Steffen Nurpmeso wrote in <20240816212216.accse4FG@steffen%sdaoden.eu>: |Hello kre@. | |Robert Elz wrote in | <16443.1723841...@jacaranda.noi.kre.to>: || Date: Thu, 15 Aug 2024 23:33:42 +0200 || From: Steffen Nurpmeso <stef...@sdaoden.eu> || Message-ID: <20240815213342.t6-hdjZT@steffen%sdaoden.eu> || ||| I have extended the test a bit, and i also see word split ||| differences. || ||There are so many problems with this test, that I'm not sure ||it is worth bothering looking to see what happens, or why there ||are differences. | |Might very well be it is true what you say for IFS, i should place |a semicolon after its' assignment. |This does not change that there is a state machine problem with |the (d)ash (derived) codebase, however. |How is it going for NetBSD, i have not looked? (Nor FreeBSD.)
Ok i see when i *do* place a semicolon the problem vanishs. Luckily i do not maintain a shell where i now would have to double the test, because for my MUA it only works with the semicolon per se. (And yes, you were right with that eval, but this comes from my MUA, and that is not the sh(1)ell language: it reads input lines, and scans for a command, and when it knows which command it is, *then* it knows how to handle the rest of the line content. Ie, if the command uses sh(1)ell style argument processing, then the arguments are processed accordingly. And, note, dear kre@, that "eval" is a command modifier, which performs eval-times sh(1)ell-style processing of the command line *before* all that happens. (Ie, like this, you can place several commands on one line and separate them with sh(1)ell-style semicolon, even if one of those commands does not actually use shell-style argument processing.) --steffen | |Der Kragenbaer, The moon bear, |der holt sich munter he cheerfully and one by one |einen nach dem anderen runter wa.ks himself off |(By Robert Gernhardt)
