Re: parsing empty alternatives: case foo|) :;;

[Peter Stephenson <p.w.stephenson@ntlworld.com>]

On Mon, 7 Aug 2017 19:21:03 +0100
Peter Stephenson <p.w.stephenson@ntlworld.com> wrote:
> It's straightforward to fix, but note I can't fix "foo||bar)" without
> more work because || is a different token.  I think we can probably
> avoid parsing it that way here if we want to go down that route.

Looks like this isn't too difficult, either.

pws

diff --git a/Src/lex.c b/Src/lex.c
index b2d9b3f..8493d47 100644
--- a/Src/lex.c
+++ b/Src/lex.c
@@ -760,7 +760,7 @@ gettok(void)
 	return AMPER;
     case LX1_BAR:
 	d = hgetc();
-	if (d == '|')
+	if (d == '|' && !incasepat)
 	    return DBAR;
 	else if (d == '&')
 	    return BARAMP;
@@ -1058,7 +1058,7 @@ gettokstr(int c, int sub)
 	    if (isset(SHGLOB)) {
 		if (sub || in_brace_param)
 		    break;
-		if (incasepat && !lexbuf.len)
+		if (incasepat > 0 && !lexbuf.len)
 		    return INPAR;
 		if (!isset(KSHGLOB) && lexbuf.len)
 		    goto brk;
@@ -1859,7 +1859,7 @@ exalias(void)
     Reswd rw;
 
     hwend();
-    if (interact && isset(SHINSTDIN) && !strin && !incasepat &&
+    if (interact && isset(SHINSTDIN) && !strin && incasepat <= 0 &&
 	tok == STRING && !nocorrect && !(inbufflags & INP_ALIAS) &&
 	(isset(CORRECTALL) || (isset(CORRECT) && incmdpos)))
 	spckword(&tokstr, 1, incmdpos, 1);
diff --git a/Src/parse.c b/Src/parse.c
index ba9cd61..2705252 100644
--- a/Src/parse.c
+++ b/Src/parse.c
@@ -48,7 +48,11 @@ mod_export int incond;
 /**/
 mod_export int inredir;
  
-/* != 0 if we are about to read a case pattern */
+/*
+ * 1 if we are about to read a case pattern
+ * -1 if we are not quite sure
+ * 0 otherwise
+ */
  
 /**/
 int incasepat;
@@ -1194,6 +1198,7 @@ par_case(int *cmplx)
 
     for (;;) {
 	char *str;
+	int skip_zshlex;
 
 	while (tok == SEPER)
 	    zshlex();
@@ -1201,11 +1206,17 @@ par_case(int *cmplx)
 	    break;
 	if (tok == INPAR)
 	    zshlex();
-	if (tok != STRING)
-	    YYERRORV(oecused);
-	if (!strcmp(tokstr, "esac"))
-	    break;
-	str = dupstring(tokstr);
+	if (tok == BAR) {
+	    str = dupstring("");
+	    skip_zshlex = 1;
+	} else {
+	    if (tok != STRING)
+		YYERRORV(oecused);
+	    if (!strcmp(tokstr, "esac"))
+		break;
+	    str = dupstring(tokstr);
+	    skip_zshlex = 0;
+	}
 	type = WC_CASE_OR;
 	pp = ecadd(0);
 	palts = ecadd(0);
@@ -1243,10 +1254,11 @@ par_case(int *cmplx)
 	 * this doesn't affect our ability to match a | or ) as
 	 * these are valid on command lines.
 	 */
-	incasepat = 0;
+	incasepat = -1;
 	incmdpos = 1;
-	for (;;) {
+	if (!skip_zshlex)
 	    zshlex();
+	for (;;) {
 	    if (tok == OUTPAR) {
 		ecstr(str);
 		ecadd(ecnpats++);
@@ -1302,10 +1314,26 @@ par_case(int *cmplx)
 	    }
 
 	    zshlex();
-	    if (tok != STRING)
+	    switch (tok) {
+	    case STRING:
+		/* Normal case */
+		str = dupstring(tokstr);
+		zshlex();
+		break;
+
+	    case OUTPAR:
+	    case BAR:
+		/* Empty string */
+		str = dupstring("");
+		break;
+
+	    default:
+		/* Oops. */
 		YYERRORV(oecused);
-	    str = dupstring(tokstr);
+		break;
+	    }
 	}
+	incasepat = 0;
 	par_save_list(cmplx);
 	if (tok == SEMIAMP)
 	    type = WC_CASE_AND;
diff --git a/Test/A01grammar.ztst b/Test/A01grammar.ztst
index 9625a15..0302c96 100644
--- a/Test/A01grammar.ztst
+++ b/Test/A01grammar.ztst
@@ -820,6 +820,43 @@
 0:case keeps exit status of last command executed in compound-list
 >37
 
+  case '' in
+    burble) print No.
+    ;;
+    spurble|) print Yes!
+    ;;
+    |burble) print Not quite.
+    ;;
+  esac
+  case '' in
+    burble) print No.
+    ;;
+    |burble) print Wow!
+    ;;
+    spurble|) print Sorry.
+    ;;
+  esac
+  case '' in
+    gurgle) print No.
+    ;;
+    wurgle||jurgle) print Yikes!
+    ;;
+    durgle|) print Hmm.
+    ;;
+    |zurgle) print Hah.
+    ;;
+  esac
+  case '' in
+    # Useless doubled empty string to check special case.
+    ||jurgle) print Ok.
+    ;;
+  esac
+0: case with no opening parentheses and empty string 
+>Yes!
+>Wow!
+>Yikes!
+>Ok.
+
   x=1
   x=2 | echo $x
   echo $x