aboutsummaryrefslogtreecommitdiff
path: root/build2/test/script/regex.cxx
blob: 38ac0966a080626831e5db0ca18c049e268d722c (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
// file      : build2/test/script/regex.cxx -*- C++ -*-
// copyright : Copyright (c) 2014-2017 Code Synthesis Ltd
// license   : MIT; see accompanying LICENSE file

#include <build2/test/script/regex.hxx>

#include <algorithm> // copy(), copy_backward()

using namespace std;

namespace build2
{
  namespace test
  {
    namespace script
    {
      namespace regex
      {
        static_assert (alignof (char_string) % 4 == 0,
                       "unexpected char_string alignment");

        static_assert (alignof (char_regex) % 4 == 0,
                       "unexpected char_regex alignment");

        static_assert (sizeof (uintptr_t) > sizeof (int16_t),
                       "unexpected uintptr_t size");

        const line_char line_char::nul (0);
        const line_char line_char::eof (-1);

        // line_char
        //
        // We package the special character into uintptr_t with the following
        // steps:
        //
        // - narrow down int value to int16_t (preserves all the valid values)
        //
        // - convert to uint16_t (bitwise representation stays the same, but no
        //   need to bother with signed value widening, leftmost bits loss on
        //   left shift, etc)
        //
        // - convert to uintptr_t (storage type)
        //
        // - shift left by two bits (the operation is fully reversible as
        //   uintptr_t is wider then uint16_t)
        //
        line_char::
        line_char (int c)
            : data_ (
              (static_cast <uintptr_t> (
                static_cast<uint16_t> (
                  static_cast<int16_t> (c))) << 2) |
              static_cast <uintptr_t> (line_type::special))
        {
          // @@ How can we allow anything for basic_regex but only subset
          //    for our own code?
          //
          const char ex[] = "pn\n\r";

          assert (c == 0  || // Null character.

                  // EOF. Note that is also passed by msvcrt as _Meta_eos
                  // enum value.
                  //
                  c == -1 ||

                  // libstdc++ line/paragraph separators.
                  //
                  c == u'\u2028' || c == u'\u2029' ||

                  (c > 0 && c <= 255 && (
                    // Supported regex special characters.
                    //
                    syntax (c) ||

                    // libstdc++ look-ahead tokens, newline chars.
                    //
                    string::traits_type::find (ex, 4, c) != nullptr)));
        }

        line_char::
        line_char (const char_string& s, line_pool& p)
            : line_char (&(*p.strings.emplace (s).first))
        {
        }

        line_char::
        line_char (char_string&& s, line_pool& p)
            : line_char (&(*p.strings.emplace (move (s)).first))
        {
        }

        line_char::
        line_char (char_regex r, line_pool& p)
            // Note: in C++17 can write as p.regexes.emplace_front(move (r))
            //
            : line_char (&(*p.regexes.emplace (p.regexes.begin (), move (r))))
        {
        }

        bool
        line_char::syntax (char c)
        {
          return string::traits_type::find (
            "()|.*+?{}\\0123456789,=!", 23, c) != nullptr;
        }

        bool
        operator== (const line_char& l, const line_char& r)
        {
          line_type lt (l.type ());
          line_type rt (r.type ());

          if (lt == rt)
          {
            bool res (true);

            switch (lt)
            {
            case line_type::special: res = l.special () == r.special (); break;
            case line_type::regex:   assert (false); break;

              // Note that we use pointers (rather than vales) comparison
              // assuming that the strings must belong to the same pool.
              //
            case line_type::literal: res = l.literal () == r.literal (); break;
            }

            return res;
          }

          // Match literal with regex.
          //
          if (lt == line_type::literal && rt == line_type::regex)
            return regex_match (*l.literal (), *r.regex ());
          else if (rt == line_type::literal && lt == line_type::regex)
            return regex_match (*r.literal (), *l.regex ());

          return false;
        }

        bool
        operator< (const line_char& l, const line_char& r)
        {
          if (l == r)
            return false;

          line_type lt (l.type ());
          line_type rt (r.type ());

          if (lt != rt)
            return lt < rt;

          bool res (false);

          switch (lt)
          {
          case line_type::special: res =  l.special () <  r.special (); break;
          case line_type::literal: res = *l.literal () < *r.literal (); break;
          case line_type::regex:   assert (false); break;
          }

          return res;
        }

        // line_char_locale
        //
        line_char_locale::
        line_char_locale ()
            : locale (locale (),
                      new std::ctype<line_char> ()) // Hidden by ctype bitmask.
        {
        }

        // char_regex
        //
        // Transform regex according to the extended flags {idot}. If regex is
        // malformed then keep transforming, so the resulting string is
        // malformed the same way. We expect the error to be reported by the
        // char_regex ctor.
        //
        static string
        transform (const string& s, char_flags f)
        {
          assert ((f & char_flags::idot) != char_flags::none);

          string r;
          bool escape (false);
          bool cclass (false);

          for (char c: s)
          {
            // Inverse escaping for a dot which is out of the char class
            // brackets.
            //
            bool inverse (c == '.' && !cclass);

            // Handle the escape case. Note that we delay adding the backslash
            // since we may have to inverse things.
            //
            if (escape)
            {
              if (!inverse)
                r += '\\';

              r += c;
              escape = false;

              continue;
            }
            else if (c == '\\')
            {
              escape = true;
              continue;
            }

            // Keep track of being inside the char class brackets, escape if
            // inversion. Note that we never inverse square brackets.
            //
            if (c == '[' && !cclass)
              cclass = true;
            else if (c == ']' && cclass)
              cclass = false;
            else if (inverse)
              r += '\\';

            r += c;
          }

          if (escape) // Regex is malformed but that's not our problem.
            r += '\\';

          return r;
        }

        static char_regex::flag_type
        to_std_flags (char_flags f)
        {
          // Note that ECMAScript flag is implied in the absense of a grammar
          // flag.
          //
          return (f & char_flags::icase) != char_flags::none
            ? char_regex::icase
            : char_regex::flag_type ();
        }

        char_regex::
        char_regex (const char_string& s, char_flags f)
            : base_type ((f & char_flags::idot) != char_flags::none
                         ? transform (s, f)
                         : s,
                         to_std_flags (f))
        {
        }
      }
    }
  }
}

namespace std
{
  using namespace build2::test::script::regex;

  // char_traits<line_char>
  //
  line_char* char_traits<line_char>::
  assign (char_type* s, size_t n, char_type c)
  {
    for (size_t i (0); i != n; ++i)
      s[i] = c;
    return s;
  }

  line_char* char_traits<line_char>::
  move (char_type* d, const char_type* s, size_t n)
  {
    if (n > 0 && d != s)
    {
      // If d < s then it can't be in [s, s + n) range and so using copy() is
      // safe. Otherwise d + n is out of (s, s + n] range and so using
      // copy_backward() is safe.
      //
      if (d < s)
        std::copy (s, s + n, d); // Hidden by char_traits<line_char>::copy().
      else
        copy_backward (s, s + n, d + n);
    }

    return d;
  }

  line_char* char_traits<line_char>::
  copy (char_type* d, const char_type* s, size_t n)
  {
    std::copy (s, s + n, d); // Hidden by char_traits<line_char>::copy().
    return d;
  }

  int char_traits<line_char>::
  compare (const char_type* s1, const char_type* s2, size_t n)
  {
    for (size_t i (0); i != n; ++i)
    {
      if (s1[i] < s2[i])
        return -1;
      else if (s2[i] < s1[i])
        return 1;
    }

    return 0;
  }

  size_t char_traits<line_char>::
  length (const char_type* s)
  {
    size_t i (0);
    while (s[i] != char_type::nul)
      ++i;

    return i;
  }

  const line_char* char_traits<line_char>::
  find (const char_type* s, size_t n, const char_type& c)
  {
    for (size_t i (0); i != n; ++i)
    {
      if (s[i] == c)
        return s + i;
    }

    return nullptr;
  }

  // ctype<line_char>
  //
  locale::id ctype<line_char>::id;

  const line_char* ctype<line_char>::
  is (const char_type* b, const char_type* e, mask* m) const
  {
    while (b != e)
    {
      const char_type& c (*b++);

      *m++ = c.type () == line_type::special && build2::digit (c.special ())
        ? digit
        : 0;
    }

    return e;
  }

  const line_char* ctype<line_char>::
  scan_is (mask m, const char_type* b, const char_type* e) const
  {
    for (; b != e; ++b)
    {
      if (is (m, *b))
        return b;
    }

    return e;
  }

  const line_char* ctype<line_char>::
  scan_not (mask m, const char_type* b, const char_type* e) const
  {
    for (; b != e; ++b)
    {
      if (!is (m, *b))
        return b;
    }

    return e;
  }

  const char* ctype<line_char>::
  widen (const char* b, const char* e, char_type* c) const
  {
    while (b != e)
      *c++ = widen (*b++);

    return e;
  }

  const line_char* ctype<line_char>::
  narrow (const char_type* b, const char_type* e, char def, char* c) const
  {
    while (b != e)
      *c++ = narrow (*b++, def);

    return e;
  }

  // regex_traits<line_char>
  //
  int regex_traits<line_char>::
  value (char_type c, int radix) const
  {
    assert (radix == 8 || radix == 10 || radix == 16);

    if (c.type () != line_type::special)
      return -1;

    const char digits[] = "0123456789ABCDEF";
    const char* d (string::traits_type::find (digits, radix, c.special ()));
    return d != nullptr ? static_cast<int> (d - digits) : -1;
  }
}