Backport 'printf -v' from ksh 93v-

'printf' on bash and zsh has a popular -v option that allows
assigning formatted output directly to variables without using a
command substitution. This is much faster and avoids snags with
stripping final linefeeds. AT&T had replicated this feature in the
abandoned 93v- beta version. This backports it with a few tweaks
and one user-visible improvement.

The 93v- version prohibited specifying a variable name with an
array subscript, such as printf -v var\[3\] foo. This works fine on
bash and zsh, so I see no reason why this should not work on ksh,
as nv_putval() deals with array subscripts just fine.

src/cmd/ksh93/bltins/print.c: b_print():
- While processing the -v option when called as printf, get a
  pointer to the variable, creating it if necessary. Pass only the
  NV_VARNAME flag to enforce a valid variable name, and not (as
  93v- does) the NV_NOARRAY flag to prohibit array subscripts.
- If a variable was given, set the output file to an internal
  string buffer and jump straight to processing the format.
- After processing the format, assign the contents to the string
  buffer to the variable.

src/cmd/ksh93/data/builtins.c:
- Document the new option, adding a warning that unquoted square
  brackets may trigger pathname expansion.

src/cmd/ksh93/bltins/print.c

@@ -1,7 +1,7 @@
 /***********************************************************************
 *                                                                      *
 *               This software is part of the ast package               *
-*          Copyright (c) 1982-2012 AT&T Intellectual Property          *
+*          Copyright (c) 1982-2014 AT&T Intellectual Property          *
 *          Copyright (c) 2020-2021 Contributors to ksh 93u+m           *
 *                      and is licensed under the                       *
 *                 Eclipse Public License, Version 1.0                  *
@@ -176,6 +176,7 @@ int    b_print(int argc, char *argv[], Shbltin_t *context)
 	const char *options, *msg = e_file+4;
 	char *format = 0;
 	int sflag = 0, nflag=0, rflag=0, vflag=0;
+	Namval_t *vname=0;
 	Optdisc_t disc;
 	exitval = 0;
 	disc.version = OPT_VERSION;
@@ -236,11 +237,21 @@ int    b_print(int argc, char *argv[], Shbltin_t *context)
 			else if(!sh_iovalidfd(shp,fd))
 				fd = -1;
 			else if(!(shp->inuse_bits&(1<<fd)) && (sh_inuse(shp,fd) || (shp->gd->hist_ptr && fd==sffileno(shp->gd->hist_ptr->histfp))))
-
 				fd = -1;
 			break;
 		case 'v':
-			vflag='v';
+			if(argc < 0)
+			{
+				/* prepare variable for printf -v varname */
+				vname = nv_open(opt_info.arg, shp->var_tree, NV_VARNAME);
+				if(!vname)
+				{
+					errormsg(SH_DICT, ERROR_exit(2), e_create, opt_info.arg);
+					UNREACHABLE();
+				}
+			}
+			else
+				vflag='v';
 			break;
 		case 'C':
 			vflag='C';
@@ -288,6 +299,13 @@ skip:
 	/* handle special case of '-' operand for print */
 	if(argc>0 && *argv && strcmp(*argv,"-")==0 && strcmp(argv[-1],"--"))
 		argv++;
+	if(vname)
+	{
+		if(!shp->strbuf2)
+			shp->strbuf2 = sfstropen();
+		outfile = shp->strbuf2;
+		goto printf_v;
+	}
 skip2:
 	if(fd < 0)
 	{
@@ -314,6 +332,7 @@ skip2:
 	}
 	/* turn off share to guarantee atomic writes for printf */
 	n = sfset(outfile,SF_SHARE|SF_PUBLIC,0);
+printf_v:
 	if(format)
 	{
 		/* printf style print */
@@ -363,7 +382,9 @@ skip2:
 			if(sfputc(outfile,'\n') < 0)
 				exitval = 1;
 	}
-	if(sflag)
+	if(vname)
+		nv_putval(vname, sfstruse(outfile), 0);
+	else if(sflag)
 	{
 		hist_flush(shp->gd->hist_ptr);
 		sh_offstate(SH_HISTORY);

src/cmd/ksh93/data/builtins.c

@@ -1193,7 +1193,7 @@ const char sh_optprint[] =
 ;
 
 const char sh_optprintf[] =
-"[-1c?\n@(#)$Id: printf (ksh 93u+m) 2021-09-13 $\n]"
+"[-1c?\n@(#)$Id: printf (ksh 93u+m) 2021-11-18 $\n]"
 "[--catalog?" SH_DICT "]"
 "[+NAME?printf - write formatted output]"
 "[+DESCRIPTION?\bprintf\b writes each \astring\a operand to "
@@ -1358,6 +1358,9 @@ const char sh_optprintf[] =
 	"time conversions will be treated as if \bnow\b were supplied.]"
 "[+?\bprintf\b is equivalent to \bprint -f\b which allows additional "
 	"options to be specified.]"
+"[v]:[name?Put the output in the variable \aname\a instead of writing to "
+	"standard output. \aname\a may include an array subscript (note that "
+	"the square brackets should be quoted to avoid pathname expansion).]"
 "\n"
 "\nformat [string ...]\n"
 "\n"

src/cmd/ksh93/include/version.h

@@ -21,7 +21,7 @@
 
 #define SH_RELEASE_FORK	"93u+m"		/* only change if you develop a new ksh93 fork */
 #define SH_RELEASE_SVER	"1.0.0-beta.2"	/* semantic version number: https://semver.org */
-#define SH_RELEASE_DATE	"2021-11-16"	/* must be in this format for $((.sh.version)) */
+#define SH_RELEASE_DATE	"2021-11-18"	/* must be in this format for $((.sh.version)) */
 #define SH_RELEASE_CPYR	"(c) 2020-2021 Contributors to ksh " SH_RELEASE_FORK
 
 /* Scripts sometimes field-split ${.sh.version}, so don't change amount of whitespace. */

src/cmd/ksh93/tests/builtins.sh

@@ -1298,5 +1298,18 @@ got="$($SHELL -i "$hist_error_leak" 2>&1)"
 [[ $exp == "$got" ]] || err_exit "file descriptor leak after substitution error in hist builtin" \
 	"(expected $(printf %q "$exp"), got $(printf %q "$got"))"
 
+# ======
+# printf -v works as of 2021-11-18
+integer ver=.sh.version
+exp=ok$'\f'0000$ver$'\n'
+printf -v got 'ok\f%012d\n' $ver 2>/dev/null
+[[ $got == "$exp" ]] || err_exit "printf -v not working" \
+	"(expected $(printf %q "$exp"), got $(printf %q "$got"))"
+unset got
+printf -v 'got[1][two][3]' 'ok\f%012d\n' $ver 2>/dev/null
+[[ ${got[1]["two"][3]} == "$exp" ]] || err_exit "printf -v not working with array subscripts" \
+	"(expected $(printf %q "$exp"), got $(printf %q "$got"))"
+unset got ver
+
 # ======
 exit $((Errors<125?Errors:125))