apibuild.py 94.4 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13
#!/usr/bin/python -u
#
# This is the API builder, it parses the C sources and build the
# API formal description in XML.
#
# See Copyright for the status of this software.
#
# daniel@veillard.com
#
import os, sys
import string
import glob

14 15 16
quiet=True
warnings=0
debug=False
17 18 19 20 21
debugsym=None

#
# C parser analysis code
#
22 23 24 25 26
included_files = {
  "libvirt.h": "header with general libvirt API definitions",
  "virterror.h": "header with error specific API definitions",
  "libvirt.c": "Main interfaces for the libvirt library",
  "virterror.c": "implements error handling and reporting code for libvirt",
27
  "event.c": "event loop for monitoring file handles",
28 29
}

30 31 32 33 34
qemu_included_files = {
  "libvirt-qemu.h": "header with QEMU specific API definitions",
  "libvirt-qemu.c": "Implementations for the QEMU specific APIs",
}

35 36
ignored_words = {
  "ATTRIBUTE_UNUSED": (0, "macro keyword"),
37
  "ATTRIBUTE_SENTINEL": (0, "macro keyword"),
38
  "VIR_DEPRECATED": (0, "macro keyword"),
39
  "VIR_EXPORT_VAR": (0, "macro keyword"),
40 41 42
  "WINAPI": (0, "Windows keyword"),
  "__declspec": (3, "Windows keyword"),
  "__stdcall": (0, "Windows keyword"),
43 44
}

D
Daniel Veillard 已提交
45 46 47 48 49 50
ignored_functions = {
  "virDomainMigrateFinish": "private function for migration",
  "virDomainMigrateFinish2": "private function for migration",
  "virDomainMigratePerform": "private function for migration",
  "virDomainMigratePrepare": "private function for migration",
  "virDomainMigratePrepare2": "private function for migration",
C
Chris Lalancette 已提交
51
  "virDomainMigratePrepareTunnel": "private function for tunnelled migration",
52 53 54 55 56 57
  "virDomainMigrateBegin3": "private function for migration",
  "virDomainMigrateFinish3": "private function for migration",
  "virDomainMigratePerform3": "private function for migration",
  "virDomainMigratePrepare3": "private function for migration",
  "virDomainMigrateConfirm3": "private function for migration",
  "virDomainMigratePrepareTunnel3": "private function for tunnelled migration",
D
Daniel Veillard 已提交
58
  "virDrvSupportsFeature": "private function for remote access",
59
  "DllMain": "specific function for Win32",
60 61 62 63 64 65
  "virEventAddHandle": "internal function in event.c",
  "virEventUpdateHandle": "internal function in event.c",
  "virEventRemoveHandle": "internal function in event.c",
  "virEventAddTimeout": "internal function in event.c",
  "virEventUpdateTimeout": "internal function in event.c",
  "virEventRemoveTimeout": "internal function in event.c",
D
Daniel Veillard 已提交
66 67
}

68 69 70 71 72 73
ignored_macros = {
  "_virSchedParameter": "backward compatibility macro for virTypedParameter",
  "_virBlkioParameter": "backward compatibility macro for virTypedParameter",
  "_virMemoryParameter": "backward compatibility macro for virTypedParameter",
}

74 75 76 77 78 79 80 81 82 83 84 85
def escape(raw):
    raw = string.replace(raw, '&', '&')
    raw = string.replace(raw, '<', '&lt;')
    raw = string.replace(raw, '>', '&gt;')
    raw = string.replace(raw, "'", '&apos;')
    raw = string.replace(raw, '"', '&quot;')
    return raw

def uniq(items):
    d = {}
    for item in items:
        d[item]=1
86 87 88
    k = d.keys()
    k.sort()
    return k
89 90 91 92 93

class identifier:
    def __init__(self, name, header=None, module=None, type=None, lineno = 0,
                 info=None, extra=None, conditionals = None):
        self.name = name
94 95 96 97 98 99 100 101 102 103 104
        self.header = header
        self.module = module
        self.type = type
        self.info = info
        self.extra = extra
        self.lineno = lineno
        self.static = 0
        if conditionals == None or len(conditionals) == 0:
            self.conditionals = None
        else:
            self.conditionals = conditionals[:]
105
        if self.name == debugsym and not quiet:
106 107
            print "=> define %s : %s" % (debugsym, (module, type, info,
                                         extra, conditionals))
108 109 110

    def __repr__(self):
        r = "%s %s:" % (self.type, self.name)
111 112 113 114 115 116 117 118 119 120 121
        if self.static:
            r = r + " static"
        if self.module != None:
            r = r + " from %s" % (self.module)
        if self.info != None:
            r = r + " " +  `self.info`
        if self.extra != None:
            r = r + " " + `self.extra`
        if self.conditionals != None:
            r = r + " " + `self.conditionals`
        return r
122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138


    def set_header(self, header):
        self.header = header
    def set_module(self, module):
        self.module = module
    def set_type(self, type):
        self.type = type
    def set_info(self, info):
        self.info = info
    def set_extra(self, extra):
        self.extra = extra
    def set_lineno(self, lineno):
        self.lineno = lineno
    def set_static(self, static):
        self.static = static
    def set_conditionals(self, conditionals):
139 140 141 142
        if conditionals == None or len(conditionals) == 0:
            self.conditionals = None
        else:
            self.conditionals = conditionals[:]
143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164

    def get_name(self):
        return self.name
    def get_header(self):
        return self.module
    def get_module(self):
        return self.module
    def get_type(self):
        return self.type
    def get_info(self):
        return self.info
    def get_lineno(self):
        return self.lineno
    def get_extra(self):
        return self.extra
    def get_static(self):
        return self.static
    def get_conditionals(self):
        return self.conditionals

    def update(self, header, module, type = None, info = None, extra=None,
               conditionals=None):
165
        if self.name == debugsym and not quiet:
166 167
            print "=> update %s : %s" % (debugsym, (module, type, info,
                                         extra, conditionals))
168
        if header != None and self.header == None:
169
            self.set_header(module)
170
        if module != None and (self.module == None or self.header == self.module):
171
            self.set_module(module)
172
        if type != None and self.type == None:
173
            self.set_type(type)
174
        if info != None:
175
            self.set_info(info)
176
        if extra != None:
177
            self.set_extra(extra)
178
        if conditionals != None:
179
            self.set_conditionals(conditionals)
180 181 182 183 184 185

class index:
    def __init__(self, name = "noname"):
        self.name = name
        self.identifiers = {}
        self.functions = {}
186 187 188
        self.variables = {}
        self.includes = {}
        self.structs = {}
189
        self.unions = {}
190 191 192 193 194
        self.enums = {}
        self.typedefs = {}
        self.macros = {}
        self.references = {}
        self.info = {}
195 196 197

    def add_ref(self, name, header, module, static, type, lineno, info=None, extra=None, conditionals = None):
        if name[0:2] == '__':
198
            return None
199 200
        d = None
        try:
201 202 203 204 205
           d = self.identifiers[name]
           d.update(header, module, type, lineno, info, extra, conditionals)
        except:
           d = identifier(name, header, module, type, lineno, info, extra, conditionals)
           self.identifiers[name] = d
206

207 208
        if d != None and static == 1:
            d.set_static(1)
209

210 211
        if d != None and name != None and type != None:
            self.references[name] = d
212

213
        if name == debugsym and not quiet:
214
            print "New ref: %s" % (d)
215

216
        return d
217 218 219

    def add(self, name, header, module, static, type, lineno, info=None, extra=None, conditionals = None):
        if name[0:2] == '__':
220
            return None
221 222
        d = None
        try:
223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242
           d = self.identifiers[name]
           d.update(header, module, type, lineno, info, extra, conditionals)
        except:
           d = identifier(name, header, module, type, lineno, info, extra, conditionals)
           self.identifiers[name] = d

        if d != None and static == 1:
            d.set_static(1)

        if d != None and name != None and type != None:
            if type == "function":
                self.functions[name] = d
            elif type == "functype":
                self.functions[name] = d
            elif type == "variable":
                self.variables[name] = d
            elif type == "include":
                self.includes[name] = d
            elif type == "struct":
                self.structs[name] = d
243 244
            elif type == "union":
                self.unions[name] = d
245 246 247 248 249 250 251
            elif type == "enum":
                self.enums[name] = d
            elif type == "typedef":
                self.typedefs[name] = d
            elif type == "macro":
                self.macros[name] = d
            else:
252
                self.warning("Unable to register type ", type)
253

254
        if name == debugsym and not quiet:
255 256 257
            print "New symbol: %s" % (d)

        return d
258 259 260 261 262 263 264

    def merge(self, idx):
        for id in idx.functions.keys():
              #
              # macro might be used to override functions or variables
              # definitions
              #
265 266 267
             if self.macros.has_key(id):
                 del self.macros[id]
             if self.functions.has_key(id):
268 269
                 self.warning("function %s from %s redeclared in %s" % (
                    id, self.functions[id].header, idx.functions[id].header))
270 271 272
             else:
                 self.functions[id] = idx.functions[id]
                 self.identifiers[id] = idx.functions[id]
273 274 275 276 277
        for id in idx.variables.keys():
              #
              # macro might be used to override functions or variables
              # definitions
              #
278 279 280
             if self.macros.has_key(id):
                 del self.macros[id]
             if self.variables.has_key(id):
281 282
                 self.warning("variable %s from %s redeclared in %s" % (
                    id, self.variables[id].header, idx.variables[id].header))
283 284 285
             else:
                 self.variables[id] = idx.variables[id]
                 self.identifiers[id] = idx.variables[id]
286
        for id in idx.structs.keys():
287
             if self.structs.has_key(id):
288 289
                 self.warning("struct %s from %s redeclared in %s" % (
                    id, self.structs[id].header, idx.structs[id].header))
290 291 292
             else:
                 self.structs[id] = idx.structs[id]
                 self.identifiers[id] = idx.structs[id]
293 294 295 296 297 298 299
        for id in idx.unions.keys():
             if self.unions.has_key(id):
                 print "union %s from %s redeclared in %s" % (
                    id, self.unions[id].header, idx.unions[id].header)
             else:
                 self.unions[id] = idx.unions[id]
                 self.identifiers[id] = idx.unions[id]
300
        for id in idx.typedefs.keys():
301
             if self.typedefs.has_key(id):
302 303
                 self.warning("typedef %s from %s redeclared in %s" % (
                    id, self.typedefs[id].header, idx.typedefs[id].header))
304 305 306
             else:
                 self.typedefs[id] = idx.typedefs[id]
                 self.identifiers[id] = idx.typedefs[id]
307 308 309 310 311 312 313 314 315 316 317
        for id in idx.macros.keys():
              #
              # macro might be used to override functions or variables
              # definitions
              #
             if self.variables.has_key(id):
                 continue
             if self.functions.has_key(id):
                 continue
             if self.enums.has_key(id):
                 continue
318
             if self.macros.has_key(id):
319 320
                 self.warning("macro %s from %s redeclared in %s" % (
                    id, self.macros[id].header, idx.macros[id].header))
321 322 323
             else:
                 self.macros[id] = idx.macros[id]
                 self.identifiers[id] = idx.macros[id]
324
        for id in idx.enums.keys():
325
             if self.enums.has_key(id):
326 327
                 self.warning("enum %s from %s redeclared in %s" % (
                    id, self.enums[id].header, idx.enums[id].header))
328 329 330
             else:
                 self.enums[id] = idx.enums[id]
                 self.identifiers[id] = idx.enums[id]
331 332 333

    def merge_public(self, idx):
        for id in idx.functions.keys():
334 335 336 337
             if self.functions.has_key(id):
                 # check that function condition agrees with header
                 if idx.functions[id].conditionals != \
                    self.functions[id].conditionals:
338 339 340 341
                     self.warning("Header condition differs from Function for %s:" \
                                      % id)
                     self.warning("  H: %s" % self.functions[id].conditionals)
                     self.warning("  C: %s" % idx.functions[id].conditionals)
342 343 344 345 346 347
                 up = idx.functions[id]
                 self.functions[id].update(None, up.module, up.type, up.info, up.extra)
         #     else:
         #         print "Function %s from %s is not declared in headers" % (
         #              id, idx.functions[id].module)
         # TODO: do the same for variables.
348 349 350

    def analyze_dict(self, type, dict):
        count = 0
351
        public = 0
352
        for name in dict.keys():
353 354 355 356
            id = dict[name]
            count = count + 1
            if id.static == 0:
                public = public + 1
357
        if count != public:
358 359 360
            print "  %d %s , %d public" % (count, type, public)
        elif count != 0:
            print "  %d public %s" % (count, type)
361 362 363


    def analyze(self):
364 365 366 367 368 369 370
        if not quiet:
            self.analyze_dict("functions", self.functions)
            self.analyze_dict("variables", self.variables)
            self.analyze_dict("structs", self.structs)
            self.analyze_dict("unions", self.unions)
            self.analyze_dict("typedefs", self.typedefs)
            self.analyze_dict("macros", self.macros)
371

372 373 374 375 376
class CLexer:
    """A lexer for the C language, tokenize the input by reading and
       analyzing it line by line"""
    def __init__(self, input):
        self.input = input
377 378 379
        self.tokens = []
        self.line = ""
        self.lineno = 0
380 381 382

    def getline(self):
        line = ''
383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401
        while line == '':
            line = self.input.readline()
            if not line:
                return None
            self.lineno = self.lineno + 1
            line = string.lstrip(line)
            line = string.rstrip(line)
            if line == '':
                continue
            while line[-1] == '\\':
                line = line[:-1]
                n = self.input.readline()
                self.lineno = self.lineno + 1
                n = string.lstrip(n)
                n = string.rstrip(n)
                if not n:
                    break
                else:
                    line = line + n
402
        return line
403

404 405 406 407 408 409 410 411
    def getlineno(self):
        return self.lineno

    def push(self, token):
        self.tokens.insert(0, token);

    def debug(self):
        print "Last token: ", self.last
412 413
        print "Token queue: ", self.tokens
        print "Line %d end: " % (self.lineno), self.line
414 415 416

    def token(self):
        while self.tokens == []:
417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519
            if self.line == "":
                line = self.getline()
            else:
                line = self.line
                self.line = ""
            if line == None:
                return None

            if line[0] == '#':
                self.tokens = map((lambda x: ('preproc', x)),
                                  string.split(line))
                break;
            l = len(line)
            if line[0] == '"' or line[0] == "'":
                end = line[0]
                line = line[1:]
                found = 0
                tok = ""
                while found == 0:
                    i = 0
                    l = len(line)
                    while i < l:
                        if line[i] == end:
                            self.line = line[i+1:]
                            line = line[:i]
                            l = i
                            found = 1
                            break
                        if line[i] == '\\':
                            i = i + 1
                        i = i + 1
                    tok = tok + line
                    if found == 0:
                        line = self.getline()
                        if line == None:
                            return None
                self.last = ('string', tok)
                return self.last

            if l >= 2 and line[0] == '/' and line[1] == '*':
                line = line[2:]
                found = 0
                tok = ""
                while found == 0:
                    i = 0
                    l = len(line)
                    while i < l:
                        if line[i] == '*' and i+1 < l and line[i+1] == '/':
                            self.line = line[i+2:]
                            line = line[:i-1]
                            l = i
                            found = 1
                            break
                        i = i + 1
                    if tok != "":
                        tok = tok + "\n"
                    tok = tok + line
                    if found == 0:
                        line = self.getline()
                        if line == None:
                            return None
                self.last = ('comment', tok)
                return self.last
            if l >= 2 and line[0] == '/' and line[1] == '/':
                line = line[2:]
                self.last = ('comment', line)
                return self.last
            i = 0
            while i < l:
                if line[i] == '/' and i+1 < l and line[i+1] == '/':
                    self.line = line[i:]
                    line = line[:i]
                    break
                if line[i] == '/' and i+1 < l and line[i+1] == '*':
                    self.line = line[i:]
                    line = line[:i]
                    break
                if line[i] == '"' or line[i] == "'":
                    self.line = line[i:]
                    line = line[:i]
                    break
                i = i + 1
            l = len(line)
            i = 0
            while i < l:
                if line[i] == ' ' or line[i] == '\t':
                    i = i + 1
                    continue
                o = ord(line[i])
                if (o >= 97 and o <= 122) or (o >= 65 and o <= 90) or \
                   (o >= 48 and o <= 57):
                    s = i
                    while i < l:
                        o = ord(line[i])
                        if (o >= 97 and o <= 122) or (o >= 65 and o <= 90) or \
                           (o >= 48 and o <= 57) or string.find(
                               " \t(){}:;,+-*/%&!|[]=><", line[i]) == -1:
                            i = i + 1
                        else:
                            break
                    self.tokens.append(('name', line[s:i]))
                    continue
                if string.find("(){}:;,[]", line[i]) != -1:
520
#                 if line[i] == '(' or line[i] == ')' or line[i] == '{' or \
521 522 523 524 525 526
#                   line[i] == '}' or line[i] == ':' or line[i] == ';' or \
#                   line[i] == ',' or line[i] == '[' or line[i] == ']':
                    self.tokens.append(('sep', line[i]))
                    i = i + 1
                    continue
                if string.find("+-*><=/%&!|.", line[i]) != -1:
527
#                 if line[i] == '+' or line[i] == '-' or line[i] == '*' or \
528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576
#                   line[i] == '>' or line[i] == '<' or line[i] == '=' or \
#                   line[i] == '/' or line[i] == '%' or line[i] == '&' or \
#                   line[i] == '!' or line[i] == '|' or line[i] == '.':
                    if line[i] == '.' and  i + 2 < l and \
                       line[i+1] == '.' and line[i+2] == '.':
                        self.tokens.append(('name', '...'))
                        i = i + 3
                        continue

                    j = i + 1
                    if j < l and (
                       string.find("+-*><=/%&!|", line[j]) != -1):
#                       line[j] == '+' or line[j] == '-' or line[j] == '*' or \
#                       line[j] == '>' or line[j] == '<' or line[j] == '=' or \
#                       line[j] == '/' or line[j] == '%' or line[j] == '&' or \
#                       line[j] == '!' or line[j] == '|'):
                        self.tokens.append(('op', line[i:j+1]))
                        i = j + 1
                    else:
                        self.tokens.append(('op', line[i]))
                        i = i + 1
                    continue
                s = i
                while i < l:
                    o = ord(line[i])
                    if (o >= 97 and o <= 122) or (o >= 65 and o <= 90) or \
                       (o >= 48 and o <= 57) or (
                        string.find(" \t(){}:;,+-*/%&!|[]=><", line[i]) == -1):
#                        line[i] != ' ' and line[i] != '\t' and
#                        line[i] != '(' and line[i] != ')' and
#                        line[i] != '{'  and line[i] != '}' and
#                        line[i] != ':' and line[i] != ';' and
#                        line[i] != ',' and line[i] != '+' and
#                        line[i] != '-' and line[i] != '*' and
#                        line[i] != '/' and line[i] != '%' and
#                        line[i] != '&' and line[i] != '!' and
#                        line[i] != '|' and line[i] != '[' and
#                        line[i] != ']' and line[i] != '=' and
#                        line[i] != '*' and line[i] != '>' and
#                        line[i] != '<'):
                        i = i + 1
                    else:
                        break
                self.tokens.append(('name', line[s:i]))

        tok = self.tokens[0]
        self.tokens = self.tokens[1:]
        self.last = tok
        return tok
577

578 579 580 581
class CParser:
    """The C module parser"""
    def __init__(self, filename, idx = None):
        self.filename = filename
582 583 584 585
        if len(filename) > 2 and filename[-2:] == '.h':
            self.is_header = 1
        else:
            self.is_header = 0
586
        self.input = open(filename)
587 588 589 590 591 592 593 594 595 596 597 598
        self.lexer = CLexer(self.input)
        if idx == None:
            self.index = index()
        else:
            self.index = idx
        self.top_comment = ""
        self.last_comment = ""
        self.comment = None
        self.collect_ref = 0
        self.no_error = 0
        self.conditionals = []
        self.defines = []
599 600 601 602 603 604 605 606 607 608 609 610 611 612

    def collect_references(self):
        self.collect_ref = 1

    def stop_error(self):
        self.no_error = 1

    def start_error(self):
        self.no_error = 0

    def lineno(self):
        return self.lexer.getlineno()

    def index_add(self, name, module, static, type, info=None, extra = None):
613 614 615 616 617 618
        if self.is_header == 1:
            self.index.add(name, module, module, static, type, self.lineno(),
                           info, extra, self.conditionals)
        else:
            self.index.add(name, None, module, static, type, self.lineno(),
                           info, extra, self.conditionals)
619 620 621

    def index_add_ref(self, name, module, static, type, info=None,
                      extra = None):
622 623 624 625 626 627
        if self.is_header == 1:
            self.index.add_ref(name, module, module, static, type,
                               self.lineno(), info, extra, self.conditionals)
        else:
            self.index.add_ref(name, None, module, static, type, self.lineno(),
                               info, extra, self.conditionals)
628 629

    def warning(self, msg):
630 631
        global warnings
        warnings = warnings + 1
632
        if self.no_error:
633 634
            return
        print msg
635 636 637

    def error(self, msg, token=-1):
        if self.no_error:
638
            return
639 640

        print "Parse Error: " + msg
641 642 643 644
        if token != -1:
            print "Got token ", token
        self.lexer.debug()
        sys.exit(1)
645 646 647

    def debug(self, msg, token=-1):
        print "Debug: " + msg
648 649 650
        if token != -1:
            print "Got token ", token
        self.lexer.debug()
651 652

    def parseTopComment(self, comment):
653 654 655 656
        res = {}
        lines = string.split(comment, "\n")
        item = None
        for line in lines:
C
Claudio Bley 已提交
657
            line = line.lstrip().lstrip('*').lstrip()
658 659 660
            try:
                (it, line) = string.split(line, ":", 1)
                item = it
C
Claudio Bley 已提交
661
                line = line.lstrip()
662 663 664 665 666 667 668 669 670 671 672
                if res.has_key(item):
                    res[item] = res[item] + " " + line
                else:
                    res[item] = line
            except:
                if item != None:
                    if res.has_key(item):
                        res[item] = res[item] + " " + line
                    else:
                        res[item] = line
        self.index.info = res
673

674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695
    def strip_lead_star(self, line):
        l = len(line)
        i = 0
        while i < l:
            if line[i] == ' ' or line[i] == '\t':
                i += 1
            elif line[i] == '*':
                return line[:i] + line[i + 1:]
            else:
                 return line
        return line

    def cleanupComment(self):
        if type(self.comment) != type(""):
            return
        # remove the leading * on multi-line comments
        lines = self.comment.splitlines(True)
        com = ""
        for line in lines:
            com = com + self.strip_lead_star(line)
        self.comment = com.strip()

696
    def parseComment(self, token):
697
        com = token[1]
698
        if self.top_comment == "":
699 700 701
            self.top_comment = com
        if self.comment == None or com[0] == '*':
            self.comment = com;
702
        else:
703
            self.comment = self.comment + com
704
        token = self.lexer.token()
705 706

        if string.find(self.comment, "DOC_DISABLE") != -1:
707
            self.stop_error()
708 709

        if string.find(self.comment, "DOC_ENABLE") != -1:
710
            self.start_error()
711

712
        return token
713 714 715 716 717 718

    #
    # Parse a comment block associate to a typedef
    #
    def parseTypeComment(self, name, quiet = 0):
        if name[0:2] == '__':
719
            quiet = 1
720 721

        args = []
722
        desc = ""
723 724

        if self.comment == None:
725 726 727
            if not quiet:
                self.warning("Missing comment for type %s" % (name))
            return((args, desc))
728
        if self.comment[0] != '*':
729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758
            if not quiet:
                self.warning("Missing * in type comment for %s" % (name))
            return((args, desc))
        lines = string.split(self.comment, '\n')
        if lines[0] == '*':
            del lines[0]
        if lines[0] != "* %s:" % (name):
            if not quiet:
                self.warning("Misformatted type comment for %s" % (name))
                self.warning("  Expecting '* %s:' got '%s'" % (name, lines[0]))
            return((args, desc))
        del lines[0]
        while len(lines) > 0 and lines[0] == '*':
            del lines[0]
        desc = ""
        while len(lines) > 0:
            l = lines[0]
            while len(l) > 0 and l[0] == '*':
                l = l[1:]
            l = string.strip(l)
            desc = desc + " " + l
            del lines[0]

        desc = string.strip(desc)

        if quiet == 0:
            if desc == "":
                self.warning("Type comment for %s lack description of the macro" % (name))

        return(desc)
759 760 761 762
    #
    # Parse a comment block associate to a macro
    #
    def parseMacroComment(self, name, quiet = 0):
763 764
        global ignored_macros

765
        if name[0:2] == '__':
766
            quiet = 1
767 768
        if ignored_macros.has_key(name):
            quiet = 1
769 770

        args = []
771
        desc = ""
772 773

        if self.comment == None:
774 775 776
            if not quiet:
                self.warning("Missing comment for macro %s" % (name))
            return((args, desc))
777
        if self.comment[0] != '*':
778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797
            if not quiet:
                self.warning("Missing * in macro comment for %s" % (name))
            return((args, desc))
        lines = string.split(self.comment, '\n')
        if lines[0] == '*':
            del lines[0]
        if lines[0] != "* %s:" % (name):
            if not quiet:
                self.warning("Misformatted macro comment for %s" % (name))
                self.warning("  Expecting '* %s:' got '%s'" % (name, lines[0]))
            return((args, desc))
        del lines[0]
        while lines[0] == '*':
            del lines[0]
        while len(lines) > 0 and lines[0][0:3] == '* @':
            l = lines[0][3:]
            try:
                (arg, desc) = string.split(l, ':', 1)
                desc=string.strip(desc)
                arg=string.strip(arg)
798
            except:
799 800 801 802 803 804 805 806 807 808 809 810 811 812 813
                if not quiet:
                    self.warning("Misformatted macro comment for %s" % (name))
                    self.warning("  problem with '%s'" % (lines[0]))
                del lines[0]
                continue
            del lines[0]
            l = string.strip(lines[0])
            while len(l) > 2 and l[0:3] != '* @':
                while l[0] == '*':
                    l = l[1:]
                desc = desc + ' ' + string.strip(l)
                del lines[0]
                if len(lines) == 0:
                    break
                l = lines[0]
814
            args.append((arg, desc))
815 816 817 818 819 820 821 822 823 824
        while len(lines) > 0 and lines[0] == '*':
            del lines[0]
        desc = ""
        while len(lines) > 0:
            l = lines[0]
            while len(l) > 0 and l[0] == '*':
                l = l[1:]
            l = string.strip(l)
            desc = desc + " " + l
            del lines[0]
825

826
        desc = string.strip(desc)
827

828 829 830
        if quiet == 0:
            if desc == "":
                self.warning("Macro comment for %s lack description of the macro" % (name))
831

832
        return((args, desc))
833 834

     #
835
     # Parse a comment block and merge the information found in the
836 837 838 839
     # parameters descriptions, finally returns a block as complete
     # as possible
     #
    def mergeFunctionComment(self, name, description, quiet = 0):
D
Daniel Veillard 已提交
840 841
        global ignored_functions

842
        if name == 'main':
843
            quiet = 1
844
        if name[0:2] == '__':
845
            quiet = 1
D
Daniel Veillard 已提交
846 847
        if ignored_functions.has_key(name):
            quiet = 1
848

849 850 851
        (ret, args) = description
        desc = ""
        retdesc = ""
852 853

        if self.comment == None:
854 855 856
            if not quiet:
                self.warning("Missing comment for function %s" % (name))
            return(((ret[0], retdesc), args, desc))
857
        if self.comment[0] != '*':
858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878
            if not quiet:
                self.warning("Missing * in function comment for %s" % (name))
            return(((ret[0], retdesc), args, desc))
        lines = string.split(self.comment, '\n')
        if lines[0] == '*':
            del lines[0]
        if lines[0] != "* %s:" % (name):
            if not quiet:
                self.warning("Misformatted function comment for %s" % (name))
                self.warning("  Expecting '* %s:' got '%s'" % (name, lines[0]))
            return(((ret[0], retdesc), args, desc))
        del lines[0]
        while lines[0] == '*':
            del lines[0]
        nbargs = len(args)
        while len(lines) > 0 and lines[0][0:3] == '* @':
            l = lines[0][3:]
            try:
                (arg, desc) = string.split(l, ':', 1)
                desc=string.strip(desc)
                arg=string.strip(arg)
879
            except:
880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960
                if not quiet:
                    self.warning("Misformatted function comment for %s" % (name))
                    self.warning("  problem with '%s'" % (lines[0]))
                del lines[0]
                continue
            del lines[0]
            l = string.strip(lines[0])
            while len(l) > 2 and l[0:3] != '* @':
                while l[0] == '*':
                    l = l[1:]
                desc = desc + ' ' + string.strip(l)
                del lines[0]
                if len(lines) == 0:
                    break
                l = lines[0]
            i = 0
            while i < nbargs:
                if args[i][1] == arg:
                    args[i] = (args[i][0], arg, desc)
                    break;
                i = i + 1
            if i >= nbargs:
                if not quiet:
                    self.warning("Unable to find arg %s from function comment for %s" % (
                       arg, name))
        while len(lines) > 0 and lines[0] == '*':
            del lines[0]
        desc = None
        while len(lines) > 0:
            l = lines[0]
            i = 0
            # Remove all leading '*', followed by at most one ' ' character
            # since we need to preserve correct identation of code examples
            while i < len(l) and l[i] == '*':
                i = i + 1
            if i > 0:
                if i < len(l) and l[i] == ' ':
                    i = i + 1
                l = l[i:]
            if len(l) >= 6 and  l[0:7] == "returns" or l[0:7] == "Returns":
                try:
                    l = string.split(l, ' ', 1)[1]
                except:
                    l = ""
                retdesc = string.strip(l)
                del lines[0]
                while len(lines) > 0:
                    l = lines[0]
                    while len(l) > 0 and l[0] == '*':
                        l = l[1:]
                    l = string.strip(l)
                    retdesc = retdesc + " " + l
                    del lines[0]
            else:
                if desc is not None:
                    desc = desc + "\n" + l
                else:
                    desc = l
                del lines[0]

        if desc is None:
            desc = ""
        retdesc = string.strip(retdesc)
        desc = string.strip(desc)

        if quiet == 0:
             #
             # report missing comments
             #
            i = 0
            while i < nbargs:
                if args[i][2] == None and args[i][0] != "void" and args[i][1] != None:
                    self.warning("Function comment for %s lacks description of arg %s" % (name, args[i][1]))
                i = i + 1
            if retdesc == "" and ret[0] != "void":
                self.warning("Function comment for %s lacks description of return value" % (name))
            if desc == "":
                self.warning("Function comment for %s lacks description of the function" % (name))


        return(((ret[0], retdesc), args, desc))
961 962

    def parsePreproc(self, token):
963 964
        if debug:
            print "=> preproc ", token, self.lexer.tokens
965
        name = token[1]
966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987
        if name == "#include":
            token = self.lexer.token()
            if token == None:
                return None
            if token[0] == 'preproc':
                self.index_add(token[1], self.filename, not self.is_header,
                                "include")
                return self.lexer.token()
            return token
        if name == "#define":
            token = self.lexer.token()
            if token == None:
                return None
            if token[0] == 'preproc':
                 # TODO macros with arguments
                name = token[1]
                lst = []
                token = self.lexer.token()
                while token != None and token[0] == 'preproc' and \
                      token[1][0] != '#':
                    lst.append(token[1])
                    token = self.lexer.token()
988
                try:
989
                    name = string.split(name, '(') [0]
990 991 992
                except:
                    pass
                info = self.parseMacroComment(name, not self.is_header)
993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055
                self.index_add(name, self.filename, not self.is_header,
                                "macro", info)
                return token

        #
        # Processing of conditionals modified by Bill 1/1/05
        #
        # We process conditionals (i.e. tokens from #ifdef, #ifndef,
        # #if, #else and #endif) for headers and mainline code,
        # store the ones from the header in libxml2-api.xml, and later
        # (in the routine merge_public) verify that the two (header and
        # mainline code) agree.
        #
        # There is a small problem with processing the headers. Some of
        # the variables are not concerned with enabling / disabling of
        # library functions (e.g. '__XML_PARSER_H__'), and we don't want
        # them to be included in libxml2-api.xml, or involved in
        # the check between the header and the mainline code.  To
        # accomplish this, we ignore any conditional which doesn't include
        # the string 'ENABLED'
        #
        if name == "#ifdef":
            apstr = self.lexer.tokens[0][1]
            try:
                self.defines.append(apstr)
                if string.find(apstr, 'ENABLED') != -1:
                    self.conditionals.append("defined(%s)" % apstr)
            except:
                pass
        elif name == "#ifndef":
            apstr = self.lexer.tokens[0][1]
            try:
                self.defines.append(apstr)
                if string.find(apstr, 'ENABLED') != -1:
                    self.conditionals.append("!defined(%s)" % apstr)
            except:
                pass
        elif name == "#if":
            apstr = ""
            for tok in self.lexer.tokens:
                if apstr != "":
                    apstr = apstr + " "
                apstr = apstr + tok[1]
            try:
                self.defines.append(apstr)
                if string.find(apstr, 'ENABLED') != -1:
                    self.conditionals.append(apstr)
            except:
                pass
        elif name == "#else":
            if self.conditionals != [] and \
               string.find(self.defines[-1], 'ENABLED') != -1:
                self.conditionals[-1] = "!(%s)" % self.conditionals[-1]
        elif name == "#endif":
            if self.conditionals != [] and \
               string.find(self.defines[-1], 'ENABLED') != -1:
                self.conditionals = self.conditionals[:-1]
            self.defines = self.defines[:-1]
        token = self.lexer.token()
        while token != None and token[0] == 'preproc' and \
            token[1][0] != '#':
            token = self.lexer.token()
        return token
1056 1057 1058 1059 1060 1061

     #
     # token acquisition on top of the lexer, it handle internally
     # preprocessor and comments since they are logically not part of
     # the program structure.
     #
1062 1063 1064
    def push(self, tok):
        self.lexer.push(tok)

1065 1066 1067 1068
    def token(self):
        global ignored_words

        token = self.lexer.token()
1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096
        while token != None:
            if token[0] == 'comment':
                token = self.parseComment(token)
                continue
            elif token[0] == 'preproc':
                token = self.parsePreproc(token)
                continue
            elif token[0] == "name" and token[1] == "__const":
                token = ("name", "const")
                return token
            elif token[0] == "name" and token[1] == "__attribute":
                token = self.lexer.token()
                while token != None and token[1] != ";":
                    token = self.lexer.token()
                return token
            elif token[0] == "name" and ignored_words.has_key(token[1]):
                (n, info) = ignored_words[token[1]]
                i = 0
                while i < n:
                    token = self.lexer.token()
                    i = i + 1
                token = self.lexer.token()
                continue
            else:
                if debug:
                    print "=> ", token
                return token
        return None
1097 1098 1099 1100 1101 1102

     #
     # Parse a typedef, it records the type and its name.
     #
    def parseTypedef(self, token):
        if token == None:
1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151
            return None
        token = self.parseType(token)
        if token == None:
            self.error("parsing typedef")
            return None
        base_type = self.type
        type = base_type
         #self.debug("end typedef type", token)
        while token != None:
            if token[0] == "name":
                name = token[1]
                signature = self.signature
                if signature != None:
                    type = string.split(type, '(')[0]
                    d = self.mergeFunctionComment(name,
                            ((type, None), signature), 1)
                    self.index_add(name, self.filename, not self.is_header,
                                    "functype", d)
                else:
                    if base_type == "struct":
                        self.index_add(name, self.filename, not self.is_header,
                                        "struct", type)
                        base_type = "struct " + name
                    else:
                        # TODO report missing or misformatted comments
                        info = self.parseTypeComment(name, 1)
                        self.index_add(name, self.filename, not self.is_header,
                                    "typedef", type, info)
                token = self.token()
            else:
                self.error("parsing typedef: expecting a name")
                return token
             #self.debug("end typedef", token)
            if token != None and token[0] == 'sep' and token[1] == ',':
                type = base_type
                token = self.token()
                while token != None and token[0] == "op":
                    type = type + token[1]
                    token = self.token()
            elif token != None and token[0] == 'sep' and token[1] == ';':
                break;
            elif token != None and token[0] == 'name':
                type = base_type
                continue;
            else:
                self.error("parsing typedef: expecting ';'", token)
                return token
        token = self.token()
        return token
1152

1153 1154 1155 1156 1157 1158
     #
     # Parse a C code block, used for functions it parse till
     # the balancing } included
     #
    def parseBlock(self, token):
        while token != None:
1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190
            if token[0] == "sep" and token[1] == "{":
                token = self.token()
                token = self.parseBlock(token)
            elif token[0] == "sep" and token[1] == "}":
                self.comment = None
                token = self.token()
                return token
            else:
                if self.collect_ref == 1:
                    oldtok = token
                    token = self.token()
                    if oldtok[0] == "name" and oldtok[1][0:3] == "vir":
                        if token[0] == "sep" and token[1] == "(":
                            self.index_add_ref(oldtok[1], self.filename,
                                                0, "function")
                            token = self.token()
                        elif token[0] == "name":
                            token = self.token()
                            if token[0] == "sep" and (token[1] == ";" or
                               token[1] == "," or token[1] == "="):
                                self.index_add_ref(oldtok[1], self.filename,
                                                    0, "type")
                    elif oldtok[0] == "name" and oldtok[1][0:4] == "XEN_":
                        self.index_add_ref(oldtok[1], self.filename,
                                            0, "typedef")
                    elif oldtok[0] == "name" and oldtok[1][0:7] == "LIBXEN_":
                        self.index_add_ref(oldtok[1], self.filename,
                                            0, "typedef")

                else:
                    token = self.token()
        return token
1191 1192 1193 1194 1195 1196

     #
     # Parse a C struct definition till the balancing }
     #
    def parseStruct(self, token):
        fields = []
1197
         #self.debug("start parseStruct", token)
1198
        while token != None:
1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218
            if token[0] == "sep" and token[1] == "{":
                token = self.token()
                token = self.parseTypeBlock(token)
            elif token[0] == "sep" and token[1] == "}":
                self.struct_fields = fields
                 #self.debug("end parseStruct", token)
                 #print fields
                token = self.token()
                return token
            else:
                base_type = self.type
                 #self.debug("before parseType", token)
                token = self.parseType(token)
                 #self.debug("after parseType", token)
                if token != None and token[0] == "name":
                    fname = token[1]
                    token = self.token()
                    if token[0] == "sep" and token[1] == ";":
                        self.comment = None
                        token = self.token()
1219 1220 1221 1222 1223 1224 1225
                        self.cleanupComment()
                        if self.type == "union":
                            fields.append((self.type, fname, self.comment,
                                           self.union_fields))
                            self.union_fields = []
                        else:
                            fields.append((self.type, fname, self.comment))
1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241
                        self.comment = None
                    else:
                        self.error("parseStruct: expecting ;", token)
                elif token != None and token[0] == "sep" and token[1] == "{":
                    token = self.token()
                    token = self.parseTypeBlock(token)
                    if token != None and token[0] == "name":
                        token = self.token()
                    if token != None and token[0] == "sep" and token[1] == ";":
                        token = self.token()
                    else:
                        self.error("parseStruct: expecting ;", token)
                else:
                    self.error("parseStruct: name", token)
                    token = self.token()
                self.type = base_type;
1242
        self.struct_fields = fields
1243 1244 1245
         #self.debug("end parseStruct", token)
         #print fields
        return token
1246

1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296
     #
     # Parse a C union definition till the balancing }
     #
    def parseUnion(self, token):
        fields = []
        # self.debug("start parseUnion", token)
        while token != None:
            if token[0] == "sep" and token[1] == "{":
                token = self.token()
                token = self.parseTypeBlock(token)
            elif token[0] == "sep" and token[1] == "}":
                self.union_fields = fields
                # self.debug("end parseUnion", token)
                # print fields
                token = self.token()
                return token
            else:
                base_type = self.type
                # self.debug("before parseType", token)
                token = self.parseType(token)
                # self.debug("after parseType", token)
                if token != None and token[0] == "name":
                    fname = token[1]
                    token = self.token()
                    if token[0] == "sep" and token[1] == ";":
                        self.comment = None
                        token = self.token()
                        self.cleanupComment()
                        fields.append((self.type, fname, self.comment))
                        self.comment = None
                    else:
                        self.error("parseUnion: expecting ;", token)
                elif token != None and token[0] == "sep" and token[1] == "{":
                    token = self.token()
                    token = self.parseTypeBlock(token)
                    if token != None and token[0] == "name":
                        token = self.token()
                    if token != None and token[0] == "sep" and token[1] == ";":
                        token = self.token()
                    else:
                        self.error("parseUnion: expecting ;", token)
                else:
                    self.error("parseUnion: name", token)
                    token = self.token()
                self.type = base_type;
        self.union_fields = fields
        # self.debug("end parseUnion", token)
        # print fields
        return token

1297 1298 1299 1300 1301
     #
     # Parse a C enum block, parse till the balancing }
     #
    def parseEnumBlock(self, token):
        self.enums = []
1302 1303 1304 1305
        name = None
        self.comment = None
        comment = ""
        value = "0"
1306
        while token != None:
1307 1308 1309 1310 1311
            if token[0] == "sep" and token[1] == "{":
                token = self.token()
                token = self.parseTypeBlock(token)
            elif token[0] == "sep" and token[1] == "}":
                if name != None:
1312
                    self.cleanupComment()
1313 1314 1315 1316 1317 1318 1319
                    if self.comment != None:
                        comment = self.comment
                        self.comment = None
                    self.enums.append((name, value, comment))
                token = self.token()
                return token
            elif token[0] == "name":
1320
                    self.cleanupComment()
1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348
                    if name != None:
                        if self.comment != None:
                            comment = string.strip(self.comment)
                            self.comment = None
                        self.enums.append((name, value, comment))
                    name = token[1]
                    comment = ""
                    token = self.token()
                    if token[0] == "op" and token[1][0] == "=":
                        value = ""
                        if len(token[1]) > 1:
                            value = token[1][1:]
                        token = self.token()
                        while token[0] != "sep" or (token[1] != ',' and
                              token[1] != '}'):
                            value = value + token[1]
                            token = self.token()
                    else:
                        try:
                            value = "%d" % (int(value) + 1)
                        except:
                            self.warning("Failed to compute value of enum %s" % (name))
                            value=""
                    if token[0] == "sep" and token[1] == ",":
                        token = self.token()
            else:
                token = self.token()
        return token
1349

1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438
    def parseVirEnumDecl(self, token):
        if token[0] != "name":
            self.error("parsing VIR_ENUM_DECL: expecting name", token)

        token = self.token()

        if token[0] != "sep":
            self.error("parsing VIR_ENUM_DECL: expecting ')'", token)

        if token[1] != ')':
            self.error("parsing VIR_ENUM_DECL: expecting ')'", token)

        token = self.token()
        if token[0] == "sep" and token[1] == ';':
            token = self.token()

        return token

    def parseVirEnumImpl(self, token):
        # First the type name
        if token[0] != "name":
            self.error("parsing VIR_ENUM_IMPL: expecting name", token)

        token = self.token()

        if token[0] != "sep":
            self.error("parsing VIR_ENUM_IMPL: expecting ','", token)

        if token[1] != ',':
            self.error("parsing VIR_ENUM_IMPL: expecting ','", token)
        token = self.token()

        # Now the sentinel name
        if token[0] != "name":
            self.error("parsing VIR_ENUM_IMPL: expecting name", token)

        token = self.token()

        if token[0] != "sep":
            self.error("parsing VIR_ENUM_IMPL: expecting ','", token)

        if token[1] != ',':
            self.error("parsing VIR_ENUM_IMPL: expecting ','", token)

        token = self.token()

        # Now a list of strings (optional comments)
        while token is not None:
            isGettext = False
            # First a string, optionally with N_(...)
            if token[0] == 'name':
                if token[1] != 'N_':
                    self.error("parsing VIR_ENUM_IMPL: expecting 'N_'", token)
                token = self.token()
                if token[0] != "sep" or token[1] != '(':
                    self.error("parsing VIR_ENUM_IMPL: expecting '('", token)
                token = self.token()
                isGettext = True

                if token[0] != "string":
                    self.error("parsing VIR_ENUM_IMPL: expecting a string", token)
                token = self.token()
            elif token[0] == "string":
                token = self.token()
            else:
                self.error("parsing VIR_ENUM_IMPL: expecting a string", token)

            # Then a separator
            if token[0] == "sep":
                if isGettext and token[1] == ')':
                    token = self.token()

                if token[1] == ',':
                    token = self.token()

                if token[1] == ')':
                    token = self.token()
                    break

            # Then an optional comment
            if token[0] == "comment":
                token = self.token()


        if token[0] == "sep" and token[1] == ';':
            token = self.token()

        return token

1439
     #
1440
     # Parse a C definition block, used for structs or unions it parse till
1441 1442 1443 1444
     # the balancing }
     #
    def parseTypeBlock(self, token):
        while token != None:
1445 1446 1447 1448 1449 1450 1451 1452 1453
            if token[0] == "sep" and token[1] == "{":
                token = self.token()
                token = self.parseTypeBlock(token)
            elif token[0] == "sep" and token[1] == "}":
                token = self.token()
                return token
            else:
                token = self.token()
        return token
1454 1455 1456 1457 1458 1459 1460 1461

     #
     # Parse a type: the fact that the type name can either occur after
     #    the definition or within the definition makes it a little harder
     #    if inside, the name token is pushed back before returning
     #
    def parseType(self, token):
        self.type = ""
1462
        self.struct_fields = []
1463
        self.union_fields = []
1464
        self.signature = None
1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476
        if token == None:
            return token

        while token[0] == "name" and (
              token[1] == "const" or \
              token[1] == "unsigned" or \
              token[1] == "signed"):
            if self.type == "":
                self.type = token[1]
            else:
                self.type = self.type + " " + token[1]
            token = self.token()
1477

1478
        if token[0] == "name" and token[1] == "long":
1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492
            if self.type == "":
                self.type = token[1]
            else:
                self.type = self.type + " " + token[1]

            # some read ahead for long long
            oldtmp = token
            token = self.token()
            if token[0] == "name" and token[1] == "long":
                self.type = self.type + " " + token[1]
            else:
                self.push(token)
                token = oldtmp

1493 1494
            oldtmp = token
            token = self.token()
1495
            if token[0] == "name" and token[1] == "int":
1496 1497 1498 1499
                self.type = self.type + " " + token[1]
            else:
                self.push(token)
                token = oldtmp
1500 1501

        elif token[0] == "name" and token[1] == "short":
1502 1503 1504 1505
            if self.type == "":
                self.type = token[1]
            else:
                self.type = self.type + " " + token[1]
1506

1507
        elif token[0] == "name" and token[1] == "struct":
1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539
            if self.type == "":
                self.type = token[1]
            else:
                self.type = self.type + " " + token[1]
            token = self.token()
            nametok = None
            if token[0] == "name":
                nametok = token
                token = self.token()
            if token != None and token[0] == "sep" and token[1] == "{":
                token = self.token()
                token = self.parseStruct(token)
            elif token != None and token[0] == "op" and token[1] == "*":
                self.type = self.type + " " + nametok[1] + " *"
                token = self.token()
                while token != None and token[0] == "op" and token[1] == "*":
                    self.type = self.type + " *"
                    token = self.token()
                if token[0] == "name":
                    nametok = token
                    token = self.token()
                else:
                    self.error("struct : expecting name", token)
                    return token
            elif token != None and token[0] == "name" and nametok != None:
                self.type = self.type + " " + nametok[1]
                return token

            if nametok != None:
                self.lexer.push(token)
                token = nametok
            return token
1540

1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562
        elif token[0] == "name" and token[1] == "union":
            if self.type == "":
                self.type = token[1]
            else:
                self.type = self.type + " " + token[1]
            token = self.token()
            nametok = None
            if token[0] == "name":
                nametok = token
                token = self.token()
            if token != None and token[0] == "sep" and token[1] == "{":
                token = self.token()
                token = self.parseUnion(token)
            elif token != None and token[0] == "name" and nametok != None:
                self.type = self.type + " " + nametok[1]
                return token

            if nametok != None:
                self.lexer.push(token)
                token = nametok
            return token

1563
        elif token[0] == "name" and token[1] == "enum":
1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585
            if self.type == "":
                self.type = token[1]
            else:
                self.type = self.type + " " + token[1]
            self.enums = []
            token = self.token()
            if token != None and token[0] == "sep" and token[1] == "{":
                token = self.token()
                token = self.parseEnumBlock(token)
            else:
                self.error("parsing enum: expecting '{'", token)
            enum_type = None
            if token != None and token[0] != "name":
                self.lexer.push(token)
                token = ("name", "enum")
            else:
                enum_type = token[1]
            for enum in self.enums:
                self.index_add(enum[0], self.filename,
                               not self.is_header, "enum",
                               (enum[1], enum[2], enum_type))
            return token
1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608
        elif token[0] == "name" and token[1] == "VIR_ENUM_DECL":
            token = self.token()
            if token != None and token[0] == "sep" and token[1] == "(":
                token = self.token()
                token = self.parseVirEnumDecl(token)
            else:
                self.error("parsing VIR_ENUM_DECL: expecting '('", token)
            if token != None:
                self.lexer.push(token)
                token = ("name", "virenumdecl")
            return token

        elif token[0] == "name" and token[1] == "VIR_ENUM_IMPL":
            token = self.token()
            if token != None and token[0] == "sep" and token[1] == "(":
                token = self.token()
                token = self.parseVirEnumImpl(token)
            else:
                self.error("parsing VIR_ENUM_IMPL: expecting '('", token)
            if token != None:
                self.lexer.push(token)
                token = ("name", "virenumimpl")
            return token
1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619

        elif token[0] == "name":
            if self.type == "":
                self.type = token[1]
            else:
                self.type = self.type + " " + token[1]
        else:
            self.error("parsing type %s: expecting a name" % (self.type),
                       token)
            return token
        token = self.token()
1620
        while token != None and (token[0] == "op" or
1621 1622 1623
              token[0] == "name" and token[1] == "const"):
            self.type = self.type + " " + token[1]
            token = self.token()
1624 1625

         #
1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664
         # if there is a parenthesis here, this means a function type
         #
        if token != None and token[0] == "sep" and token[1] == '(':
            self.type = self.type + token[1]
            token = self.token()
            while token != None and token[0] == "op" and token[1] == '*':
                self.type = self.type + token[1]
                token = self.token()
            if token == None or token[0] != "name" :
                self.error("parsing function type, name expected", token);
                return token
            self.type = self.type + token[1]
            nametok = token
            token = self.token()
            if token != None and token[0] == "sep" and token[1] == ')':
                self.type = self.type + token[1]
                token = self.token()
                if token != None and token[0] == "sep" and token[1] == '(':
                    token = self.token()
                    type = self.type;
                    token = self.parseSignature(token);
                    self.type = type;
                else:
                    self.error("parsing function type, '(' expected", token);
                    return token
            else:
                self.error("parsing function type, ')' expected", token);
                return token
            self.lexer.push(token)
            token = nametok
            return token

         #
         # do some lookahead for arrays
         #
        if token != None and token[0] == "name":
            nametok = token
            token = self.token()
            if token != None and token[0] == "sep" and token[1] == '[':
1665
                self.type = self.type + " " + nametok[1]
1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686
                while token != None and token[0] == "sep" and token[1] == '[':
                    self.type = self.type + token[1]
                    token = self.token()
                    while token != None and token[0] != 'sep' and \
                          token[1] != ']' and token[1] != ';':
                        self.type = self.type + token[1]
                        token = self.token()
                if token != None and token[0] == 'sep' and token[1] == ']':
                    self.type = self.type + token[1]
                    token = self.token()
                else:
                    self.error("parsing array type, ']' expected", token);
                    return token
            elif token != None and token[0] == "sep" and token[1] == ':':
                 # remove :12 in case it's a limited int size
                token = self.token()
                token = self.token()
            self.lexer.push(token)
            token = nametok

        return token
1687 1688 1689 1690 1691 1692

     #
     # Parse a signature: '(' has been parsed and we scan the type definition
     #    up to the ')' included
    def parseSignature(self, token):
        signature = []
1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719
        if token != None and token[0] == "sep" and token[1] == ')':
            self.signature = []
            token = self.token()
            return token
        while token != None:
            token = self.parseType(token)
            if token != None and token[0] == "name":
                signature.append((self.type, token[1], None))
                token = self.token()
            elif token != None and token[0] == "sep" and token[1] == ',':
                token = self.token()
                continue
            elif token != None and token[0] == "sep" and token[1] == ')':
                 # only the type was provided
                if self.type == "...":
                    signature.append((self.type, "...", None))
                else:
                    signature.append((self.type, None, None))
            if token != None and token[0] == "sep":
                if token[1] == ',':
                    token = self.token()
                    continue
                elif token[1] == ')':
                    token = self.token()
                    break
        self.signature = signature
        return token
1720

1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752
    # this dict contains the functions that are allowed to use [unsigned]
    # long for legacy reasons in their signature and return type. this list is
    # fixed. new procedures and public APIs have to use [unsigned] long long
    long_legacy_functions = \
      { "virGetVersion"                  : (False, ("libVer", "typeVer")),
        "virConnectGetLibVersion"        : (False, ("libVer")),
        "virConnectGetVersion"           : (False, ("hvVer")),
        "virDomainGetMaxMemory"          : (True,  ()),
        "virDomainMigrate"               : (False, ("flags", "bandwidth")),
        "virDomainMigrate2"              : (False, ("flags", "bandwidth")),
        "virDomainMigrateBegin3"         : (False, ("flags", "bandwidth")),
        "virDomainMigrateConfirm3"       : (False, ("flags", "bandwidth")),
        "virDomainMigrateDirect"         : (False, ("flags", "bandwidth")),
        "virDomainMigrateFinish"         : (False, ("flags")),
        "virDomainMigrateFinish2"        : (False, ("flags")),
        "virDomainMigrateFinish3"        : (False, ("flags")),
        "virDomainMigratePeer2Peer"      : (False, ("flags", "bandwidth")),
        "virDomainMigratePerform"        : (False, ("flags", "bandwidth")),
        "virDomainMigratePerform3"       : (False, ("flags", "bandwidth")),
        "virDomainMigratePrepare"        : (False, ("flags", "bandwidth")),
        "virDomainMigratePrepare2"       : (False, ("flags", "bandwidth")),
        "virDomainMigratePrepare3"       : (False, ("flags", "bandwidth")),
        "virDomainMigratePrepareTunnel"  : (False, ("flags", "bandwidth")),
        "virDomainMigratePrepareTunnel3" : (False, ("flags", "bandwidth")),
        "virDomainMigrateToURI"          : (False, ("flags", "bandwidth")),
        "virDomainMigrateToURI2"         : (False, ("flags", "bandwidth")),
        "virDomainMigrateVersion1"       : (False, ("flags", "bandwidth")),
        "virDomainMigrateVersion2"       : (False, ("flags", "bandwidth")),
        "virDomainMigrateVersion3"       : (False, ("flags", "bandwidth")),
        "virDomainMigrateSetMaxSpeed"    : (False, ("bandwidth")),
        "virDomainSetMaxMemory"          : (False, ("memory")),
        "virDomainSetMemory"             : (False, ("memory")),
1753
        "virDomainSetMemoryFlags"        : (False, ("memory")),
E
Eric Blake 已提交
1754
        "virDomainBlockCommit"           : (False, ("bandwidth")),
1755
        "virDomainBlockJobSetSpeed"      : (False, ("bandwidth")),
1756
        "virDomainBlockPull"             : (False, ("bandwidth")),
1757
        "virDomainBlockRebase"           : (False, ("bandwidth")),
1758
        "virDomainMigrateGetMaxSpeed"    : (False, ("bandwidth")) }
1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783

    def checkLongLegacyFunction(self, name, return_type, signature):
        if "long" in return_type and "long long" not in return_type:
            try:
                if not CParser.long_legacy_functions[name][0]:
                    raise Exception()
            except:
                self.error(("function '%s' is not allowed to return long, "
                            "use long long instead") % (name))

        for param in signature:
            if "long" in param[0] and "long long" not in param[0]:
                try:
                    if param[1] not in CParser.long_legacy_functions[name][1]:
                        raise Exception()
                except:
                    self.error(("function '%s' is not allowed to take long "
                                "parameter '%s', use long long instead")
                               % (name, param[1]))

    # this dict contains the structs that are allowed to use [unsigned]
    # long for legacy reasons. this list is fixed. new structs have to use
    # [unsigned] long long
    long_legacy_struct_fields = \
      { "_virDomainInfo"                 : ("maxMem", "memory"),
1784 1785
        "_virNodeInfo"                   : ("memory"),
        "_virDomainBlockJobInfo"         : ("bandwidth") }
1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797

    def checkLongLegacyStruct(self, name, fields):
        for field in fields:
            if "long" in field[0] and "long long" not in field[0]:
                try:
                    if field[1] not in CParser.long_legacy_struct_fields[name]:
                        raise Exception()
                except:
                    self.error(("struct '%s' is not allowed to contain long "
                                "field '%s', use long long instead") \
                               % (name, field[1]))

1798 1799 1800 1801 1802 1803 1804
     #
     # Parse a global definition, be it a type, variable or function
     # the extern "C" blocks are a bit nasty and require it to recurse.
     #
    def parseGlobal(self, token):
        static = 0
        if token[1] == 'extern':
1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879
            token = self.token()
            if token == None:
                return token
            if token[0] == 'string':
                if token[1] == 'C':
                    token = self.token()
                    if token == None:
                        return token
                    if token[0] == 'sep' and token[1] == "{":
                        token = self.token()
#                        print 'Entering extern "C line ', self.lineno()
                        while token != None and (token[0] != 'sep' or
                              token[1] != "}"):
                            if token[0] == 'name':
                                token = self.parseGlobal(token)
                            else:
                                self.error(
                                 "token %s %s unexpected at the top level" % (
                                        token[0], token[1]))
                                token = self.parseGlobal(token)
#                        print 'Exiting extern "C" line', self.lineno()
                        token = self.token()
                        return token
                else:
                    return token
        elif token[1] == 'static':
            static = 1
            token = self.token()
            if token == None or  token[0] != 'name':
                return token

        if token[1] == 'typedef':
            token = self.token()
            return self.parseTypedef(token)
        else:
            token = self.parseType(token)
            type_orig = self.type
        if token == None or token[0] != "name":
            return token
        type = type_orig
        self.name = token[1]
        token = self.token()
        while token != None and (token[0] == "sep" or token[0] == "op"):
            if token[0] == "sep":
                if token[1] == "[":
                    type = type + token[1]
                    token = self.token()
                    while token != None and (token[0] != "sep" or \
                          token[1] != ";"):
                        type = type + token[1]
                        token = self.token()

            if token != None and token[0] == "op" and token[1] == "=":
                 #
                 # Skip the initialization of the variable
                 #
                token = self.token()
                if token[0] == 'sep' and token[1] == '{':
                    token = self.token()
                    token = self.parseBlock(token)
                else:
                    self.comment = None
                    while token != None and (token[0] != "sep" or \
                          (token[1] != ';' and token[1] != ',')):
                            token = self.token()
                self.comment = None
                if token == None or token[0] != "sep" or (token[1] != ';' and
                   token[1] != ','):
                    self.error("missing ';' or ',' after value")

            if token != None and token[0] == "sep":
                if token[1] == ";":
                    self.comment = None
                    token = self.token()
                    if type == "struct":
1880
                        self.checkLongLegacyStruct(self.name, self.struct_fields)
1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892
                        self.index_add(self.name, self.filename,
                             not self.is_header, "struct", self.struct_fields)
                    else:
                        self.index_add(self.name, self.filename,
                             not self.is_header, "variable", type)
                    break
                elif token[1] == "(":
                    token = self.token()
                    token = self.parseSignature(token)
                    if token == None:
                        return None
                    if token[0] == "sep" and token[1] == ";":
1893
                        self.checkLongLegacyFunction(self.name, type, self.signature)
1894 1895 1896 1897 1898 1899
                        d = self.mergeFunctionComment(self.name,
                                ((type, None), self.signature), 1)
                        self.index_add(self.name, self.filename, static,
                                        "function", d)
                        token = self.token()
                    elif token[0] == "sep" and token[1] == "{":
1900
                        self.checkLongLegacyFunction(self.name, type, self.signature)
1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922
                        d = self.mergeFunctionComment(self.name,
                                ((type, None), self.signature), static)
                        self.index_add(self.name, self.filename, static,
                                        "function", d)
                        token = self.token()
                        token = self.parseBlock(token);
                elif token[1] == ',':
                    self.comment = None
                    self.index_add(self.name, self.filename, static,
                                    "variable", type)
                    type = type_orig
                    token = self.token()
                    while token != None and token[0] == "sep":
                        type = type + token[1]
                        token = self.token()
                    if token != None and token[0] == "name":
                        self.name = token[1]
                        token = self.token()
                else:
                    break

        return token
1923 1924

    def parse(self):
1925 1926
        if not quiet:
            print "Parsing %s" % (self.filename)
1927
        token = self.token()
1928
        while token != None:
1929
            if token[0] == 'name':
1930
                token = self.parseGlobal(token)
1931
            else:
1932 1933 1934 1935 1936
                self.error("token %s %s unexpected at the top level" % (
                       token[0], token[1]))
                token = self.parseGlobal(token)
                return
        self.parseTopComment(self.top_comment)
1937
        return self.index
1938

1939 1940 1941

class docBuilder:
    """A documentation builder"""
J
Jiri Denemark 已提交
1942
    def __init__(self, name, path='.', directories=['.'], includes=[]):
1943
        self.name = name
J
Jiri Denemark 已提交
1944
        self.path = path
1945
        self.directories = directories
1946 1947 1948 1949
        if name == "libvirt":
            self.includes = includes + included_files.keys()
        elif name == "libvirt-qemu":
            self.includes = includes + qemu_included_files.keys()
1950 1951 1952
        self.modules = {}
        self.headers = {}
        self.idx = index()
1953
        self.xref = {}
1954 1955
        self.index = {}
        self.basename = name
1956

1957 1958 1959 1960 1961
    def warning(self, msg):
        global warnings
        warnings = warnings + 1
        print msg

1962
    def indexString(self, id, str):
1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998
        if str == None:
            return
        str = string.replace(str, "'", ' ')
        str = string.replace(str, '"', ' ')
        str = string.replace(str, "/", ' ')
        str = string.replace(str, '*', ' ')
        str = string.replace(str, "[", ' ')
        str = string.replace(str, "]", ' ')
        str = string.replace(str, "(", ' ')
        str = string.replace(str, ")", ' ')
        str = string.replace(str, "<", ' ')
        str = string.replace(str, '>', ' ')
        str = string.replace(str, "&", ' ')
        str = string.replace(str, '#', ' ')
        str = string.replace(str, ",", ' ')
        str = string.replace(str, '.', ' ')
        str = string.replace(str, ';', ' ')
        tokens = string.split(str)
        for token in tokens:
            try:
                c = token[0]
                if string.find(string.letters, c) < 0:
                    pass
                elif len(token) < 3:
                    pass
                else:
                    lower = string.lower(token)
                    # TODO: generalize this a bit
                    if lower == 'and' or lower == 'the':
                        pass
                    elif self.xref.has_key(token):
                        self.xref[token].append(id)
                    else:
                        self.xref[token] = [id]
            except:
                pass
1999 2000

    def analyze(self):
2001 2002
        if not quiet:
            print "Project %s : %d headers, %d modules" % (self.name, len(self.headers.keys()), len(self.modules.keys()))
2003
        self.idx.analyze()
2004 2005

    def scanHeaders(self):
2006 2007 2008 2009 2010
        for header in self.headers.keys():
            parser = CParser(header)
            idx = parser.parse()
            self.headers[header] = idx;
            self.idx.merge(idx)
2011 2012

    def scanModules(self):
2013 2014 2015 2016 2017 2018
        for module in self.modules.keys():
            parser = CParser(module)
            idx = parser.parse()
            # idx.analyze()
            self.modules[module] = idx
            self.idx.merge_public(idx)
2019 2020 2021

    def scan(self):
        for directory in self.directories:
2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041
            files = glob.glob(directory + "/*.c")
            for file in files:
                skip = 1
                for incl in self.includes:
                    if string.find(file, incl) != -1:
                        skip = 0;
                        break
                if skip == 0:
                    self.modules[file] = None;
            files = glob.glob(directory + "/*.h")
            for file in files:
                skip = 1
                for incl in self.includes:
                    if string.find(file, incl) != -1:
                        skip = 0;
                        break
                if skip == 0:
                    self.headers[file] = None;
        self.scanHeaders()
        self.scanModules()
2042

2043 2044
    def modulename_file(self, file):
        module = os.path.basename(file)
2045 2046 2047 2048 2049
        if module[-2:] == '.h':
            module = module[:-2]
        elif module[-2:] == '.c':
            module = module[:-2]
        return module
2050 2051 2052 2053

    def serialize_enum(self, output, name):
        id = self.idx.enums[name]
        output.write("    <enum name='%s' file='%s'" % (name,
2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066
                     self.modulename_file(id.header)))
        if id.info != None:
            info = id.info
            if info[0] != None and info[0] != '':
                try:
                    val = eval(info[0])
                except:
                    val = info[0]
                output.write(" value='%s'" % (val));
            if info[2] != None and info[2] != '':
                output.write(" type='%s'" % info[2]);
            if info[1] != None and info[1] != '':
                output.write(" info='%s'" % escape(info[1]));
2067 2068 2069 2070 2071
        output.write("/>\n")

    def serialize_macro(self, output, name):
        id = self.idx.macros[name]
        output.write("    <macro name='%s' file='%s'>\n" % (name,
2072 2073
                     self.modulename_file(id.header)))
        if id.info != None:
2074
            try:
2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086
                (args, desc) = id.info
                if desc != None and desc != "":
                    output.write("      <info><![CDATA[%s]]></info>\n" % (desc))
                    self.indexString(name, desc)
                for arg in args:
                    (name, desc) = arg
                    if desc != None and desc != "":
                        output.write("      <arg name='%s' info='%s'/>\n" % (
                                     name, escape(desc)))
                        self.indexString(name, desc)
                    else:
                        output.write("      <arg name='%s'/>\n" % (name))
2087 2088 2089 2090
            except:
                pass
        output.write("    </macro>\n")

2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104
    def serialize_union(self, output, field, desc):
        output.write("      <field name='%s' type='union' info='%s'>\n" % (field[1] , desc))
        output.write("        <union>\n")
        for f in field[3]:
            desc = f[2]
            if desc == None:
                desc = ''
            else:
                desc = escape(desc)
            output.write("          <field name='%s' type='%s' info='%s'/>\n" % (f[1] , f[0], desc))

        output.write("        </union>\n")
        output.write("      </field>\n")

2105 2106
    def serialize_typedef(self, output, name):
        id = self.idx.typedefs[name]
2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122
        if id.info[0:7] == 'struct ':
            output.write("    <struct name='%s' file='%s' type='%s'" % (
                     name, self.modulename_file(id.header), id.info))
            name = id.info[7:]
            if self.idx.structs.has_key(name) and ( \
               type(self.idx.structs[name].info) == type(()) or
                type(self.idx.structs[name].info) == type([])):
                output.write(">\n");
                try:
                    for field in self.idx.structs[name].info:
                        desc = field[2]
                        self.indexString(name, desc)
                        if desc == None:
                            desc = ''
                        else:
                            desc = escape(desc)
2123 2124 2125 2126
                        if field[0] == "union":
                            self.serialize_union(output, field, desc)
                        else:
                            output.write("      <field name='%s' type='%s' info='%s'/>\n" % (field[1] , field[0], desc))
2127
                except:
2128
                    self.warning("Failed to serialize struct %s" % (name))
2129 2130 2131 2132 2133 2134
                output.write("    </struct>\n")
            else:
                output.write("/>\n");
        else :
            output.write("    <typedef name='%s' file='%s' type='%s'" % (
                         name, self.modulename_file(id.header), id.info))
2135
            try:
2136 2137 2138 2139 2140 2141 2142 2143
                desc = id.extra
                if desc != None and desc != "":
                    output.write(">\n      <info><![CDATA[%s]]></info>\n" % (desc))
                    output.write("    </typedef>\n")
                else:
                    output.write("/>\n")
            except:
                output.write("/>\n")
2144 2145 2146

    def serialize_variable(self, output, name):
        id = self.idx.variables[name]
2147 2148 2149 2150 2151 2152
        if id.info != None:
            output.write("    <variable name='%s' file='%s' type='%s'/>\n" % (
                    name, self.modulename_file(id.header), id.info))
        else:
            output.write("    <variable name='%s' file='%s'/>\n" % (
                    name, self.modulename_file(id.header)))
2153

2154 2155
    def serialize_function(self, output, name):
        id = self.idx.functions[name]
2156
        if name == debugsym and not quiet:
2157
            print "=>", id
2158 2159

        output.write("    <%s name='%s' file='%s' module='%s'>\n" % (id.type,
2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191
                     name, self.modulename_file(id.header),
                     self.modulename_file(id.module)))
        #
        # Processing of conditionals modified by Bill 1/1/05
        #
        if id.conditionals != None:
            apstr = ""
            for cond in id.conditionals:
                if apstr != "":
                    apstr = apstr + " &amp;&amp; "
                apstr = apstr + cond
            output.write("      <cond>%s</cond>\n"% (apstr));
        try:
            (ret, params, desc) = id.info
            output.write("      <info><![CDATA[%s]]></info>\n" % (desc))
            self.indexString(name, desc)
            if ret[0] != None:
                if ret[0] == "void":
                    output.write("      <return type='void'/>\n")
                else:
                    output.write("      <return type='%s' info='%s'/>\n" % (
                             ret[0], escape(ret[1])))
                    self.indexString(name, ret[1])
            for param in params:
                if param[0] == 'void':
                    continue
                if param[2] == None:
                    output.write("      <arg name='%s' type='%s' info=''/>\n" % (param[1], param[0]))
                else:
                    output.write("      <arg name='%s' type='%s' info='%s'/>\n" % (param[1], param[0], escape(param[2])))
                    self.indexString(name, param[2])
        except:
2192
            self.warning("Failed to save function %s info: " % name, `id.info`)
2193 2194 2195 2196
        output.write("    </%s>\n" % (id.type))

    def serialize_exports(self, output, file):
        module = self.modulename_file(file)
2197 2198 2199 2200 2201 2202 2203 2204 2205 2206
        output.write("    <file name='%s'>\n" % (module))
        dict = self.headers[file]
        if dict.info != None:
            for data in ('Summary', 'Description', 'Author'):
                try:
                    output.write("     <%s>%s</%s>\n" % (
                                 string.lower(data),
                                 escape(dict.info[data]),
                                 string.lower(data)))
                except:
2207
                    self.warning("Header %s lacks a %s description" % (module, data))
2208 2209 2210 2211
            if dict.info.has_key('Description'):
                desc = dict.info['Description']
                if string.find(desc, "DEPRECATED") != -1:
                    output.write("     <deprecated/>\n")
2212 2213

        ids = dict.macros.keys()
2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224
        ids.sort()
        for id in uniq(ids):
            # Macros are sometime used to masquerade other types.
            if dict.functions.has_key(id):
                continue
            if dict.variables.has_key(id):
                continue
            if dict.typedefs.has_key(id):
                continue
            if dict.structs.has_key(id):
                continue
2225 2226
            if dict.unions.has_key(id):
                continue
2227 2228 2229
            if dict.enums.has_key(id):
                continue
            output.write("     <exports symbol='%s' type='macro'/>\n" % (id))
2230
        ids = dict.enums.keys()
2231 2232 2233
        ids.sort()
        for id in uniq(ids):
            output.write("     <exports symbol='%s' type='enum'/>\n" % (id))
2234
        ids = dict.typedefs.keys()
2235 2236 2237
        ids.sort()
        for id in uniq(ids):
            output.write("     <exports symbol='%s' type='typedef'/>\n" % (id))
2238
        ids = dict.structs.keys()
2239 2240 2241
        ids.sort()
        for id in uniq(ids):
            output.write("     <exports symbol='%s' type='struct'/>\n" % (id))
2242
        ids = dict.variables.keys()
2243 2244 2245
        ids.sort()
        for id in uniq(ids):
            output.write("     <exports symbol='%s' type='variable'/>\n" % (id))
2246
        ids = dict.functions.keys()
2247 2248 2249 2250
        ids.sort()
        for id in uniq(ids):
            output.write("     <exports symbol='%s' type='function'/>\n" % (id))
        output.write("    </file>\n")
2251 2252 2253 2254 2255

    def serialize_xrefs_files(self, output):
        headers = self.headers.keys()
        headers.sort()
        for file in headers:
2256 2257 2258 2259 2260 2261 2262 2263 2264 2265
            module = self.modulename_file(file)
            output.write("    <file name='%s'>\n" % (module))
            dict = self.headers[file]
            ids = uniq(dict.functions.keys() + dict.variables.keys() + \
                  dict.macros.keys() + dict.typedefs.keys() + \
                  dict.structs.keys() + dict.enums.keys())
            ids.sort()
            for id in ids:
                output.write("      <ref name='%s'/>\n" % (id))
            output.write("    </file>\n")
2266 2267 2268 2269
        pass

    def serialize_xrefs_functions(self, output):
        funcs = {}
2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297
        for name in self.idx.functions.keys():
            id = self.idx.functions[name]
            try:
                (ret, params, desc) = id.info
                for param in params:
                    if param[0] == 'void':
                        continue
                    if funcs.has_key(param[0]):
                        funcs[param[0]].append(name)
                    else:
                        funcs[param[0]] = [name]
            except:
                pass
        typ = funcs.keys()
        typ.sort()
        for type in typ:
            if type == '' or type == 'void' or type == "int" or \
               type == "char *" or type == "const char *" :
                continue
            output.write("    <type name='%s'>\n" % (type))
            ids = funcs[type]
            ids.sort()
            pid = ''    # not sure why we have dups, but get rid of them!
            for id in ids:
                if id != pid:
                    output.write("      <ref name='%s'/>\n" % (id))
                    pid = id
            output.write("    </type>\n")
2298 2299 2300

    def serialize_xrefs_constructors(self, output):
        funcs = {}
2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324
        for name in self.idx.functions.keys():
            id = self.idx.functions[name]
            try:
                (ret, params, desc) = id.info
                if ret[0] == "void":
                    continue
                if funcs.has_key(ret[0]):
                    funcs[ret[0]].append(name)
                else:
                    funcs[ret[0]] = [name]
            except:
                pass
        typ = funcs.keys()
        typ.sort()
        for type in typ:
            if type == '' or type == 'void' or type == "int" or \
               type == "char *" or type == "const char *" :
                continue
            output.write("    <type name='%s'>\n" % (type))
            ids = funcs[type]
            ids.sort()
            for id in ids:
                output.write("      <ref name='%s'/>\n" % (id))
            output.write("    </type>\n")
2325 2326

    def serialize_xrefs_alpha(self, output):
2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338
        letter = None
        ids = self.idx.identifiers.keys()
        ids.sort()
        for id in ids:
            if id[0] != letter:
                if letter != None:
                    output.write("    </letter>\n")
                letter = id[0]
                output.write("    <letter name='%s'>\n" % (letter))
            output.write("      <ref name='%s'/>\n" % (id))
        if letter != None:
            output.write("    </letter>\n")
2339 2340 2341

    def serialize_xrefs_references(self, output):
        typ = self.idx.identifiers.keys()
2342 2343 2344 2345 2346 2347 2348 2349
        typ.sort()
        for id in typ:
            idf = self.idx.identifiers[id]
            module = idf.header
            output.write("    <reference name='%s' href='%s'/>\n" % (id,
                         'html/' + self.basename + '-' +
                         self.modulename_file(module) + '.html#' +
                         id))
2350 2351 2352

    def serialize_xrefs_index(self, output):
        index = self.xref
2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396
        typ = index.keys()
        typ.sort()
        letter = None
        count = 0
        chunk = 0
        chunks = []
        for id in typ:
            if len(index[id]) > 30:
                continue
            if id[0] != letter:
                if letter == None or count > 200:
                    if letter != None:
                        output.write("      </letter>\n")
                        output.write("    </chunk>\n")
                        count = 0
                        chunks.append(["chunk%s" % (chunk -1), first_letter, letter])
                    output.write("    <chunk name='chunk%s'>\n" % (chunk))
                    first_letter = id[0]
                    chunk = chunk + 1
                elif letter != None:
                    output.write("      </letter>\n")
                letter = id[0]
                output.write("      <letter name='%s'>\n" % (letter))
            output.write("        <word name='%s'>\n" % (id))
            tokens = index[id];
            tokens.sort()
            tok = None
            for token in tokens:
                if tok == token:
                    continue
                tok = token
                output.write("          <ref name='%s'/>\n" % (token))
                count = count + 1
            output.write("        </word>\n")
        if letter != None:
            output.write("      </letter>\n")
            output.write("    </chunk>\n")
            if count != 0:
                chunks.append(["chunk%s" % (chunk -1), first_letter, letter])
            output.write("    <chunks>\n")
            for ch in chunks:
                output.write("      <chunk name='%s' start='%s' end='%s'/>\n" % (
                             ch[0], ch[1], ch[2]))
            output.write("    </chunks>\n")
2397 2398

    def serialize_xrefs(self, output):
2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416
        output.write("  <references>\n")
        self.serialize_xrefs_references(output)
        output.write("  </references>\n")
        output.write("  <alpha>\n")
        self.serialize_xrefs_alpha(output)
        output.write("  </alpha>\n")
        output.write("  <constructors>\n")
        self.serialize_xrefs_constructors(output)
        output.write("  </constructors>\n")
        output.write("  <functions>\n")
        self.serialize_xrefs_functions(output)
        output.write("  </functions>\n")
        output.write("  <files>\n")
        self.serialize_xrefs_files(output)
        output.write("  </files>\n")
        output.write("  <index>\n")
        self.serialize_xrefs_index(output)
        output.write("  </index>\n")
2417 2418

    def serialize(self):
J
Jiri Denemark 已提交
2419
        filename = "%s/%s-api.xml" % (self.path, self.name)
2420 2421
        if not quiet:
            print "Saving XML description %s" % (filename)
2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455
        output = open(filename, "w")
        output.write('<?xml version="1.0" encoding="ISO-8859-1"?>\n')
        output.write("<api name='%s'>\n" % self.name)
        output.write("  <files>\n")
        headers = self.headers.keys()
        headers.sort()
        for file in headers:
            self.serialize_exports(output, file)
        output.write("  </files>\n")
        output.write("  <symbols>\n")
        macros = self.idx.macros.keys()
        macros.sort()
        for macro in macros:
            self.serialize_macro(output, macro)
        enums = self.idx.enums.keys()
        enums.sort()
        for enum in enums:
            self.serialize_enum(output, enum)
        typedefs = self.idx.typedefs.keys()
        typedefs.sort()
        for typedef in typedefs:
            self.serialize_typedef(output, typedef)
        variables = self.idx.variables.keys()
        variables.sort()
        for variable in variables:
            self.serialize_variable(output, variable)
        functions = self.idx.functions.keys()
        functions.sort()
        for function in functions:
            self.serialize_function(output, function)
        output.write("  </symbols>\n")
        output.write("</api>\n")
        output.close()

J
Jiri Denemark 已提交
2456
        filename = "%s/%s-refs.xml" % (self.path, self.name)
2457 2458
        if not quiet:
            print "Saving XML Cross References %s" % (filename)
2459 2460 2461 2462 2463 2464 2465 2466
        output = open(filename, "w")
        output.write('<?xml version="1.0" encoding="ISO-8859-1"?>\n')
        output.write("<apirefs name='%s'>\n" % self.name)
        self.serialize_xrefs(output)
        output.write("</apirefs>\n")
        output.close()


2467 2468
def rebuild(name):
    if name not in ["libvirt", "libvirt-qemu"]:
J
Ján Tomko 已提交
2469
        self.warning("rebuild() failed, unknown module %s") % name
2470
        return None
2471
    builder = None
2472 2473
    srcdir = os.environ["srcdir"]
    if glob.glob(srcdir + "/../src/libvirt.c") != [] :
2474
        if not quiet:
2475
            print "Rebuilding API description for %s" % name
J
Jiri Denemark 已提交
2476 2477 2478 2479 2480
        dirs = [srcdir + "/../src",
                srcdir + "/../src/util",
                srcdir + "/../include/libvirt"]
        if glob.glob(srcdir + "/../include/libvirt/libvirt.h") == [] :
            dirs.append("../include/libvirt")
2481
        builder = docBuilder(name, srcdir, dirs, [])
D
Daniel Veillard 已提交
2482
    elif glob.glob("src/libvirt.c") != [] :
2483
        if not quiet:
2484 2485
            print "Rebuilding API description for %s" % name
        builder = docBuilder(name, srcdir,
J
Jiri Denemark 已提交
2486
                             ["src", "src/util", "include/libvirt"],
2487
                             [])
2488
    else:
2489
        self.warning("rebuild() failed, unable to guess the module")
2490
        return None
2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508
    builder.scan()
    builder.analyze()
    builder.serialize()
    return builder

#
# for debugging the parser
#
def parse(filename):
    parser = CParser(filename)
    idx = parser.parse()
    return idx

if __name__ == "__main__":
    if len(sys.argv) > 1:
        debug = 1
        parse(sys.argv[1])
    else:
2509 2510
        rebuild("libvirt")
        rebuild("libvirt-qemu")
2511 2512 2513 2514
    if warnings > 0:
        sys.exit(2)
    else:
        sys.exit(0)