On Thu, 28 Aug 2025, 04:20 Chet Ramey, <chet.ra...@case.edu> wrote: > It was in early drafts of POSIX.2 (1003.2-1992). > > The feature started out as the compound command ((expr)) (draft 8), evolved > into a word expansion $[...] (draft 9), and then changed to $((...)) > (draft 11, the version I have was published in September, 1991). >
In hindsight, that was an unfortunate decision by the POSIX committee, creating a permanent mess parsing $((…)) as both command and numeric substitutions, depending on what's inside, like $((…)…(…)) or $(((case $thing in *$pattern) /bin/true & esac)). What was the rationale for preferring $((…)) over $[…] back when they had a free choice? POSIX has a warning tor effect that writing a command substitution starting with $(( is inadvisable, but not formally deprecated (in favour of $( (…)) with whitespace); would there be any appetite for issuing a formal deprecation? -Martin
