word: implement twa word parsing

* spot/twaalgos/word.hh: add parse_word method and a new constructor
* spot/twaalgos/word.cc: implement word parsing
* python/spot/__init__.py: add parse_word method binding
* tests/python/word.ipynb: add word parsing tests

python/spot/__init__.py

@@ -860,3 +860,7 @@ def sat_minimize(aut, acc=None, colored=False,
         args += ',dichotomy';
     from spot.impl import sat_minimize as sm
     return sm(aut, args, state_based)
+
+def parse_word(word, dic=_bdd_dict):
+    from spot.impl import parse_word as pw
+    return pw(word, dic)

spot/twaalgos/word.cc

@@ -1,6 +1,6 @@
 // -*- coding: utf-8 -*-
-// Copyright (C) 2013, 2014, 2015 Laboratoire de Recherche et Développement
+// Copyright (C) 2013, 2014, 2015, 2016 Laboratoire de Recherche et
-// de l'Epita (LRDE).
+// Développement de l'Epita (LRDE).
 //
 // This file is part of Spot, a model checking library.
 //
@@ -20,10 +20,13 @@
 #include <spot/twaalgos/word.hh>
 #include <spot/twa/bddprint.hh>
 #include <spot/twa/bdddict.hh>
+#include <spot/tl/parse.hh>
+#include <spot/tl/simplify.hh>
+#include <spot/tl/apcollect.hh>
 
 namespace spot
 {
-  twa_word::twa_word(const twa_run_ptr run)
+  twa_word::twa_word(const twa_run_ptr& run)
     : dict_(run->aut->get_dict())
   {
     for (auto& i: run->prefix)
@@ -33,6 +36,11 @@ namespace spot
     dict_->register_all_variables_of(run->aut, this);
   }
 
+  twa_word::twa_word(const bdd_dict_ptr& dict)
+    : dict_(dict)
+  {
+  }
+
   void
   twa_word::simplify()
   {
@@ -105,4 +113,143 @@ namespace spot
     os << '}';
     return os;
   }
+
+  namespace
+  {
+    static void word_parse_error(const std::string& word,
+				 size_t i, parse_error_list pel)
+    {
+      for (auto& err: pel)
+	{
+	  err.first.begin.column += i;
+	  err.first.end.column += i;
+	}
+      std::ostringstream os;
+      format_parse_errors(os, word, pel);
+      throw parse_error(os.str());
+    }
+
+    static void word_parse_error(const std::string& word, size_t i,
+				 const std::string& message)
+    {
+      if (i == std::string::npos)
+	i = word.size();
+      std::ostringstream s;
+      s << ">>> " << word << '\n';
+      for (auto j = i + 4; j > 0; --j)
+	s << ' ';
+      s << '^' << '\n';
+      s << message << '\n';
+      throw parse_error(s.str());
+    }
+
+    static size_t skip_next_formula(const std::string& word, size_t begin)
+    {
+      bool quoted = false;
+      auto size = word.size();
+      for (auto j = begin; j < size; ++j)
+	{
+	  auto c = word[j];
+	  if (!quoted && (c == ';' || c == '}'))
+	    return j;
+	  if (c == '"')
+	    quoted = !quoted;
+	  else if (quoted && c == '\\')
+	    ++j;
+	}
+      if (quoted)
+	word_parse_error(word, word.size(), "Unclosed string");
+      return std::string::npos;
+    }
+  }
+
+  twa_word_ptr parse_word(const std::string& word, const bdd_dict_ptr& dict)
+  {
+    atomic_prop_set aps;
+    parse_error_list pel;
+    tl_simplifier tls(dict);
+    twa_word_ptr tw = make_twa_word(dict);
+    size_t i = 0;
+    auto ind = i;
+
+    auto extract_bdd =
+      [&](typename twa_word::seq_t& seq)
+      {
+	auto sub = word.substr(i, ind - i);
+	formula f = spot::parse_infix_boolean(sub, pel);
+	if (!pel.empty())
+	  word_parse_error(word, i, pel);
+	atomic_prop_collect(f, &aps);
+	seq.push_back(tls.as_bdd(f));
+	if (word[ind] == '}')
+	  return true;
+	// Skip blanks after semi-colon
+	i = word.find_first_not_of(' ', ind + 1);
+	return false;
+      };
+
+    // Parse the prefix part. Can be empty.
+    while (word.substr(i, 6) != std::string("cycle{"))
+      {
+	ind = skip_next_formula(word, i);
+	if (ind == std::string::npos)
+	  word_parse_error(word, word.size(),
+			   "A twa_word must contain a cycle");
+	if (word[ind] == '}')
+	  word_parse_error(word, ind, "Expected ';' delimiter :"
+			   "'}' stands for ending a cycle");
+	// Exract formula, convert it to bdd and add it to the prefix sequence
+	extract_bdd(tw->prefix);
+	if (i == std::string::npos)
+	  word_parse_error(word, ind + 1, "Missing cycle in formula");
+      }
+    // Consume "cycle{"
+    i += 6;
+    while (true)
+      {
+	ind = skip_next_formula(word, i);
+	if (ind == std::string::npos)
+	  word_parse_error(word, word.size(),
+			   "Missing ';' or '}' after formula");
+	// Extract formula, convert it to bdd and add it to the cycle sequence
+	// Break if an '}' is encountered
+	if (extract_bdd(tw->cycle))
+	  break;
+	if (i == std::string::npos)
+	  word_parse_error(word, ind + 1,
+			   "Missing end of cycle character : '}'");
+      }
+    if (ind != word.size() - 1)
+      word_parse_error(word, ind + 1, "Input should be finished after cycle");
+    for (auto ap: aps)
+      dict->register_proposition(ap, tw.get());
+    return tw;
+  }
+
+  twa_graph_ptr twa_word::as_automaton() const
+  {
+    twa_graph_ptr aut = make_twa_graph(dict_);
+
+    aut->prop_weak(true);
+    aut->prop_deterministic(true);
+
+    size_t i = 0;
+    aut->new_states(prefix.size() + cycle.size());
+    for (auto b: prefix)
+      {
+	aut->new_edge(i, i + 1, b);
+	++i;
+      }
+    size_t j = i;
+    auto b = cycle.begin();
+    auto end = --cycle.end();
+    for (; b != end; ++b)
+      {
+	aut->new_edge(i, i + 1, *b);
+	++i;
+      }
+    // Close the loop
+    aut->new_edge(i, j, *b);
+    return aut;
+  }
 }

spot/twaalgos/word.hh

@@ -1,5 +1,5 @@
 // -*- coding: utf-8 -*-
-// Copyright (C) 2013, 2014, 2015 Laboratoire de Recherche et
+// Copyright (C) 2013, 2014, 2015, 2016 Laboratoire de Recherche et
 // Développement de l'Epita (LRDE).
 //
 // This file is part of Spot, a model checking library.
@@ -28,7 +28,8 @@ namespace spot
   /// \brief An infinite word stored as a lasso.
   struct SPOT_API twa_word final
   {
-    twa_word(const twa_run_ptr run);
+    twa_word(const bdd_dict_ptr& dict);
+    twa_word(const twa_run_ptr& run);
     ~twa_word()
     {
       dict_->unregister_all_my_variables(this);
@@ -45,10 +46,26 @@ namespace spot
       return dict_;
     }
 
+    twa_graph_ptr as_automaton() const;
+
     SPOT_API
     friend std::ostream& operator<<(std::ostream& os, const twa_word& w);
   private:
     bdd_dict_ptr dict_;
   };
 
+  typedef std::shared_ptr<twa_word> twa_word_ptr;
+
+  inline twa_word_ptr make_twa_word(const bdd_dict_ptr& dict)
+  {
+    return std::make_shared<twa_word>(dict);
+  }
+
+  inline twa_word_ptr make_twa_word(const twa_run_ptr& run)
+  {
+    return std::make_shared<twa_word>(run);
+  }
+
+  SPOT_API
+  twa_word_ptr parse_word(const std::string& word, const bdd_dict_ptr& dict);
 }

tests/python/word.ipynb

@@ -1,7 +1,7 @@
 {
  "metadata": {
   "name": "",
-  "signature": "sha256:e975b9a64657f38248c434a89f29b28f0bf90f9c40b7e8afdd7459734cbdcd38"
+  "signature": "sha256:6461a5fa3f321dac24ae47d88c90f911822e003b658d05d4ef7c102ff64e058e"
  },
  "nbformat": 3,
  "nbformat_minor": 0,
@@ -139,7 +139,7 @@
         "</svg>\n"
        ],
        "text": [
-        "<spot.impl.twa_graph; proxy of <Swig Object of type 'std::shared_ptr< spot::twa_graph > *' at 0x7f4f0c2e0810> >"
+        "<spot.impl.twa_graph; proxy of <Swig Object of type 'std::shared_ptr< spot::twa_graph > *' at 0x7fdd0c5d9330> >"
        ]
       }
      ],
@@ -294,6 +294,224 @@
       }
      ],
      "prompt_number": 7
+    },
+    {
+     "cell_type": "code",
+     "collapsed": false,
+     "input": [
+      "print(spot.parse_word('a; b; cycle{a&b}'))\n",
+      "print(spot.parse_word('cycle{a&bb|bac&(aaa|bbb)}'))\n",
+      "print(spot.parse_word('a; b;b; qiwuei;\"a;b&c;a\" ;cycle{a}'))"
+     ],
+     "language": "python",
+     "metadata": {},
+     "outputs": [
+      {
+       "output_type": "stream",
+       "stream": "stdout",
+       "text": [
+        "a; b; cycle{a & b}\n",
+        "cycle{(a & bb) | (aaa & bac) | (bac & bbb)}\n",
+        "a; b; b; qiwuei; \"a;b&c;a\"; cycle{a}\n"
+       ]
+      }
+     ],
+     "prompt_number": 8
+    },
+    {
+     "cell_type": "code",
+     "collapsed": false,
+     "input": [
+      "print(spot.parse_word('a; b&!a; b'))"
+     ],
+     "language": "python",
+     "metadata": {},
+     "outputs": [
+      {
+       "ename": "SyntaxError",
+       "evalue": "\n>>> a; b&!a; b\n              ^\nA twa_word must contain a cycle\n (<string>)",
+       "output_type": "pyerr",
+       "traceback": [
+        "\u001b[0;36m  File \u001b[0;32m\"<string>\"\u001b[0;36m, line \u001b[0;32munknown\u001b[0m\n\u001b[0;31mSyntaxError\u001b[0m\u001b[0;31m:\u001b[0m \n>>> a; b&!a; b\n              ^\nA twa_word must contain a cycle\n\n"
+       ]
+      }
+     ],
+     "prompt_number": 9
+    },
+    {
+     "cell_type": "code",
+     "collapsed": false,
+     "input": [
+      "print(spot.parse_word('a; b; c}'))"
+     ],
+     "language": "python",
+     "metadata": {},
+     "outputs": [
+      {
+       "ename": "SyntaxError",
+       "evalue": "\n>>> a; b; c}\n           ^\nExpected ';' delimiter :'}' stands for ending a cycle\n (<string>)",
+       "output_type": "pyerr",
+       "traceback": [
+        "\u001b[0;36m  File \u001b[0;32m\"<string>\"\u001b[0;36m, line \u001b[0;32munknown\u001b[0m\n\u001b[0;31mSyntaxError\u001b[0m\u001b[0;31m:\u001b[0m \n>>> a; b; c}\n           ^\nExpected ';' delimiter :'}' stands for ending a cycle\n\n"
+       ]
+      }
+     ],
+     "prompt_number": 10
+    },
+    {
+     "cell_type": "code",
+     "collapsed": false,
+     "input": [
+      "print(spot.parse_word('a; cycle{}'))"
+     ],
+     "language": "python",
+     "metadata": {},
+     "outputs": [
+      {
+       "ename": "SyntaxError",
+       "evalue": "\n>>> a; cycle{}\n             ^\nempty input\n\n (<string>)",
+       "output_type": "pyerr",
+       "traceback": [
+        "\u001b[0;36m  File \u001b[0;32m\"<string>\"\u001b[0;36m, line \u001b[0;32munknown\u001b[0m\n\u001b[0;31mSyntaxError\u001b[0m\u001b[0;31m:\u001b[0m \n>>> a; cycle{}\n             ^\nempty input\n\n\n"
+       ]
+      }
+     ],
+     "prompt_number": 11
+    },
+    {
+     "cell_type": "code",
+     "collapsed": false,
+     "input": [
+      "print(spot.parse_word('a; cycle{!a}; a'))"
+     ],
+     "language": "python",
+     "metadata": {},
+     "outputs": [
+      {
+       "ename": "SyntaxError",
+       "evalue": "\n>>> a; cycle{!a}; a\n                ^\nInput should be finished after cycle\n (<string>)",
+       "output_type": "pyerr",
+       "traceback": [
+        "\u001b[0;36m  File \u001b[0;32m\"<string>\"\u001b[0;36m, line \u001b[0;32munknown\u001b[0m\n\u001b[0;31mSyntaxError\u001b[0m\u001b[0;31m:\u001b[0m \n>>> a; cycle{!a}; a\n                ^\nInput should be finished after cycle\n\n"
+       ]
+      }
+     ],
+     "prompt_number": 12
+    },
+    {
+     "cell_type": "code",
+     "collapsed": false,
+     "input": [
+      "w1 = spot.parse_word('a; !a; cycle{a; !a; a}')"
+     ],
+     "language": "python",
+     "metadata": {},
+     "outputs": [],
+     "prompt_number": 13
+    },
+    {
+     "cell_type": "code",
+     "collapsed": false,
+     "input": [
+      "w1.as_automaton()"
+     ],
+     "language": "python",
+     "metadata": {},
+     "outputs": [
+      {
+       "metadata": {},
+       "output_type": "pyout",
+       "prompt_number": 14,
+       "svg": [
+        "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
+        "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
+        " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
+        "<!-- Generated by graphviz version 2.38.0 (20140413.2041)\n",
+        " -->\n",
+        "<!-- Title: G Pages: 1 -->\n",
+        "<svg width=\"406pt\" height=\"74pt\"\n",
+        " viewBox=\"0.00 0.00 406.00 74.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
+        "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 70)\">\n",
+        "<title>G</title>\n",
+        "<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-70 402,-70 402,4 -4,4\"/>\n",
+        "<!-- I -->\n",
+        "<!-- 0 -->\n",
+        "<g id=\"node2\" class=\"node\"><title>0</title>\n",
+        "<ellipse fill=\"#ffffaa\" stroke=\"black\" cx=\"56\" cy=\"-18\" rx=\"18\" ry=\"18\"/>\n",
+        "<text text-anchor=\"middle\" x=\"56\" y=\"-14.3\" font-family=\"Lato\" font-size=\"14.00\">0</text>\n",
+        "</g>\n",
+        "<!-- I&#45;&gt;0 -->\n",
+        "<g id=\"edge1\" class=\"edge\"><title>I&#45;&gt;0</title>\n",
+        "<path fill=\"none\" stroke=\"black\" d=\"M1.15491,-18C2.79388,-18 17.1543,-18 30.6317,-18\"/>\n",
+        "<polygon fill=\"black\" stroke=\"black\" points=\"37.9419,-18 30.9419,-21.1501 34.4419,-18 30.9419,-18.0001 30.9419,-18.0001 30.9419,-18.0001 34.4419,-18 30.9418,-14.8501 37.9419,-18 37.9419,-18\"/>\n",
+        "</g>\n",
+        "<!-- 1 -->\n",
+        "<g id=\"node3\" class=\"node\"><title>1</title>\n",
+        "<ellipse fill=\"#ffffaa\" stroke=\"black\" cx=\"135\" cy=\"-18\" rx=\"18\" ry=\"18\"/>\n",
+        "<text text-anchor=\"middle\" x=\"135\" y=\"-14.3\" font-family=\"Lato\" font-size=\"14.00\">1</text>\n",
+        "</g>\n",
+        "<!-- 0&#45;&gt;1 -->\n",
+        "<g id=\"edge2\" class=\"edge\"><title>0&#45;&gt;1</title>\n",
+        "<path fill=\"none\" stroke=\"black\" d=\"M74.0888,-18C84.5562,-18 98.1196,-18 109.693,-18\"/>\n",
+        "<polygon fill=\"black\" stroke=\"black\" points=\"116.959,-18 109.959,-21.1501 113.459,-18 109.959,-18.0001 109.959,-18.0001 109.959,-18.0001 113.459,-18 109.959,-14.8501 116.959,-18 116.959,-18\"/>\n",
+        "<text text-anchor=\"start\" x=\"92\" y=\"-21.8\" font-family=\"Lato\" font-size=\"14.00\">a</text>\n",
+        "</g>\n",
+        "<!-- 2 -->\n",
+        "<g id=\"node4\" class=\"node\"><title>2</title>\n",
+        "<ellipse fill=\"#ffffaa\" stroke=\"black\" cx=\"218\" cy=\"-18\" rx=\"18\" ry=\"18\"/>\n",
+        "<text text-anchor=\"middle\" x=\"218\" y=\"-14.3\" font-family=\"Lato\" font-size=\"14.00\">2</text>\n",
+        "</g>\n",
+        "<!-- 1&#45;&gt;2 -->\n",
+        "<g id=\"edge3\" class=\"edge\"><title>1&#45;&gt;2</title>\n",
+        "<path fill=\"none\" stroke=\"black\" d=\"M153.178,-18C164.669,-18 179.959,-18 192.693,-18\"/>\n",
+        "<polygon fill=\"black\" stroke=\"black\" points=\"199.847,-18 192.847,-21.1501 196.347,-18 192.847,-18.0001 192.847,-18.0001 192.847,-18.0001 196.347,-18 192.847,-14.8501 199.847,-18 199.847,-18\"/>\n",
+        "<text text-anchor=\"start\" x=\"171\" y=\"-21.8\" font-family=\"Lato\" font-size=\"14.00\">!a</text>\n",
+        "</g>\n",
+        "<!-- 3 -->\n",
+        "<g id=\"node5\" class=\"node\"><title>3</title>\n",
+        "<ellipse fill=\"#ffffaa\" stroke=\"black\" cx=\"297\" cy=\"-48\" rx=\"18\" ry=\"18\"/>\n",
+        "<text text-anchor=\"middle\" x=\"297\" y=\"-44.3\" font-family=\"Lato\" font-size=\"14.00\">3</text>\n",
+        "</g>\n",
+        "<!-- 2&#45;&gt;3 -->\n",
+        "<g id=\"edge4\" class=\"edge\"><title>2&#45;&gt;3</title>\n",
+        "<path fill=\"none\" stroke=\"black\" d=\"M234.954,-24.216C246.105,-28.5603 261.15,-34.4223 273.515,-39.2395\"/>\n",
+        "<polygon fill=\"black\" stroke=\"black\" points=\"280.066,-41.7919 272.4,-42.1858 276.805,-40.5213 273.544,-39.2507 273.544,-39.2507 273.544,-39.2507 276.805,-40.5213 274.687,-36.3156 280.066,-41.7919 280.066,-41.7919\"/>\n",
+        "<text text-anchor=\"start\" x=\"254\" y=\"-36.8\" font-family=\"Lato\" font-size=\"14.00\">a</text>\n",
+        "</g>\n",
+        "<!-- 4 -->\n",
+        "<g id=\"node6\" class=\"node\"><title>4</title>\n",
+        "<ellipse fill=\"#ffffaa\" stroke=\"black\" cx=\"380\" cy=\"-18\" rx=\"18\" ry=\"18\"/>\n",
+        "<text text-anchor=\"middle\" x=\"380\" y=\"-14.3\" font-family=\"Lato\" font-size=\"14.00\">4</text>\n",
+        "</g>\n",
+        "<!-- 3&#45;&gt;4 -->\n",
+        "<g id=\"edge5\" class=\"edge\"><title>3&#45;&gt;4</title>\n",
+        "<path fill=\"none\" stroke=\"black\" d=\"M314.003,-42.073C326.017,-37.6233 342.638,-31.4675 356.064,-26.4947\"/>\n",
+        "<polygon fill=\"black\" stroke=\"black\" points=\"362.753,-24.0174 357.283,-29.4026 359.471,-25.233 356.189,-26.4487 356.189,-26.4487 356.189,-26.4487 359.471,-25.233 355.095,-23.4948 362.753,-24.0174 362.753,-24.0174\"/>\n",
+        "<text text-anchor=\"start\" x=\"333\" y=\"-37.8\" font-family=\"Lato\" font-size=\"14.00\">!a</text>\n",
+        "</g>\n",
+        "<!-- 4&#45;&gt;2 -->\n",
+        "<g id=\"edge6\" class=\"edge\"><title>4&#45;&gt;2</title>\n",
+        "<path fill=\"none\" stroke=\"black\" d=\"M362.276,-13.9842C349.433,-11.115 331.209,-7.50646 315,-6 299.069,-4.51934 294.922,-4.4234 279,-6 266.86,-7.20214 253.537,-9.74069 242.536,-12.1747\"/>\n",
+        "<polygon fill=\"black\" stroke=\"black\" points=\"235.677,-13.744 241.798,-9.11188 239.089,-12.9632 242.501,-12.1825 242.501,-12.1825 242.501,-12.1825 239.089,-12.9632 243.204,-15.2532 235.677,-13.744 235.677,-13.744\"/>\n",
+        "<text text-anchor=\"start\" x=\"293.5\" y=\"-9.8\" font-family=\"Lato\" font-size=\"14.00\">a</text>\n",
+        "</g>\n",
+        "</g>\n",
+        "</svg>\n"
+       ],
+       "text": [
+        "<spot.impl.twa_graph; proxy of <Swig Object of type 'std::shared_ptr< spot::twa_graph > *' at 0x7fdd0c5d96c0> >"
+       ]
+      }
+     ],
+     "prompt_number": 14
+    },
+    {
+     "cell_type": "code",
+     "collapsed": false,
+     "input": [],
+     "language": "python",
+     "metadata": {},
+     "outputs": []
     }
    ],
    "metadata": {}