[PATCH 2/4] m4: catch suspicions of unevaluated macros

[Akim Demaille]

Check in m4's output if there are sequences such as m4_foo or b4_foo,
which are probably resulting from incorrect m4 processing.

It actually already is useful:

- it caught a leaking b4_lac_if leaking from glr.c, where LAC is not
  supported, hence b4_lac_if is not defined.

- it also caught references to location.hh in position.hh when
  location.hh does not exist.

- while making "Code injection" robust to these new warnings (it is
  its very purpose to let b4_canary pass unevaluated), I saw that it
  did not check lalr1.d, and when adding lalr1.d, it revealed it did
  underquote ocurrences of token value types.

* src/scan-skel.l (macro): New abbreviation.
Use it.
* data/skeletons/glr.c: Don't use b4_lac_if, we don't have it.
* data/skeletons/location.cc: Don't generate position.hh when we don't
generate location.hh.
* data/skeletons/d.m4 (b4_basic_symbol_constructor_define): Fix
underquotation.
* data/skeletons/bison.m4 (b4_canary): New.
* tests/input.at (Code injection): Use it, and check lalr1.d too.
---
 TODO                       |  4 ----
 data/skeletons/bison.m4    |  7 +++++++
 data/skeletons/d.m4        |  4 ++--
 data/skeletons/glr.c       |  9 ++-------
 data/skeletons/location.cc | 17 ++++++++++-------
 src/scan-skel.l            | 19 ++++++++++++++++++-
 tests/input.at             | 37 ++++++++++++++++++-------------------
 7 files changed, 57 insertions(+), 40 deletions(-)

diff --git a/TODO b/TODO
index 8bdee7c3..150d4072 100644
--- a/TODO
+++ b/TODO
@@ -108,10 +108,6 @@ enough.
 *** calc.at
 Stop hard-coding "Calc".  Adjust local.at (look for FIXME).
 
-** A dev warning for b4_
-Maybe we should check for m4_ and b4_ leaking out of the m4 processing, as
-Autoconf does.  It would have caught over-quotation issues.
-
 ** doc
 I feel it's ugly to use the GNU style to declare functions in the doc.  It
 generates tons of white space in the page, and may contribute to bad page
diff --git a/data/skeletons/bison.m4 b/data/skeletons/bison.m4
index 6d92e4f9..488ad610 100644
--- a/data/skeletons/bison.m4
+++ b/data/skeletons/bison.m4
@@ -261,6 +261,13 @@ m4_define([b4_fatal_at],
 [b4_error([[fatal]], $@)dnl
 m4_exit(1)])
 
+# b4_canary(MSG)
+# --------------
+# Issue a warning on stderr and in the output.  Used in the test suite
+# to catch spurious m4 evaluations.
+m4_define([b4_canary],
+[m4_errprintn([dead canary: $1])DEAD CANARY($1)])
+
 
 ## ------------ ##
 ## Data Types.  ##
diff --git a/data/skeletons/d.m4 b/data/skeletons/d.m4
index b69df1a7..c0632e47 100644
--- a/data/skeletons/d.m4
+++ b/data/skeletons/d.m4
@@ -586,7 +586,7 @@ alias ACCEPT = ]b4_parser_class[.YYACCEPT;]])[]])[
 # Create Symbol struct constructors for all the visible types.
 m4_define([b4_basic_symbol_constructor_define],
 [b4_token_visible_if([$1],
-[    this(TokenKind token]b4_symbol_if([$1], [has_type],
+[[    this(TokenKind token]b4_symbol_if([$1], [has_type],
 [[, ]b4_union_if([], [[typeof(YYSemanticType.]])b4_symbol([$1], [type])dnl
 []b4_union_if([], [[) ]])[ val]])[]b4_locations_if([[, Location loc]])[)
     {
@@ -601,7 +601,7 @@ m4_define([b4_basic_symbol_constructor_define],
       value_.]b4_symbol([$1], [type])[ = val;]])])[]b4_locations_if([
       location_ = loc;])[
     }
-])])
+]])])
 
 
 # b4_symbol_type_define
diff --git a/data/skeletons/glr.c b/data/skeletons/glr.c
index 1f956955..70ab8b1d 100644
--- a/data/skeletons/glr.c
+++ b/data/skeletons/glr.c
@@ -2204,12 +2204,7 @@ yy_syntax_error_arguments (const yyGLRStack* yystackp,
      - Don't assume there isn't a lookahead just because this state is a
        consistent state with a default action.  There might have been a
        previous inconsistent state, consistent state with a non-default
-       action, or user semantic action that manipulated yychar.]b4_lac_if([[
-       In the first two cases, it might appear that the current syntax
-       error should have been detected in the previous state when yy_lac
-       was invoked.  However, at that time, there might have been a
-       different syntax error that discarded a different initial context
-       during error recovery, leaving behind the current lookahead.]], [[
+       action, or user semantic action that manipulated yychar.
      - Of course, the expected token list depends on states to have
        correct lookahead information, and it depends on the parser not
        to perform extra reductions after fetching a lookahead from the
@@ -2217,7 +2212,7 @@ yy_syntax_error_arguments (const yyGLRStack* yystackp,
        (from LALR or IELR) and default reductions corrupt the expected
        token list.  However, the list is correct for canonical LR with
        one exception: it will still contain any token that will not be
-       accepted due to an error action in a later state.]])[
+       accepted due to an error action in a later state.
   */
   if (yytoken != ]b4_symbol(empty, kind)[)
     {
diff --git a/data/skeletons/location.cc b/data/skeletons/location.cc
index b32c1377..3870b2bc 100644
--- a/data/skeletons/location.cc
+++ b/data/skeletons/location.cc
@@ -19,13 +19,6 @@ m4_pushdef([b4_copyright_years],
            [2002-2015, 2018-2021])
 
 
-# b4_position_file
-# ----------------
-# Name of the file containing the position class, if we want this file.
-b4_header_if([b4_required_version_if([30200], [],
-                                     [m4_define([b4_position_file], 
[position.hh])])])])
-
-
 # b4_location_file
 # ----------------
 # Name of the file containing the position/location class,
@@ -54,6 +47,16 @@ m4_ifdef([b4_location_file],
  ])
 
 
+# b4_position_file
+# ----------------
+# Name of the file containing the position class, if we want this file.
+b4_header_if(
+  [b4_required_version_if(
+    [30200], [],
+    [m4_ifdef([b4_location_file],
+              [m4_define([b4_position_file], [position.hh])])])])
+
+
 
 # b4_location_define
 # ------------------
diff --git a/src/scan-skel.l b/src/scan-skel.l
index 20049cfb..60232a2e 100644
--- a/src/scan-skel.l
+++ b/src/scan-skel.l
@@ -53,6 +53,9 @@ static void fail_for_invalid_at (char const *at);
 static void output_mapped_file (char const *name);
 %}
 
+/* Identifiers of our M4 macros.  */
+macro    [bm]4_[a-zA-Z_0-9]*
+
 %x SC_AT_DIRECTIVE_ARGS
 %x SC_AT_DIRECTIVE_SKIP_WS
 
@@ -87,7 +90,21 @@ static void output_mapped_file (char const *name);
   /* This pattern must not match more than the previous @ patterns. */
 @[^@{}''(\n]*   fail_for_invalid_at (yytext);
 \n              out_lineno++; ECHO;
-[^@\n]+         ECHO;
+[^bm@\n]+       ECHO;
+
+  /* If there are still identifiers that look like macros, such as
+     b4_synbol, this probably an error, say a typo in M4, or
+     overquotation.  */
+{macro}         {
+                  location loc = empty_loc;
+                  loc.start.file = map_file_name (out_name);
+                  loc.start.line = out_lineno;
+                  loc.end = loc.start;
+                  complain (&loc, Wother,
+                            "suspicious sequence in the output: %s", yytext);
+                  ECHO;
+                }
+.               ECHO;
 
 <INITIAL><<EOF>> {
   if (out_name)
diff --git a/tests/input.at b/tests/input.at
index 5f298bc9..f41c3657 100644
--- a/tests/input.at
+++ b/tests/input.at
@@ -2889,26 +2889,26 @@ AT_SETUP([[Code injection]])
 # Try to have MACRO be run by bison.
 m4_pushdef([AT_TEST],
 [AT_DATA([[input.y]],
-[[%type <$1(DEAD %type)> exp
-%token <$1(DEAD %token)> a
+[[%type <$1(%type)> exp
+%token <$1(%token)> a
 %token b
 %initial-action
 {
   $$;
-  $<$1(DEAD %initial-action)>$
+  $<$1(%initial-action)>$
 };
 %printer
 {
   $$
-  $<$1(DEAD %printer)>$
+  $<$1(%printer)>$
 } <> <*>;
 %lex-param
 {
-  $1(DEAD %lex-param)
+  $1(%lex-param)
 };
 %parse-param
 {
-  $1(DEAD %parse-param)
+  $1(%parse-param)
 };
 %%
 exp:
@@ -2916,25 +2916,24 @@ m4_pushdef([AT_TEST],
   {
     $$;
     $][1;
-    $<$1(DEAD action 1)>$
-    $<$1(DEAD action 2)>1
-    $<$1(DEAD action 3)>name
-    $<$1(DEAD action 4)>0
+    $<$1(action 1)>$
+    $<$1(action 2)>1
+    $<$1(action 3)>name
+    $<$1(action 4)>0
     ;
   };
 ]])
 
-# FIXME: Provide a means to iterate over all the skeletons.
-AT_BISON_CHECK([[-d               input.y]])
-AT_BISON_CHECK([[-d -S glr.c      input.y]])
-AT_BISON_CHECK([[-d -S lalr1.cc   input.y]])
-AT_BISON_CHECK([[-d -S glr.cc     input.y]])
-AT_BISON_CHECK([[-d -S glr2.cc    input.y]])
-AT_BISON_CHECK([[   -S lalr1.java input.y]])
+# Disable -Wother to avoid the warnings about the suspicious presence
+# of `b4_canary` in the output.
+m4_foreach([b4_skel],
+           [[yacc.c], [glr.c], [lalr1.cc], [glr.cc], [glr2.cc], [lalr1.d], 
[lalr1.java]],
+[AT_BISON_CHECK([[-Wno-other -S ]b4_skel[ ]m4_bmatch(b4_skel, [.*\.java$], [], 
[-d])[ input.y]])
+])
 ])
 
-AT_TEST([m4_errprintn])
-AT_TEST([@:>@m4_errprintn])
+AT_TEST([b4_canary])
+AT_TEST([@:>@b4_canary])
 
 m4_popdef([AT_TEST])
 
-- 
2.32.0