PATCH: ksh $'...' syntax

[Zoltan Hidvegi <hzoli@xxxxxxxxxx>]

This implements the $'...' quoting syntax from ksh93.
Here is the ksh doc:

                                         A  single  quote  cannot
       appear  within  the single quotes.  A single quoted string
       preceded by an unquoted $ is processed as an ANSI-C string
       except  that  \0 within the string causes the remainder of
       the string to be ignored  and  \E  is  equivalent  to  the
       escape  character  (ascii 033).                           

Zsh is a little bit different, it interprets the escapes the same way the
print builtin does, and of course \0 within the string is allowed and
does not cause the reminder ignored.

This syntax is very useful to specify command line arguments with
non-printable characters or newlines.

Zoli


*** Doc/Zsh/grammar.yo	1997/06/05 04:44:57	3.1.3.0
--- Doc/Zsh/grammar.yo	1997/06/06 06:13:24	3.1.3.1
***************
*** 305,312 ****
  A character may be var(quoted) (that is, made
  to stand for itself) by preceding it with a `tt(\)'.
  `tt(\)' followed by a newline is ignored.
! All characters enclosed between a pair of single quotes (tt(''))
! are quoted. A single quote cannot appear within single quotes.
! Inside double quotes (tt("")), parameter and command substitution
! occurs, and `tt(\)' quotes the characters
  `tt(\)', `tt(`)', `tt(")', and `tt($)'.
--- 305,314 ----
  A character may be var(quoted) (that is, made
  to stand for itself) by preceding it with a `tt(\)'.
  `tt(\)' followed by a newline is ignored.
! All characters enclosed between a pair of single quotes (tt('')) that
! is not preceeded by a `tt($)' are quoted.  A single quote cannot appear
! within single quotes.  A single quoted string preceded by an unquoted
! `tt($)' is processed the same way as the string arguments of the
! tt(print) builtin.  Inside double quotes (tt("")), parameter and
! command substitution occurs, and `tt(\)' quotes the characters
  `tt(\)', `tt(`)', `tt(")', and `tt($)'.
*** Src/lex.c	1997/06/05 04:44:57	3.1.3.0
--- Src/lex.c	1997/06/06 06:13:24	3.1.3.1
*************** gettokstr(int c, int sub)
*** 936,948 ****
  	    if (lexstop)
  		goto brk;
  	    break;
! 	case LX2_QUOTE:
  	    add(Snull);
  	    cmdpush(CS_QUOTE);
  	    for (;;) {
  		STOPHIST
  		while ((c = hgetc()) != '\'' && !lexstop) {
! 		    if (!sub && isset(CSHJUNKIEQUOTES) && c == '\n') {
  			if (bptr[-1] == '\\')
  			    bptr--, len--;
  			else
--- 936,955 ----
  	    if (lexstop)
  		goto brk;
  	    break;
! 	case LX2_QUOTE: {
! 	    int strquote = (len && bptr[-1] == String);
! 
  	    add(Snull);
  	    cmdpush(CS_QUOTE);
  	    for (;;) {
  		STOPHIST
  		while ((c = hgetc()) != '\'' && !lexstop) {
! 		    if (strquote && c == '\\') {
! 			add(c);
! 			c = hgetc();
! 			if (lexstop)
! 			    break;
! 		    } else if (!sub && isset(CSHJUNKIEQUOTES) && c == '\n') {
  			if (bptr[-1] == '\\')
  			    bptr--, len--;
  			else
*************** gettokstr(int c, int sub)
*** 967,972 ****
--- 974,980 ----
  	    lexstop = 0;
  	    c = Snull;
  	    break;
+ 	}
  	case LX2_DQUOTE:
  	    add(Dnull);
  	    cmdpush(CS_DQUOTE);
*************** skipcomm(void)
*** 1347,1359 ****
  	    add(c);
  	    c = hgetc();
  	    break;
! 	case '\'':
  	    add(c);
  	    STOPHIST
! 	    while ((c = hgetc()) != '\'' && !lexstop)
  		add(c);
  	    ALLOWHIST
  	    break;
  	case '\"':
  	    add(c);
  	    while ((c = hgetc()) != '\"' && !lexstop)
--- 1355,1374 ----
  	    add(c);
  	    c = hgetc();
  	    break;
! 	case '\'': {
! 	    int strquote = bptr[-1] == '$';
  	    add(c);
  	    STOPHIST
! 	    while ((c = hgetc()) != '\'' && !lexstop) {
! 		if (c == '\\' && strquote) {
! 		    add(c);
! 		    c = hgetc();
! 		}
  		add(c);
+ 	    }
  	    ALLOWHIST
  	    break;
+ 	}
  	case '\"':
  	    add(c);
  	    while ((c = hgetc()) != '\"' && !lexstop)
*** Src/subst.c	1997/06/05 04:44:57	3.1.3.0
--- Src/subst.c	1997/06/06 06:13:24	3.1.3.1
*************** stringsubst(LinkList list, LinkNode node
*** 114,119 ****
--- 114,122 ----
  		str = arithsubst(str + 2, &str3, str2);
  		setdata(node, (void *) str3);
  		continue;
+ 	    } else if (str[1] == Snull) {
+ 		str = getkeystring(str, NULL, 4, NULL);
+ 		continue;
  	    } else {
  		node = paramsubst(list, node, &str, qt, ssub);
  		if (errflag || !node)
*** Src/utils.c	1997/06/05 04:44:57	3.1.3.0
--- Src/utils.c	1997/06/06 06:13:24	3.1.3.1
*************** dquotedzputs(char const *s, FILE *stream
*** 3142,3153 ****
  char *
  getkeystring(char *s, int *len, int fromwhere, int *misc)
  {
!     char *buf = halloc(strlen(s) + 1);
!     char *t = buf, *u = NULL;
      char svchar = '\0';
      int meta = 0, control = 0;
  
!     for (; *s; s++) {
  	if (*s == '\\' && s[1]) {
  	    switch (*++s) {
  	    case 'a':
--- 3142,3159 ----
  char *
  getkeystring(char *s, int *len, int fromwhere, int *misc)
  {
!     char *buf;
!     char *t, *u = NULL;
      char svchar = '\0';
      int meta = 0, control = 0;
  
!     if (fromwhere != 4)
! 	buf = halloc(strlen(s) + 1);
!     else {
! 	buf = s;
! 	s += 2;
!     }
!     for (t = buf; *s; s++) {
  	if (*s == '\\' && s[1]) {
  	    switch (*++s) {
  	    case 'a':
*************** getkeystring(char *s, int *len, int from
*** 3235,3240 ****
--- 3241,3249 ----
  		}
  		break;
  	    }
+ 	} else if (fromwhere == 4 && *s == Snull) {
+ 	    for (u = t; (*u++ = *s++););
+ 	    return t + 1;
  	} else if (*s == '^' && fromwhere == 2) {
  	    control = 1;
  	    continue;
*************** getkeystring(char *s, int *len, int from
*** 3257,3263 ****
--- 3266,3278 ----
  	    t[-1] |= 0x80;
  	    meta = 0;
  	}
+ 	if (fromwhere == 4 && imeta(t[-1])) {
+ 	    *t = t[-1] ^ 32;
+ 	    t[-1] = Meta;
+ 	    t++;
+ 	}
      }
+     DPUTS(fromwhere == 4, "BUG: unterminated $' substitution");
      *t = '\0';
      *len = t - buf;
      return buf;