00:00:37 <esolangs> [[Language list]] https://esolangs.org/w/index.php?diff=177825&oldid=177804 * Cleverxia * (+18) /* T */
00:06:41 <b_jonas> Sgeo: the J manual calls strings (character arrays, possibly multidimensional, as opposed to arrays of numbers or boxed) literal too, see https://www.jsoftware.com/docs/help701/dictionary/dx003.htm
00:22:02 <ais523> Sgeo: heh, I'd actually never seen it in action before (and assumed that it might have been fixed at some point, but apparently not?)
00:35:00 <Sgeo> ais523, this compiler is Fortran G, which is for FORTRAN IV. So... not very new
00:40:44 <Sgeo> I should probably delete those equal signs. I copy/pasted from MUSIC/SP's fullscreen editor
00:45:13 <Sgeo> Same result on VS FORTRAN II, which I think is 77
00:58:51 <Sgeo> FORTRAN IV subroutines can take... statement numbers as arguments. The subroutine can say which one to return control to
01:02:21 <Sgeo> "If a dummy argument is assigned a value in the subprogram, the corresponding actual argument must be a subscripted or unsubscripted variable
01:02:21 <Sgeo> name, or an array name. A constant should not be specified as an actual
01:02:21 <Sgeo> argument unless the programmer is certain that the corresponding dummy
01:02:21 <Sgeo> argument is not assigned a value in the subprogram.
01:02:51 <Sgeo> ais523, is the behavior you stated in a spec somewhere or is it something akin to UB?
01:03:28 <ais523> Sgeo: I'm not sure of the history – I knew at least some versions of Fortran acted like that, but not why
01:11:13 <b_jonas> ais523: it makes sense, we're here fifty years later and we're still not sure how to consistently handle const correctness nicely without overhead like having to define multiple versions of subscripting methods. it's no surprise then that Fortran compilers back then would have considered enforcing const correctness a priority
01:13:46 <Sgeo> https://archive.org/details/ansi-x-3.9-1966-fortran-66/page/24/mode/1up I think this is saying the same thing, and in fact states that expressions are sent by value
01:14:02 <Sgeo> I should probably look at the 77 spec next
01:22:24 <esolangs> [[]] https://esolangs.org/w/index.php?diff=177826&oldid=177546 * InLuaIKnow * (+68)
01:25:51 <esolangs> [[]] https://esolangs.org/w/index.php?diff=177827&oldid=177826 * InLuaIKnow * (+27)
01:28:13 -!- amby has quit (Remote host closed the connection).
01:56:46 -!- ais523 has quit (Quit: quit).
03:00:24 <Sgeo> https://fortranwiki.org/fortran/show/Modernizing+Old+Fortran
03:03:24 -!- lisbeths has joined.
03:21:13 <Sgeo> FORTRAN II required function names to end with F
03:22:37 <Sgeo> Actually not sure if that only applies to functions defined in single statements.
03:22:57 <Sgeo> ...actually I'm even more confused now
03:27:20 <Sgeo> Statement functions end with F, subprogram functions with names 4 characters or longer cannot have their name end with F
03:31:54 <Sgeo> https://cyber.dabamos.de/programming/fortran/transcompiler/lecture01.pdf
03:38:15 <Sgeo> "This statement is kind of boring. A CONTINUE just returns to the top of a loop." pretty sure this is flatly incorrect, CONTINUE is a noop
06:03:15 -!- Sgeo has quit (Read error: Connection reset by peer).