More TODO.

[bison.git] / src / scan-gram.l

diff --git a/src/scan-gram.l b/src/scan-gram.l
index 7ea0d6855d448e985cc3cd679b5456484b8098f6..5ef8edba25c30cd857acee5538f34e9e25d172f8 100644 (file)
--- a/src/scan-gram.l
+++ b/src/scan-gram.l
@@ -78,6 +78,8 @@ static void unexpected_newline (boundary, char const *);
  /* A identifier was just read in directives/rules.  Special state
     to capture the sequence `identifier :'. */
 %x SC_AFTER_IDENTIFIER
  /* A identifier was just read in directives/rules.  Special state
     to capture the sequence `identifier :'. */
 %x SC_AFTER_IDENTIFIER

+ /* A complex tag, with nested angles brackets. */
+%x SC_TAG

 
  /* Three types of user code:
     - prologue (code between `%{' `%}' in the first section, before %%);
 
  /* Three types of user code:
     - prologue (code between `%{' `%}' in the first section, before %%);


@@ -96,8 +98,10 @@ int    [0-9]+
 
 /* POSIX says that a tag must be both an id and a C union member, but
    historically almost any character is allowed in a tag.  We disallow
 
 /* POSIX says that a tag must be both an id and a C union member, but
    historically almost any character is allowed in a tag.  We disallow

-   NUL and newline, as this simplifies our implementation.  */
-tag     [^\0\n>]+
+   NUL, as this simplifies our implementation.  We disallow angle
+   bracket to match them in nested pairs: several languages use them
+   for generics/template types.  */
+tag     [^\0<>]+

 
 /* Zero or more instances of backslash-newline.  Following GCC, allow
    white space between the backslash and the newline.  */
 
 /* Zero or more instances of backslash-newline.  Following GCC, allow
    white space between the backslash and the newline.  */


@@ -105,8 +109,9 @@ splice       (\\[ \f\t\v]*\n)*
 
 %%
 %{
 
 %%
 %{

-  /* Nesting level of the current code in braces.  */
-  int braces_level IF_LINT (= 0);
+  /* Nesting level.  Either for nested braces, or nested angle brackets
+     (but not mixed).  */
+  int nesting IF_LINT (= 0);

 
   /* Parent context state, when applicable.  */
   int context_state IF_LINT (= 0);
 
   /* Parent context state, when applicable.  */
   int context_state IF_LINT (= 0);


@@ -205,8 +210,6 @@ splice       (\\[ \f\t\v]*\n)*
   "="                     return EQUAL;
   "|"                     return PIPE;
   ";"                     return SEMICOLON;
   "="                     return EQUAL;
   "|"                     return PIPE;
   ";"                     return SEMICOLON;

-  "<*>"                   return TYPE_TAG_ANY;
-  "<>"                    return TYPE_TAG_NONE;

 
   {id} {
     val->uniqstr = uniqstr_new (yytext);
 
   {id} {
     val->uniqstr = uniqstr_new (yytext);


@@ -235,18 +238,25 @@ splice     (\\[ \f\t\v]*\n)*
   /* Code in between braces.  */
   "{" {
     STRING_GROW;
   /* Code in between braces.  */
   "{" {
     STRING_GROW;

-    braces_level = 0;
+    nesting = 0;

     code_start = loc->start;
     BEGIN SC_BRACED_CODE;
   }
 
   /* A type. */
     code_start = loc->start;
     BEGIN SC_BRACED_CODE;
   }
 
   /* A type. */

+  "<*>"       return TAG_ANY;
+  "<>"        return TAG_NONE;

   "<"{tag}">" {
     obstack_grow (&obstack_for_string, yytext + 1, yyleng - 2);
     STRING_FINISH;
     val->uniqstr = uniqstr_new (last_string);
     STRING_FREE;
   "<"{tag}">" {
     obstack_grow (&obstack_for_string, yytext + 1, yyleng - 2);
     STRING_FINISH;
     val->uniqstr = uniqstr_new (last_string);
     STRING_FREE;

-    return TYPE;
+    return TAG;
+  }
+  "<"         {
+    nesting = 0;
+    token_start = loc->start;
+    BEGIN SC_TAG;

   }
 
   "%%" {
   }
 
   "%%" {


@@ -267,6 +277,17 @@ splice      (\\[ \f\t\v]*\n)*
 }
 
 
 }
 
 

+  /*--------------------------------------------------------------.
+  | Supporting \0 complexifies our implementation for no expected |
+  | added value.                                                  |
+  `--------------------------------------------------------------*/
+
+<SC_ESCAPED_CHARACTER,SC_ESCAPED_STRING,SC_TAG>
+{
+  \0       complain_at (*loc, _("invalid null character"));
+}
+
+

   /*-----------------------------------------------------------------.
   | Scanning after an identifier, checking whether a colon is next.  |
   `-----------------------------------------------------------------*/
   /*-----------------------------------------------------------------.
   | Scanning after an identifier, checking whether a colon is next.  |
   `-----------------------------------------------------------------*/


@@ -386,11 +407,40 @@ splice     (\\[ \f\t\v]*\n)*
   }
 }
 
   }
 }
 

-<SC_ESCAPED_CHARACTER,SC_ESCAPED_STRING>
+  /*-----------------------------------------------------------.
+  | Scanning a Bison nested tag.  The initial angle bracket is |
+  | already eaten.                                             |
+  `-----------------------------------------------------------*/
+
+<SC_TAG>

 {
 {

-  \0       complain_at (*loc, _("invalid null character"));
-}
+  ">" {
+    --nesting;
+    if (nesting < 0)
+      {
+        STRING_FINISH;
+        loc->start = token_start;
+        val->uniqstr = uniqstr_new (last_string);
+        STRING_FREE;
+        BEGIN INITIAL;
+        return TAG;
+      }
+    STRING_GROW;
+  }
+
+  [^<>]+ STRING_GROW;
+  "<"+   STRING_GROW; nesting += yyleng;

 
 

+  <<EOF>> {
+    unexpected_eof (token_start, ">");
+    STRING_FINISH;
+    loc->start = token_start;
+    val->uniqstr = uniqstr_new (last_string);
+    STRING_FREE;
+    BEGIN INITIAL;
+    return TAG;
+  }
+}

 
   /*----------------------------.
   | Decode escaped characters.  |
 
   /*----------------------------.
   | Decode escaped characters.  |


@@ -509,13 +559,13 @@ splice     (\\[ \f\t\v]*\n)*
 
 <SC_BRACED_CODE>
 {
 
 <SC_BRACED_CODE>
 {

-  "{"|"<"{splice}"%"  STRING_GROW; braces_level++;
-  "%"{splice}">"      STRING_GROW; braces_level--;
+  "{"|"<"{splice}"%"  STRING_GROW; nesting++;
+  "%"{splice}">"      STRING_GROW; nesting--;

   "}" {
     obstack_1grow (&obstack_for_string, '}');
 
   "}" {
     obstack_1grow (&obstack_for_string, '}');
 

-    --braces_level;
-    if (braces_level < 0)
+    --nesting;
+    if (nesting < 0)

       {
        STRING_FINISH;
        loc->start = code_start;
       {
        STRING_FINISH;
        loc->start = code_start;