From mboxrd@z Thu Jan 1 00:00:00 1970 From: Peter =?utf-8?q?M=C3=BCller?= To: development@lists.ipfire.org Subject: Re: [PATCH] attempt to make make.sh more shellcheck compliant Date: Fri, 12 Apr 2019 17:06:00 +0000 Message-ID: <60a1f77a-d07b-be5a-be7c-9d5a050d2353@ipfire.org> In-Reply-To: <4AE772C7-BAA2-4A28-8D4F-359DDC081A12@ipfire.org> MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="===============4877001321937515447==" List-Id: --===============4877001321937515447== Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Hello Michael, > Hello, >=20 > NACK. >=20 > I get your intention with this, but I do not think that the result is usabl= e. >=20 > It is a simple as that nobody ends a line with a ; in shell. I do not want = to edit the script that way in the future. This is not how people write shell. Since there are some lines ending with a semicolon in the script, I guess it makes more sense to adjust these now, so we stay consistent in general. >=20 > There might be some unquoted variables in this, but I really cannot find th= em in this patch. Okay, I will hand in another patch for them, so they are clearly visible. >=20 > I think there is a much bigger value in manually auditing code and understa= nd what it does and if it does it in the right way. A lexical analysis does n= ot improve quality. The script consists of almost 1950 lines. You are certainly right, but an automated analysis seemed to be the only way we can improve code quality a bit - as a full analysis requires too much time. >=20 > What you are doing here is making this compatible with other shells which w= ill never be called because the first line says #!/bin/bash. At the end of th= e day, this script does not even need to be portable because it won=E2=80=99t= run on any other platforms apart from Linux. Okay. >=20 > -Michael Thanks, and best regards, Peter M=C3=BCller --=20 The road to Hades is easy to travel. -- Bion of Borysthenes --===============4877001321937515447==--