summaryrefslogtreecommitdiffstats
path: root/src/parsers.cc
blob: 4d6c058de9ef08df6bf28f550e3c4f72bb616650 (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
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
#include "parsers.h"

#include <assert.h>
#include <errno.h>
#include <stdio.h>
#include <string.h>

#include "ninja.h"

string Token::AsString() const {
  switch (type_) {
  case IDENT:    return "'" + extra_ + "'";
  case UNKNOWN:  return "unknown '" + extra_ + "'";
  case RULE:     return "'rule'";
  case BUILD:    return "'build'";
  case SUBNINJA: return "'subninja'";
  case NEWLINE:  return "newline";
  case EQUALS:   return "'='";
  case COLON:    return "':'";
  case PIPE:     return "'|'";
  case TEOF:     return "eof";
  case INDENT:   return "indenting in";
  case OUTDENT:  return "indenting out";
  case NONE:
  default:
    assert(false);
    return "";
  }
}

void Tokenizer::Start(const char* start, const char* end) {
  cur_line_ = cur_ = start;
  end_ = end;
}

bool Tokenizer::Error(const string& message, string* err) {
  char buf[1024];
  sprintf(buf, "line %d, col %d: %s",
          line_number_,
          (int)(token_.pos_ - cur_line_) + 1,
          message.c_str());
  err->assign(buf);
  return false;
}

void Tokenizer::SkipWhitespace(bool newline) {
  if (token_.type_ == Token::NEWLINE && newline)
    Newline(NULL);

  while (cur_ < end_) {
    if (*cur_ == ' ') {
      ++cur_;
    } else if (newline && *cur_ == '\n') {
      Newline(NULL);
    } else if (*cur_ == '\\' && cur_ + 1 < end_ && cur_[1] == '\n') {
      ++cur_; ++cur_;
      cur_line_ = cur_;
      ++line_number_;
    } else if (*cur_ == '#' && cur_ == cur_line_) {
      while (cur_ < end_ && *cur_ != '\n')
        ++cur_;
      if (cur_ < end_ && *cur_ == '\n') {
        ++cur_;
        cur_line_ = cur_;
        ++line_number_;
      }
    } else {
      break;
    }
  }
}

bool Tokenizer::Newline(string* err) {
  if (!ExpectToken(Token::NEWLINE, err))
    return false;

  return true;
}

static bool IsIdentChar(char c) {
  return
    ('a' <= c && c <= 'z') ||
    ('+' <= c && c <= '9') ||  // +,-./ and numbers
    ('A' <= c && c <= 'Z') ||
    (c == '_') || (c == '@');
}

bool Tokenizer::ExpectToken(Token::Type expected, string* err) {
  PeekToken();
  if (token_.type_ != expected) {
    return Error("expected " + Token(expected).AsString() + ", "
                 "got " + token_.AsString(), err);
  }
  ConsumeToken();
  return true;
}

bool Tokenizer::ReadIdent(string* out) {
  PeekToken();
  if (token_.type_ != Token::IDENT)
    return false;
  out->assign(token_.extra_);
  ConsumeToken();
  return true;
}

bool Tokenizer::ReadToNewline(string* text, string* err) {
  // XXX token_.clear();
  while (cur_ < end_ && *cur_ != '\n') {
    if (*cur_ == '\\') {
      ++cur_;
      if (cur_ >= end_)
        return Error("unexpected eof", err);
      if (*cur_ != '\n') {
        // XXX we just let other backslashes through verbatim now.
        // This may not be wise.
        text->push_back('\\');
        text->push_back(*cur_);
        ++cur_;
        continue;
      }
      ++cur_;
      cur_line_ = cur_;
      ++line_number_;
      SkipWhitespace();
      // Collapse whitespace, but make sure we get at least one space.
      if (text->size() > 0 && text->at(text->size() - 1) != ' ')
        text->push_back(' ');
    } else {
      text->push_back(*cur_);
      ++cur_;
    }
  }
  return Newline(err);
}

Token::Type Tokenizer::PeekToken() {
  if (token_.type_ != Token::NONE)
    return token_.type_;

  token_.pos_ = cur_;
  if (cur_indent_ == -1) {
    cur_indent_ = cur_ - cur_line_;
    if (cur_indent_ != last_indent_) {
      if (cur_indent_ > last_indent_) {
        token_.type_ = Token::INDENT;
      } else if (cur_indent_ < last_indent_) {
        token_.type_ = Token::OUTDENT;
      }
      last_indent_ = cur_indent_;
      return token_.type_;
    }
  }

  if (cur_ >= end_) {
    token_.type_ = Token::TEOF;
    return token_.type_;
  }

  if (IsIdentChar(*cur_)) {
    while (cur_ < end_ && IsIdentChar(*cur_)) {
      token_.extra_.push_back(*cur_);
      ++cur_;
    }
    if (token_.extra_ == "rule")
      token_.type_ = Token::RULE;
    else if (token_.extra_ == "build")
      token_.type_ = Token::BUILD;
    else if (token_.extra_ == "subninja")
      token_.type_ = Token::SUBNINJA;
    else
      token_.type_ = Token::IDENT;
  } else if (*cur_ == ':') {
    token_.type_ = Token::COLON;
    ++cur_;
  } else if (*cur_ == '=') {
    token_.type_ = Token::EQUALS;
    ++cur_;
  } else if (*cur_ == '|') {
    token_.type_ = Token::PIPE;
    ++cur_;
  } else if (*cur_ == '\n') {
    token_.type_ = Token::NEWLINE;
    ++cur_;
    cur_line_ = cur_;
    cur_indent_ = -1;
    ++line_number_;
  }

  SkipWhitespace();

  if (token_.type_ == Token::NONE) {
    token_.type_ = Token::UNKNOWN;
    token_.extra_ = *cur_;
  }

  return token_.type_;
}

void Tokenizer::ConsumeToken() {
  token_.Clear();
}

bool MakefileParser::Parse(const string& input, string* err) {
  tokenizer_.Start(input.data(), input.data() + input.size());

  if (!tokenizer_.ReadIdent(&out_))
    return tokenizer_.Error("expected output filename", err);
  if (!tokenizer_.ExpectToken(Token::COLON, err))
    return false;
  while (tokenizer_.PeekToken() == Token::IDENT) {
    string in;
    tokenizer_.ReadIdent(&in);
    ins_.push_back(in);
  }
  if (!tokenizer_.ExpectToken(Token::NEWLINE, err))
    return false;
  if (!tokenizer_.ExpectToken(Token::TEOF, err))
    return false;

  return true;
}

ManifestParser::ManifestParser(State* state, FileReader* file_reader)
  : state_(state), file_reader_(file_reader) {
  env_ = &state->bindings_;
}
bool ManifestParser::Load(const string& filename, string* err) {
  string contents;
  if (!file_reader_->ReadFile(filename, &contents, err))
    return false;
  return Parse(contents, err);
}

bool ManifestParser::Parse(const string& input, string* err) {
  tokenizer_.Start(input.data(), input.data() + input.size());

  tokenizer_.SkipWhitespace(true);

  while (tokenizer_.token().type_ != Token::TEOF) {
    switch (tokenizer_.PeekToken()) {
      case Token::RULE:
        if (!ParseRule(err))
          return false;
        break;
      case Token::BUILD:
        if (!ParseEdge(err))
          return false;
        break;
      case Token::SUBNINJA:
        if (!ParseSubNinja(err))
          return false;
        break;
      case Token::IDENT: {
        string name, value;
        if (!ParseLet(&name, &value, err))
          return false;

        env_->AddBinding(name, value);
        if (name == "builddir") {
          builddir_ = value;
          if (builddir_.substr(0, 5) == "$root") {
            builddir_ = root_ + builddir_.substr(5);
          }
          if (!builddir_.empty() && builddir_[builddir_.size() - 1] != '/')
            builddir_.push_back('/');
        }
        break;
      }
      case Token::TEOF:
        continue;
      default:
        return tokenizer_.Error("unhandled " + tokenizer_.token().AsString(), err);
    }
    tokenizer_.SkipWhitespace(true);
  }

  return true;
}

bool ManifestParser::ParseRule(string* err) {
  if (!tokenizer_.ExpectToken(Token::RULE, err))
    return false;
  string name;
  if (!tokenizer_.ReadIdent(&name)) {
    return tokenizer_.Error("expected rule name, got " + tokenizer_.token().AsString(),
                         err);
  }
  if (!tokenizer_.Newline(err))
    return false;

  if (state_->LookupRule(name) != NULL) {
    *err = "duplicate rule '" + name + "'";
    return false;
  }

  Rule* rule = new Rule(name);  // XXX scoped_ptr

  if (tokenizer_.PeekToken() == Token::INDENT) {
    tokenizer_.ConsumeToken();

    while (tokenizer_.PeekToken() != Token::OUTDENT) {
      string key, val;
      if (!ParseLet(&key, &val, err))
        return false;

      string parse_err;
      if (key == "command") {
        if (!rule->ParseCommand(val, &parse_err))
          return tokenizer_.Error(parse_err, err);
      } else if (key == "depfile") {
        if (!rule->depfile_.Parse(val, &parse_err))
          return tokenizer_.Error(parse_err, err);
      } else if (key == "description") {
        if (!rule->description_.Parse(val, &parse_err))
          return tokenizer_.Error(parse_err, err);
      } else {
        // Die on other keyvals for now; revisit if we want to add a
        // scope here.
        return tokenizer_.Error("unexpected variable '" + key + "'", err);
      }
    }
    tokenizer_.ConsumeToken();
  }

  if (rule->command_.unparsed().empty())
    return tokenizer_.Error("expected 'command =' line", err);

  state_->AddRule(rule);
  return true;
}

bool ManifestParser::ParseLet(string* name, string* value, string* err) {
  if (!tokenizer_.ReadIdent(name))
    return tokenizer_.Error("expected variable name", err);
  if (!tokenizer_.ExpectToken(Token::EQUALS, err))
    return false;

  // XXX should we tokenize here?  it means we'll need to understand
  // command syntax, though...
  if (!tokenizer_.ReadToNewline(value, err))
    return false;

  // Do @ -> builddir substitution.
  size_t ofs;
  while ((ofs = value->find('@')) != string::npos) {
    value->replace(ofs, 1, builddir_);
    ofs += builddir_.size();
  }

  return true;
}

static string CanonicalizePath(const string& path) {
  string out;
  for (size_t i = 0; i < path.size(); ++i) {
    char in = path[i];
    if (in == '/' &&
        (!out.empty() && *out.rbegin() == '/')) {
      continue;
    }
    out.push_back(in);
  }
  return out;
}

bool ManifestParser::ParseEdge(string* err) {
  vector<string> ins, outs;

  if (!tokenizer_.ExpectToken(Token::BUILD, err))
    return false;

  for (;;) {
    if (tokenizer_.PeekToken() == Token::COLON) {
      tokenizer_.ConsumeToken();
      break;
    }

    string out;
    if (!tokenizer_.ReadIdent(&out))
      return tokenizer_.Error("expected output file list", err);
    outs.push_back(ExpandFile(out));
  }
  // XXX check outs not empty

  string rule_name;
  if (!tokenizer_.ReadIdent(&rule_name))
    return tokenizer_.Error("expected build command name", err);

  const Rule* rule = state_->LookupRule(rule_name);
  if (!rule)
    return tokenizer_.Error("unknown build rule '" + rule_name + "'", err);

  if (!rule->depfile_.empty()) {
    if (outs.size() > 1) {
      return tokenizer_.Error("dependency files only work with single-output "
                           "rules", err);
    }
  }

  for (;;) {
    string in;
    if (!tokenizer_.ReadIdent(&in))
      break;
    ins.push_back(ExpandFile(in));
  }

  // Add all order-only deps, counting how many as we go.
  int order_only = 0;
  if (tokenizer_.PeekToken() == Token::PIPE) {
    tokenizer_.ConsumeToken();
    for (;;) {
      string in;
      if (!tokenizer_.ReadIdent(&in))
        break;
      ins.push_back(ExpandFile(in));
      ++order_only;
    }
  }

  if (!tokenizer_.Newline(err))
    return false;

  // Default to using outer env.
  BindingEnv* env = env_;

  // But use a nested env if there are variables in scope.
  if (tokenizer_.PeekToken() == Token::INDENT) {
    tokenizer_.ConsumeToken();

    // XXX scoped_ptr to handle error case.
    env = new BindingEnv;
    env->parent_ = env_;
    while (tokenizer_.PeekToken() != Token::OUTDENT) {
      string key, val;
      if (!ParseLet(&key, &val, err))
        return false;
      env->AddBinding(key, val);
    }
    tokenizer_.ConsumeToken();
  }

  Edge* edge = state_->AddEdge(rule);
  edge->env_ = env;
  for (vector<string>::iterator i = ins.begin(); i != ins.end(); ++i)
    state_->AddInOut(edge, Edge::IN, *i);
  for (vector<string>::iterator i = outs.begin(); i != outs.end(); ++i)
    state_->AddInOut(edge, Edge::OUT, *i);
  edge->order_only_deps_ = order_only;

  return true;
}

bool ManifestParser::ParseSubNinja(string* err) {
  if (!tokenizer_.ExpectToken(Token::SUBNINJA, err))
    return false;
  string path;
  if (!tokenizer_.ReadIdent(&path))
    return tokenizer_.Error("expected subninja path", err);
  if (!tokenizer_.Newline(err))
    return false;

  string contents;
  if (!file_reader_->ReadFile(path, &contents, err))
    return false;

  ManifestParser subparser(state_, file_reader_);
  // Simulate variable inheritance of builddir.
  subparser.builddir_ = builddir_;
  subparser.env_ = new BindingEnv;
  subparser.env_->parent_ = env_;
  string sub_err;
  if (!subparser.Parse(contents, &sub_err))
    return tokenizer_.Error("in '" + path + "': " + sub_err, err);

  return true;
}

string ManifestParser::ExpandFile(const string& file) {
  string out = file;
  if (!file.empty() && file[0] == '@')
    out = builddir_ + file.substr(1);
  return CanonicalizePath(out);
}