sdcc/support/cpp/gcc/spellcheck.h

   1 /* Find near-matches for strings and identifiers.
   2    Copyright (C) 2015-2022 Free Software Foundation, Inc.
   3
   4 This file is part of GCC.
   5
   6 GCC is free software; you can redistribute it and/or modify it under
   7 the terms of the GNU General Public License as published by the Free
   8 Software Foundation; either version 3, or (at your option) any later
   9 version.
  10
  11 GCC is distributed in the hope that it will be useful, but WITHOUT ANY
  12 WARRANTY; without even the implied warranty of MERCHANTABILITY or
  13 FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
  14 for more details.
  15
  16 You should have received a copy of the GNU General Public License
  17 along with GCC; see the file COPYING3.  If not see
  18 <http://www.gnu.org/licenses/>.  */
  19
  20 #ifndef GCC_SPELLCHECK_H
  21 #define GCC_SPELLCHECK_H
  22
  23 typedef unsigned int edit_distance_t;
  24 const edit_distance_t MAX_EDIT_DISTANCE = UINT_MAX;
  25
  26 /* spellcheck.cc  */
  27 extern edit_distance_t
  28 get_edit_distance (const char *s, int len_s,
  29                    const char *t, int len_t);
  30
  31 extern edit_distance_t
  32 get_edit_distance (const char *s, const char *t);
  33
  34 extern const char *
  35 find_closest_string (const char *target,
  36                      const auto_vec<const char *> *candidates);
  37
  38 /* A traits class for describing a string-like type usable by
  39    class best_match.
  40    Specializations should provide the implementations of the following:
  41
  42      static size_t get_length (TYPE);
  43      static const char *get_string (TYPE);
  44
  45    get_string should return a non-NULL ptr, which does not need to be
  46    0-terminated.  */
  47
  48 template <typename TYPE>
  49 struct edit_distance_traits {
  50   static size_t get_length (TYPE const&)
  51   {
  52     gcc_assert (0);
  53          return 0;
  54   }
  55   static const char *get_string (TYPE const&)
  56   {
  57     gcc_assert (0);
  58     return 0;
  59   }
  60 };
  61
  62 /* Specialization of edit_distance_traits for C-style strings.  */
  63
  64 template <>
  65 struct edit_distance_traits<const char *>
  66 {
  67   static size_t get_length (const char *str)
  68   {
  69     gcc_assert (str);
  70     return strlen (str);
  71   }
  72
  73   static const char *get_string (const char *str)
  74   {
  75     gcc_assert (str);
  76     return str;
  77   }
  78 };
  79
  80 extern edit_distance_t get_edit_distance_cutoff (size_t goal_len,
  81                                                  size_t candidate_len);
  82
  83 /* A type for use when determining the best match against a string,
  84    expressed as a template so that we can match against various
  85    string-like types (const char *, frontend identifiers, and preprocessor
  86    macros).
  87
  88    This type accumulates the best possible match against GOAL_TYPE for
  89    a sequence of elements of CANDIDATE_TYPE, whilst minimizing the
  90    number of calls to get_edit_distance and to
  91    edit_distance_traits<T>::get_length.  */
  92
  93 template <typename GOAL_TYPE, typename CANDIDATE_TYPE>
  94 class best_match
  95 {
  96  public:
  97   typedef GOAL_TYPE goal_t;
  98   typedef CANDIDATE_TYPE candidate_t;
  99   typedef edit_distance_traits<goal_t> goal_traits;
 100   typedef edit_distance_traits<candidate_t> candidate_traits;
 101
 102   /* Constructor.  */
 103
 104   best_match (GOAL_TYPE goal,
 105               edit_distance_t best_distance_so_far = MAX_EDIT_DISTANCE)
 106   : m_goal (goal_traits::get_string (goal)),
 107     m_goal_len (goal_traits::get_length (goal)),
 108     m_best_candidate (NULL),
 109     m_best_distance (best_distance_so_far)
 110   {}
 111
 112   /* Compare the edit distance between CANDIDATE and m_goal,
 113      and if it's the best so far, record it.  */
 114
 115   void consider (candidate_t candidate)
 116   {
 117     size_t candidate_len = candidate_traits::get_length (candidate);
 118
 119     /* Calculate a lower bound on the candidate's distance to the goal,
 120        based on the difference in lengths; it will require at least
 121        this many insertions/deletions.  */
 122     edit_distance_t min_candidate_distance
 123       = abs ((ssize_t)candidate_len - (ssize_t)m_goal_len);
 124
 125     /* If the candidate's length is sufficiently different to that
 126        of the goal string, then the number of insertions/deletions
 127        may be >= the best distance so far.  If so, we can reject
 128        the candidate immediately without needing to compute
 129        the exact distance, since it won't be an improvement.  */
 130     if (min_candidate_distance >= m_best_distance)
 131       return;
 132
 133     /* If the candidate will be unable to beat the criterion in
 134        get_best_meaningful_candidate, reject it without computing
 135        the exact distance.  */
 136     edit_distance_t cutoff = get_cutoff (candidate_len);
 137     if (min_candidate_distance > cutoff)
 138       return;
 139
 140     /* Otherwise, compute the distance and see if the candidate
 141        has beaten the previous best value.  */
 142     edit_distance_t dist
 143       = get_edit_distance (m_goal, m_goal_len,
 144                            candidate_traits::get_string (candidate),
 145                            candidate_len);
 146     if (dist < m_best_distance)
 147       {
 148         m_best_distance = dist;
 149         m_best_candidate = candidate;
 150         m_best_candidate_len = candidate_len;
 151       }
 152   }
 153
 154   /* Assuming that BEST_CANDIDATE is known to be better than
 155      m_best_candidate, update (without recomputing the edit distance to
 156      the goal).  */
 157
 158   void set_best_so_far (CANDIDATE_TYPE best_candidate,
 159                         edit_distance_t best_distance,
 160                         size_t best_candidate_len)
 161   {
 162     gcc_assert (best_distance < m_best_distance);
 163     m_best_candidate = best_candidate;
 164     m_best_distance = best_distance;
 165     m_best_candidate_len = best_candidate_len;
 166   }
 167
 168   /* Generate the maximum edit distance for which we consider a suggestion
 169      to be meaningful, given a candidate of length CANDIDATE_LEN.  */
 170
 171   edit_distance_t get_cutoff (size_t candidate_len) const
 172   {
 173     return ::get_edit_distance_cutoff (m_goal_len, candidate_len);
 174   }
 175
 176   /* Get the best candidate so far, but applying a filter to ensure
 177      that we return NULL if none of the candidates are close to the goal,
 178      to avoid offering nonsensical suggestions to the user.  */
 179
 180   candidate_t get_best_meaningful_candidate () const
 181   {
 182     /* If the edit distance is too high, the suggestion is likely to be
 183        meaningless.  */
 184     if (m_best_candidate)
 185       {
 186         edit_distance_t cutoff = get_cutoff (m_best_candidate_len);
 187         if (m_best_distance > cutoff)
 188           return NULL;
 189     }
 190
 191     /* If the goal string somehow makes it into the candidate list, offering
 192        it as a suggestion will be nonsensical e.g.
 193          'constexpr' does not name a type; did you mean 'constexpr'?
 194        Ultimately such suggestions are due to bugs in constructing the
 195        candidate list, but as a band-aid, do not offer suggestions for
 196        distance == 0 (where candidate == goal).  */
 197     if (m_best_distance == 0)
 198       return NULL;
 199
 200     return m_best_candidate;
 201   }
 202
 203   /* Get the closest candidate so far, without applying any filtering.  */
 204
 205   candidate_t blithely_get_best_candidate () const
 206   {
 207     return m_best_candidate;
 208   }
 209
 210   edit_distance_t get_best_distance () const { return m_best_distance; }
 211   size_t get_best_candidate_length () const { return m_best_candidate_len; }
 212
 213  private:
 214   const char *m_goal;
 215   size_t m_goal_len;
 216   candidate_t m_best_candidate;
 217   edit_distance_t m_best_distance;
 218   size_t m_best_candidate_len;
 219 };
 220
 221 #endif  /* GCC_SPELLCHECK_H  */