
OS                             Microsoft Windows 10.0.19044
Platform                       Win32NT
PSCompatibleVersions           {1.0, 2.0, 3.0, 4.0…}
PSRemotingProtocolVersion      2.3
SerializationVersion           1.1.0.1
WSManStackVersion              3.0

Visuals

[mklement0]

While it is understandable to expect 1..-1 inside [...] to refer to the range of elements from the second (1) through to the last (-1), this is - by design - not how it works.

.. is the range operator, which exists independently of index expressions ([...]), and, if used in the latter, has no knowledge of the array / collection being addressed: 1..-1 invariably results in the following array of indices: 1, 0, -1; e.g.:

PS> (1..10)[1..-1]  # Gets the second (1), the first (0), and the last (-1) element.
2
1
10

See the following proposal for introducing a syntax that does what you're looking for:

• Support array (collection) slicing with all-remaining and except-last-N elements logic - enhance the range operator in index operations #7940

[237dmitry]

I always wondered why this was the case, and it has always been like this. Then I got used to it, the main thing is to know this behavior.

Workaround:

($a = 1..10)[1..$a.Count]

[mklement0]

Indeed, @237dmitry, though in the case at hand it would be: ($a = 1..10)[1..($a.Count-2)] in order to get 2, 3, ... 9

[237dmitry]

though in the case at hand it would be

Yes, for [1..-2]

For [1..-1] the error does not occur if you specify the last index that is obviously larger than the actual one.

(1..10)[1..100]   # more iterations but the same result
(1..10)[1..1e6]   # very slow

[mklement0]

Good point, I confused the -2 and -1 use case.
It may not often arise in practice, but note that with Set-StrictMode -Version 3 or higher in effect,
($a = 1..10)[1..$a.Count] breaks. That is, specifying indices outside the array's bounds then causes an error.

[237dmitry]

but note that with Set-StrictMode -Version 3

I did not know it. Now I will.

[microsoft-github-policy-service]

This issue has been marked as by-design and has not had any activity for 1 day. It has been closed for housekeeping purposes.

[microsoft-github-policy-service]

This issue has been marked as by-design and has not had any activity for 1 day. It has been closed for housekeeping purposes.

[microsoft-github-policy-service]

This issue has been marked as by-design and has not had any activity for 1 day. It has been closed for housekeeping purposes.

[microsoft-github-policy-service]

This issue has been marked as by-design and has not had any activity for 1 day. It has been closed for housekeeping purposes.

[microsoft-github-policy-service]

This issue has been marked as by-design and has not had any activity for 1 day. It has been closed for housekeeping purposes.