never iterate on the edge_vector()

This fixes #552, reported by Rüdiger and Ayrat.

* tests/sanity/style.test: Warn aginst iterations on edge_vector.
* spot/parseaut/parseaut.yy, spot/twaalgos/complete.cc,
spot/twaalgos/parity.cc: Iterate over edges(), not edge_vector().
* tests/core/ltlcross.test: Add a test case for #552.
* NEWS: Mention the bug.

NEWS

@@ -156,6 +156,16 @@ New in spot 2.11.6.dev  (not yet released)
     could fail to equate two "t" condition, because we have two ways
     to represent "t": the empty condition, or the empty "Inf({})".
 
+  - Functions complement() and change_parity() could incorrectly read
+    or write the unused edge #0.  In the case of complement(), writing
+    that edge was usually harmless.  However in some scenario,
+    complement could need to stick a ⓪ acceptance mark on edge #0,
+    then the acceptance condition could be simplified to "t", and
+    finally change_parity could be confused to find such an accepting
+    mark in an automaton that declares no colors, and perform some
+    computation on that color that caused it to crash with a "Too many
+    acceptance sets used" message.  (issue #552)
+
 New in spot 2.11.6  (2023-08-01)
 
   Bug fixes:

spot/parseaut/parseaut.yy

@@ -2556,7 +2556,7 @@ static void fix_acceptance(result_& r)
   auto onlyneg = r.neg_acc_sets - r.pos_acc_sets;
   if (onlyneg)
     {
-      for (auto& t: r.h->aut->edge_vector())
+      for (auto& t: r.h->aut->edges())
 	t.acc ^= onlyneg;
     }
 
@@ -2570,7 +2570,7 @@ static void fix_acceptance(result_& r)
   if (both)
     {
       base = acc.add_sets(both.count());
-      for (auto& t: r.h->aut->edge_vector())
+      for (auto& t: r.h->aut->edges())
         {
           unsigned i = 0;
 	  if ((t.acc & both) != both)

spot/twaalgos/complete.cc

@@ -144,7 +144,7 @@ namespace spot
             if (need_acc_fix)
               {
                 auto a = aut->set_buchi();
-                for (auto& t: aut->edge_vector())
+                for (auto& t: aut->edges())
                   t.acc = a;
                 if (aut->num_edges())
                   acc = a;

spot/twaalgos/parity.cc

@@ -50,7 +50,7 @@ namespace spot
     change_acc(twa_graph_ptr& aut, unsigned num_sets, bool change_kind,
                bool change_style, bool output_max, bool input_max)
     {
-      for (auto& e: aut->edge_vector())
+      for (auto& e: aut->edges())
         if (e.acc)
           {
             unsigned msb = (input_max ? e.acc.max_set() : e.acc.min_set()) - 1;

tests/core/ltlcross.test

@@ -69,7 +69,12 @@ ltlcross --verbose ltl2tgba ltl2tgba \
 # Issue #524.
 ltlcross ltl2tgba -f '!(X(v3 | G!v5) | ((Xv5 & !(v5 & !X!v3)) U !v5))'
 
-# Issue #546.
-ltlcross 'ltl2tgba --medium -p' 'ltl2tgba -p' 'ltl2tgba --medium -D -p' \
+# Issue #546, Issue #552.
+ltlcross 'ltl2tgba --medium -p' \
+         'ltl2tgba -p' \
+         'ltl2tgba --medium -D -p' \
+         'ltl2tgba --medium --colored-parity="min even" -C -D' \
+         'ltl2tgba --colored-parity="min even" -C -D' \
          -f 'a | FGa | GF(!b | Gb)' \
-         -f '(~ v1 U ~ v5) | G(F v9 & F G v1) | G F(~ v7 | G v7 | G v3)'
+         -f '(~ v1 U ~ v5) | G(F v9 & F G v1) | G F(~ v7 | G v7 | G v3)' \
+         -f 'FG((a | Fb | FG!b) & !G(c & d))'

tests/sanity/style.test

@@ -298,6 +298,11 @@ for dir in "$TOP/spot" "$TOP/bin" "$TOP/tests"; do
       $GREP 'catch *([^.]' $tmp | $GREP -v 'const.*&' &&
 	diag 'Always capture exceptions by const reference.'
 
+      # iterating over edge_vector() is suspicious, because the
+      # first edge should not be used.
+      $GREP 'for (.*:.*edge_vector()' $tmp &&
+	diag 'Did you mean to iterate over edges()?'
+
       case $file in
 	*.hh | *.hxx)
 	  if $GREP -E '(<<|>>)' $tmp >/dev/null; then