On Wednesday, October 31, 2018 10:19 PM, Joseph Mayer <[email protected]> wrote:
> On Tuesday, October 30, 2018 4:33 PM, Sebastien Marie [email protected] wrote: > > > On Tue, Oct 30, 2018 at 08:11:24AM +0000, Joseph Mayer wrote: > > > > > On a quick sourcecode check I didn't see any code paths e.g. > > > "if (argv[0] matches "ksh") { something } else { something else }" > > > however I presume I missed something. > > > > yes, the check was too quick, it seems :) > > src/bin/ksh/main.c > > 134 int > > 135 main(int argc, char *argv[]) > > 136 { > > 137 int i; > > 138 int argi; > > 139 Source *s; > > 140 struct block *l; > > 141 int restricted, errexit; > > 142 char *wp; > > 143 struct env env; > > 144 pid_t ppid; > > 145 > > 146 kshname = argv[0]; > > 147 > > ... > > 236 / Check to see if we're /bin/sh. */237 if (!strcmp(kshname, "sh") || > > !strcmp(kshname, "-sh") || > > 238 (strlen(kshname) >= 3 && > > 239 !strcmp(&kshname[strlen(kshname) - 3], "/sh"))) { > > 240 Flag(FSH) = 1; > > 241 version_param = "SH_VERSION"; > > 242 } > > 243 > > > > Sebastien Marie > > Hi Sebastien, > > Thanks a lot for clarifying. > > sh's man page (http://man.openbsd.org/sh#DESCRIPTION) says: > > "This version of sh is actually ksh in disguise. As such, it also > supports the features described in ksh(1). This manual page describes > only the parts relevant to a POSIX compliant sh." > > When I read that originally, I perceived it as that "sh" and "ksh" > normally would have equivalent behavior - which also seems logical > given that their binaries are byte-equivalent. > > The meaning I gather from the sentence is that sh and ksh are > equivalent and that instead the man pages will describe different > functionalities that are actually available in both. > > Can that phrase in sh's man page be tweaked so that my misunderstanding > no longer is possible? > > Do you find my misunderstanding a reasonable reading? > > Joseph Moving this to bugs@ , sorry for spam.
