Path: csiph.com!goblin1!goblin.stu.neva.ru!usenet.stanford.edu!not-for-mail
From: Chet Ramey <chet.ramey@case.edu>
Newsgroups: gnu.bash.bug
Subject: Re: Bash parameter transforamtion on empty array triggers unset variable.
Date: Wed, 12 Aug 2020 09:49:05 -0400
Organization: ITS, Case Western Reserve University
Lines: 57
Approved: bug-bash@gnu.org
Message-ID: <mailman.1714.1597240155.2739.bug-bash@gnu.org>
References: <CAJjXSE9b8G0aXyc+t58cbYRwFbiN5eX2rpKnyFwg728x5ZQhKQ@mail.gmail.com> <3ddb8a56-c7d0-b87e-8bb4-520acecc90d5@case.edu> <CAJjXSE8vfCpx7r4owo5V0tA=G1qi5nr5oDwqJwyQoM41oC2W3Q@mail.gmail.com> <6afdb56d-5dff-e780-e794-8c053fb166ee@case.edu>
Reply-To: chet.ramey@case.edu
NNTP-Posting-Host: lists.gnu.org
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 8bit
Cc: chet.ramey@case.edu, chet@cwru.edu, bash-testers@cwru.edu, bug-bash@gnu.org
To: Andrew Neff <andrew.neff@visionsystemsinc.com>
Envelope-to: bug-bash@gnu.org
Autocrypt: addr=chet.ramey@case.edu; prefer-encrypt=mutual; keydata= mQGiBEEOsGwRBACFa0A1oa71HSZLWxAx0svXzhOZNQZOzqHmSuGOG92jIpQpr8DpvgRh40Yp AwdcXb8QG1J5yGAKeevNE1zCFaA725vGSdHUyypHouV0xoWwukYO6qlyyX+2BZU+okBUqoWQ koWxiYaCSfzB2Ln7pmdys1fJhcgBKf3VjWCjd2XJTwCgoFJOwyBFJdugjfwjSoRSwDOIMf0D /iQKqlWhIO1LGpMrGX0il0/x4zj0NAcSwAk7LaPZbN4UPjn5pqGEHBlf1+xDDQCkAoZ/VqES GZragl4VqJfxBr29Ag0UDvNbUbXoxQsARdero1M8GiAIRc50hj7HXFoERwenbNDJL86GPLAQ OTGOCa4W2o29nFfFjQrsrrYHzVtyA/9oyKvTeEMJ7NA3VJdWcmn7gOu0FxEmSNhSoV1T4vP2 1Wf7f5niCCRKQLNyUy0wEApQi4tSysdz+AbgAc0b/bHYVzIf2uO2lIEZQNNt+3g2bmXgloWm W5fsm/di50Gm1l1Na63d3RZ00SeFQos6WEwLUHEB0yp6KXluXLLIZitEJLQwQ2hldCBSYW1l eSAoQ2FzZSBzdGFuZGFyZCkgPGNoZXQucmFtZXlAY2FzZS5lZHU+iF8EExECAB8FAkPi19EC GwMHCwkIBwMCAQMVAgMDFgIBAh4BAheAAAoJELtYafBk6nSrelkAn31Gsuib7GcCZHbv5L5t VKYR9LklAJ4hzUHKA49Z0QXR+qCb80osIcmPSbkBDQRBDrBvEAQAkK6TAOKBEM+EC4j6V/7o /riVZqcgU5cid2qG9TXdwNtD9a3kvA/ObZBO93sX59wc6Bnwo4VJxsOmMlpGrAjJsxNwg3QH akEtf8LXRbVpj5xStdmBdQZUhIQyalo/2/TZq5OijtddUQcL5cs70hTv/FpT3wUvr2Xr8rjF 41IFEz8AAwcD/A0CZEGlzIrT5WCBnl6xBog/8vKiUCbarByat3d1mL6DbizvKNXQRTC9E/vE dENAWCQCjr75Bu55xT8n3SXGtWdDC5xmZ/P3OBYORP8yl8H8I1FIosWOFirbIeYdZPq8SPD1 HL+EXo9zSiHVrrZRJ19ooCKKbSdXHFCY+aJG+0KZiEkEGBECAAkFAkEOsG8CGwwACgkQu1hp 8GTqdKvjcACfZlkVCDwaz/NTO9cy3t69oWpVPNwAnRwe0qk/WL/gfhH346xh5B3HFbFN
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:68.0) Gecko/20100101 Thunderbird/68.11.0
In-Reply-To: <CAJjXSE8vfCpx7r4owo5V0tA=G1qi5nr5oDwqJwyQoM41oC2W3Q@mail.gmail.com>
Content-Language: en-US
Received-SPF: pass client-ip=129.22.103.194; envelope-from=chet.ramey@case.edu; helo=mpv3-2015.case.edu
X-Spam_score_int: -43
X-Spam_score: -4.4
X-Spam_bar: ----
X-Spam_report: (-4.4 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001 autolearn=ham autolearn_force=no
X-Spam_action: no action
X-BeenThere: bug-bash@gnu.org
X-Mailman-Version: 2.1.23
Precedence: list
List-Id: Bug reports for the GNU Bourne Again SHell <bug-bash.gnu.org>
List-Unsubscribe: <https://lists.gnu.org/mailman/options/bug-bash>, <mailto:bug-bash-request@gnu.org?subject=unsubscribe>
List-Archive: <https://lists.gnu.org/archive/html/bug-bash>
List-Post: <mailto:bug-bash@gnu.org>
List-Help: <mailto:bug-bash-request@gnu.org?subject=help>
List-Subscribe: <https://lists.gnu.org/mailman/listinfo/bug-bash>, <mailto:bug-bash-request@gnu.org?subject=subscribe>
X-Mailman-Original-Message-ID: <6afdb56d-5dff-e780-e794-8c053fb166ee@case.edu>
X-Mailman-Original-References: <CAJjXSE9b8G0aXyc+t58cbYRwFbiN5eX2rpKnyFwg728x5ZQhKQ@mail.gmail.com> <3ddb8a56-c7d0-b87e-8bb4-520acecc90d5@case.edu> <CAJjXSE8vfCpx7r4owo5V0tA=G1qi5nr5oDwqJwyQoM41oC2W3Q@mail.gmail.com>
Xref: csiph.com gnu.bash.bug:16743

On 8/11/20 11:45 AM, Andrew Neff wrote:
> Ah, I see the confusion.
> 
> The issue you pointed out, "@Q breaks set -o nounset" refers to quote
> parameter expansion, as does the line in CHANGES-5.1, 1.q, which makes
> sense to call this a bug that was allowed in bash 4.4 and 5.0.

Not quite. The report revealed that none of the parameter transformations
obeyed `set -u'. The fix was to make that happen, so that parameter
transformations were aligned with the other word expansions.


> I should have specified, the focus of this issue is the "@a" expansion. 

Sure, you want an exception.

> It
> makes sense that @Q/E/P/A expansion should not work on unset variables with
> nounset enabled. However, @a is uniquely different, in that it does not
> have to do with the value of the variable, but rather the variable type.

This is why it works on unset variables when `set -u' isn't enabled. The
question is whether that makes it special enough to be an exception.


> Here are 3 specific details I would like to address:
> 
> 1. @a expansion should work on unset variables with "set -u" in bash 5.1.
> It seems like the correct thing to do. Only @a expansion. This has been a
> very useful feature in bash 4.4 and 5.0.

OK, you found it useful in something you wanted to do. You're restating the
proposition.

> 
> 2. With "set -u", the following works in bash 4.4 or newer (and makes sense
> that it works): (set -eu; x=(); echo "${x[@]}")
> Here x is not unset, it is set to an empty array. This expansion make sense
> with nounset turned on because x is not unset, it is set to ()

That's not why. The variable is indeed unset. The reason it doesn't fail
is because there's a special POSIX carveout for $@ with no positional
parameters and I extended it to ${var[@]}. The expansions of $@ and ${A[@]}
parallel each other in just about every other way, so it made sense.

> 3. The same as #2, but for associative arrays
> Works: (set -eu; declare -A x=(); echo "${x[@]}")

The same carveout for ${A[@]}.

So I guess the question is whether or not @A and @a are exceptional enough.

-- 
``The lyf so short, the craft so long to lerne.'' - Chaucer
		 ``Ars longa, vita brevis'' - Hippocrates
Chet Ramey, UTech, CWRU    chet@case.edu    http://tiswww.cwru.edu/~chet/