#!/usr/bin/env python # encoding: utf-8 # Thomas Nagy, 2006-2009 (ita) """ C/C++ preprocessor for finding dependencies Reasons for using the Waf preprocessor by default 1. Some c/c++ extensions (Qt) require a custom preprocessor for obtaining the dependencies (.moc files) 2. Not all compilers provide .d files for obtaining the dependencies (portability) 3. A naive file scanner will not catch the constructs such as "#include foo()" 4. A naive file scanner will catch unnecessary dependencies (change an unused header -> recompile everything) Regarding the speed concerns: a. the preprocessing is performed only when files must be compiled b. the macros are evaluated only for #if/#elif/#include c. the time penalty is about 10% d. system headers are not scanned Now if you do not want the Waf preprocessor, the tool "gccdeps" uses the .d files produced during the compilation to track the dependencies (useful when used with the boost libraries). It only works with gcc though, and it cannot be used with Qt builds. A dumb file scanner will be added in the future, so we will have most bahaviours. """ # TODO: more varargs, pragma once # TODO: dumb file scanner tracking all includes import re, sys, os, string import Logs, Build, Utils from Logs import debug, error import traceback class PreprocError(Utils.WafError): pass POPFILE = '-' recursion_limit = 5000 "do not loop too much on header inclusion" go_absolute = 0 "set to 1 to track headers on files in /usr/include - else absolute paths are ignored" standard_includes = ['/usr/include'] if sys.platform == "win32": standard_includes = [] use_trigraphs = 0 'apply the trigraph rules first' strict_quotes = 0 "Keep <> for system includes (do not search for those includes)" g_optrans = { 'not':'!', 'and':'&&', 'bitand':'&', 'and_eq':'&=', 'or':'||', 'bitor':'|', 'or_eq':'|=', 'xor':'^', 'xor_eq':'^=', 'compl':'~', } "these ops are for c++, to reset, set an empty dict" # ignore #warning and #error re_lines = re.compile(\ '^[ \t]*(#|%:)[ \t]*(ifdef|ifndef|if|else|elif|endif|include|import|define|undef|pragma)[ \t]*(.*)\r*$', re.IGNORECASE | re.MULTILINE) re_mac = re.compile("^[a-zA-Z_]\w*") re_fun = re.compile('^[a-zA-Z_][a-zA-Z0-9_]*[(]') re_pragma_once = re.compile('^\s*once\s*', re.IGNORECASE) re_nl = re.compile('\\\\\r*\n', re.MULTILINE) re_cpp = re.compile( r"""(/\*[^*]*\*+(?:[^/*][^*]*\*+)*/)|//[^\n]*|("(?:\\.|[^"\\])*"|'(?:\\.|[^'\\])*'|.[^/"'\\]*)""", re.MULTILINE) trig_def = [('??'+a, b) for a, b in zip("=-/!'()<>", r'#~\|^[]{}')] chr_esc = {'0':0, 'a':7, 'b':8, 't':9, 'n':10, 'f':11, 'v':12, 'r':13, '\\':92, "'":39} NUM = 'i' OP = 'O' IDENT = 'T' STR = 's' CHAR = 'c' tok_types = [NUM, STR, IDENT, OP] exp_types = [ r"""0[xX](?P[a-fA-F0-9]+)(?P[uUlL]*)|L*?'(?P(\\.|[^\\'])+)'|(?P\d+)[Ee](?P[+-]*?\d+)(?P[fFlL]*)|(?P\d*\.\d+)([Ee](?P[+-]*?\d+))?(?P[fFlL]*)|(?P\d+\.\d*)([Ee](?P[+-]*?\d+))?(?P[fFlL]*)|(?P0*)(?P\d+)(?P[uUlL]*)""", r'L?"([^"\\]|\\.)*"', r'[a-zA-Z_]\w*', r'%:%:|<<=|>>=|\.\.\.|<<|<%|<:|<=|>>|>=|\+\+|\+=|--|->|-=|\*=|/=|%:|%=|%>|==|&&|&=|\|\||\|=|\^=|:>|!=|##|[\(\)\{\}\[\]<>\?\|\^\*\+&=:!#;,%/\-\?\~\.]', ] re_clexer = re.compile('|'.join(["(?P<%s>%s)" % (name, part) for name, part in zip(tok_types, exp_types)]), re.M) accepted = 'a' ignored = 'i' undefined = 'u' skipped = 's' def repl(m): if m.group(1): return ' ' s = m.group(2) if s is None: return '' return s def filter_comments(filename): # return a list of tuples : keyword, line code = Utils.readf(filename) if use_trigraphs: for (a, b) in trig_def: code = code.split(a).join(b) code = re_nl.sub('', code) code = re_cpp.sub(repl, code) return [(m.group(2), m.group(3)) for m in re.finditer(re_lines, code)] prec = {} # op -> number, needed for such expressions: #if 1 && 2 != 0 ops = ['* / %', '+ -', '<< >>', '< <= >= >', '== !=', '& | ^', '&& ||', ','] for x in range(len(ops)): syms = ops[x] for u in syms.split(): prec[u] = x def reduce_nums(val_1, val_2, val_op): """apply arithmetic rules and try to return an integer result""" #print val_1, val_2, val_op # now perform the operation, make certain a and b are numeric try: a = 0 + val_1 except TypeError: a = int(val_1) try: b = 0 + val_2 except TypeError: b = int(val_2) d = val_op if d == '%': c = a%b elif d=='+': c = a+b elif d=='-': c = a-b elif d=='*': c = a*b elif d=='/': c = a/b elif d=='^': c = a^b elif d=='|': c = a|b elif d=='||': c = int(a or b) elif d=='&': c = a&b elif d=='&&': c = int(a and b) elif d=='==': c = int(a == b) elif d=='!=': c = int(a != b) elif d=='<=': c = int(a <= b) elif d=='<': c = int(a < b) elif d=='>': c = int(a > b) elif d=='>=': c = int(a >= b) elif d=='^': c = int(a^b) elif d=='<<': c = a<>': c = a>>b else: c = 0 return c def get_num(lst): if not lst: raise PreprocError("empty list for get_num") (p, v) = lst[0] if p == OP: if v == '(': count_par = 1 i = 1 while i < len(lst): (p, v) = lst[i] if p == OP: if v == ')': count_par -= 1 if count_par == 0: break elif v == '(': count_par += 1 i += 1 else: raise PreprocError("rparen expected %r" % lst) (num, _) = get_term(lst[1:i]) return (num, lst[i+1:]) elif v == '+': return get_num(lst[1:]) elif v == '-': num, lst = get_num(lst[1:]) return (reduce_nums('-1', num, '*'), lst) elif v == '!': num, lst = get_num(lst[1:]) return (int(not int(num)), lst) elif v == '~': return (~ int(num), lst) else: raise PreprocError("invalid op token %r for get_num" % lst) elif p == NUM: return v, lst[1:] elif p == IDENT: # all macros should have been replaced, remaining identifiers eval to 0 return 0, lst[1:] else: raise PreprocError("invalid token %r for get_num" % lst) def get_term(lst): if not lst: raise PreprocError("empty list for get_term") num, lst = get_num(lst) if not lst: return (num, []) (p, v) = lst[0] if p == OP: if v == '&&' and not num: return (num, []) elif v == '||' and num: return (num, []) elif v == ',': # skip return get_term(lst[1:]) elif v == '?': count_par = 0 i = 1 while i < len(lst): (p, v) = lst[i] if p == OP: if v == ')': count_par -= 1 elif v == '(': count_par += 1 elif v == ':': if count_par == 0: break i += 1 else: raise PreprocError("rparen expected %r" % lst) if int(num): return get_term(lst[1:i]) else: return get_term(lst[i+1:]) else: num2, lst = get_num(lst[1:]) if not lst: # no more tokens to process num2 = reduce_nums(num, num2, v) return get_term([(NUM, num2)] + lst) # operator precedence p2, v2 = lst[0] if p2 != OP: raise PreprocError("op expected %r" % lst) if prec[v2] >= prec[v]: num2 = reduce_nums(num, num2, v) return get_term([(NUM, num2)] + lst) else: num3, lst = get_num(lst[1:]) num3 = reduce_nums(num2, num3, v2) return get_term([(NUM, num), (p, v), (NUM, num3)] + lst) raise PreprocError("cannot reduce %r" % lst) def reduce_eval(lst): """take a list of tokens and output true or false (#if/#elif conditions)""" num, lst = get_term(lst) return (NUM, num) def stringize(lst): """use for converting a list of tokens to a string""" lst = [str(v2) for (p2, v2) in lst] return "".join(lst) def paste_tokens(t1, t2): """ here is what we can paste: a ## b -> ab > ## = -> >= a ## 2 -> a2 """ p1 = None if t1[0] == OP and t2[0] == OP: p1 = OP elif t1[0] == IDENT and (t2[0] == IDENT or t2[0] == NUM): p1 = IDENT elif t1[0] == NUM and t2[0] == NUM: p1 = NUM if not p1: raise PreprocError('tokens do not make a valid paste %r and %r' % (t1, t2)) return (p1, t1[1] + t2[1]) def reduce_tokens(lst, defs, ban=[]): """replace the tokens in lst, using the macros provided in defs, and a list of macros that cannot be re-applied""" i = 0 while i < len(lst): (p, v) = lst[i] if p == IDENT and v == "defined": del lst[i] if i < len(lst): (p2, v2) = lst[i] if p2 == IDENT: if v2 in defs: lst[i] = (NUM, 1) else: lst[i] = (NUM, 0) elif p2 == OP and v2 == '(': del lst[i] (p2, v2) = lst[i] del lst[i] # remove the ident, and change the ) for the value if v2 in defs: lst[i] = (NUM, 1) else: lst[i] = (NUM, 0) else: raise PreprocError("invalid define expression %r" % lst) elif p == IDENT and v in defs: if isinstance(defs[v], str): a, b = extract_macro(defs[v]) defs[v] = b macro_def = defs[v] to_add = macro_def[1] if isinstance(macro_def[0], list): # macro without arguments del lst[i] for x in xrange(len(to_add)): lst.insert(i, to_add[x]) i += 1 else: # collect the arguments for the funcall args = [] del lst[i] if i >= len(lst): raise PreprocError("expected '(' after %r (got nothing)" % v) (p2, v2) = lst[i] if p2 != OP or v2 != '(': raise PreprocError("expected '(' after %r" % v) del lst[i] one_param = [] count_paren = 0 while i < len(lst): p2, v2 = lst[i] del lst[i] if p2 == OP and count_paren == 0: if v2 == '(': one_param.append((p2, v2)) count_paren += 1 elif v2 == ')': if one_param: args.append(one_param) break elif v2 == ',': if not one_param: raise PreprocError("empty param in funcall %s" % p) args.append(one_param) one_param = [] else: one_param.append((p2, v2)) else: one_param.append((p2, v2)) if v2 == '(': count_paren += 1 elif v2 == ')': count_paren -= 1 else: raise PreprocError('malformed macro') # substitute the arguments within the define expression accu = [] arg_table = macro_def[0] j = 0 while j < len(to_add): (p2, v2) = to_add[j] if p2 == OP and v2 == '#': # stringize is for arguments only if j+1 < len(to_add) and to_add[j+1][0] == IDENT and to_add[j+1][1] in arg_table: toks = args[arg_table[to_add[j+1][1]]] accu.append((STR, stringize(toks))) j += 1 else: accu.append((p2, v2)) elif p2 == OP and v2 == '##': # token pasting, how can man invent such a complicated system? if accu and j+1 < len(to_add): # we have at least two tokens t1 = accu[-1] if to_add[j+1][0] == IDENT and to_add[j+1][1] in arg_table: toks = args[arg_table[to_add[j+1][1]]] if toks: accu[-1] = paste_tokens(t1, toks[0]) #(IDENT, accu[-1][1] + toks[0][1]) accu.extend(toks[1:]) else: # error, case "a##" accu.append((p2, v2)) accu.extend(toks) elif to_add[j+1][0] == IDENT and to_add[j+1][1] == '__VA_ARGS__': # TODO not sure # first collect the tokens va_toks = [] st = len(macro_def[0]) pt = len(args) for x in args[pt-st+1:]: va_toks.extend(x) va_toks.append((OP, ',')) if va_toks: va_toks.pop() # extra comma if len(accu)>1: (p3, v3) = accu[-1] (p4, v4) = accu[-2] if v3 == '##': # remove the token paste accu.pop() if v4 == ',' and pt < st: # remove the comma accu.pop() accu += va_toks else: accu[-1] = paste_tokens(t1, to_add[j+1]) j += 1 else: # invalid paste, case "##a" or "b##" accu.append((p2, v2)) elif p2 == IDENT and v2 in arg_table: toks = args[arg_table[v2]] reduce_tokens(toks, defs, ban+[v]) accu.extend(toks) else: accu.append((p2, v2)) j += 1 reduce_tokens(accu, defs, ban+[v]) for x in xrange(len(accu)-1, -1, -1): lst.insert(i, accu[x]) i += 1 def eval_macro(lst, adefs): """reduce the tokens from the list lst, and try to return a 0/1 result""" reduce_tokens(lst, adefs, []) if not lst: raise PreprocError("missing tokens to evaluate") (p, v) = reduce_eval(lst) return int(v) != 0 def extract_macro(txt): """process a macro definition from "#define f(x, y) x * y" into a function or a simple macro without arguments""" t = tokenize(txt) if re_fun.search(txt): p, name = t[0] p, v = t[1] if p != OP: raise PreprocError("expected open parenthesis") i = 1 pindex = 0 params = {} prev = '(' while 1: i += 1 p, v = t[i] if prev == '(': if p == IDENT: params[v] = pindex pindex += 1 prev = p elif p == OP and v == ')': break else: raise PreprocError("unexpected token (3)") elif prev == IDENT: if p == OP and v == ',': prev = v elif p == OP and v == ')': break else: raise PreprocError("comma or ... expected") elif prev == ',': if p == IDENT: params[v] = pindex pindex += 1 prev = p elif p == OP and v == '...': raise PreprocError("not implemented (1)") else: raise PreprocError("comma or ... expected (2)") elif prev == '...': raise PreprocError("not implemented (2)") else: raise PreprocError("unexpected else") #~ print (name, [params, t[i+1:]]) return (name, [params, t[i+1:]]) else: (p, v) = t[0] return (v, [[], t[1:]]) re_include = re.compile('^\s*(<(?P.*)>|"(?P.*)")') def extract_include(txt, defs): """process a line in the form "#include foo" to return a string representing the file""" m = re_include.search(txt) if m: if m.group('a'): return '<', m.group('a') if m.group('b'): return '"', m.group('b') # perform preprocessing and look at the result, it must match an include toks = tokenize(txt) reduce_tokens(toks, defs, ['waf_include']) if not toks: raise PreprocError("could not parse include %s" % txt) if len(toks) == 1: if toks[0][0] == STR: return '"', toks[0][1] else: if toks[0][1] == '<' and toks[-1][1] == '>': return stringize(toks).lstrip('<').rstrip('>') raise PreprocError("could not parse include %s." % txt) def parse_char(txt): if not txt: raise PreprocError("attempted to parse a null char") if txt[0] != '\\': return ord(txt) c = txt[1] if c == 'x': if len(txt) == 4 and txt[3] in string.hexdigits: return int(txt[2:], 16) return int(txt[2:], 16) elif c.isdigit(): if c == '0' and len(txt)==2: return 0 for i in 3, 2, 1: if len(txt) > i and txt[1:1+i].isdigit(): return (1+i, int(txt[1:1+i], 8)) else: try: return chr_esc[c] except KeyError: raise PreprocError("could not parse char literal '%s'" % txt) @Utils.run_once def tokenize_private(s): ret = [] for match in re_clexer.finditer(s): m = match.group for name in tok_types: v = m(name) if v: if name == IDENT: try: v = g_optrans[v]; name = OP except KeyError: # c++ specific if v.lower() == "true": v = 1 name = NUM elif v.lower() == "false": v = 0 name = NUM elif name == NUM: if m('oct'): v = int(v, 8) elif m('hex'): v = int(m('hex'), 16) elif m('n0'): v = m('n0') else: v = m('char') if v: v = parse_char(v) else: v = m('n2') or m('n4') elif name == OP: if v == '%:': v = '#' elif v == '%:%:': v = '##' elif name == STR: # remove the quotes around the string v = v[1:-1] ret.append((name, v)) break return ret def tokenize(s): """convert a string into a list of tokens (shlex.split does not apply to c/c++/d)""" return tokenize_private(s)[:] @Utils.run_once def define_name(line): return re_mac.match(line).group(0) class c_parser(object): def __init__(self, nodepaths=None, defines=None): #self.lines = txt.split('\n') self.lines = [] if defines is None: self.defs = {} else: self.defs = dict(defines) # make a copy self.state = [] self.env = None # needed for the variant when searching for files self.count_files = 0 self.currentnode_stack = [] self.nodepaths = nodepaths or [] self.nodes = [] self.names = [] # file added self.curfile = '' self.ban_includes = set([]) def cached_find_resource(self, node, filename): try: nd = node.bld.cache_nd except: nd = node.bld.cache_nd = {} tup = (node.id, filename) try: return nd[tup] except KeyError: ret = node.find_resource(filename) nd[tup] = ret return ret def tryfind(self, filename): self.curfile = filename # for msvc it should be a for loop on the whole stack found = self.cached_find_resource(self.currentnode_stack[-1], filename) for n in self.nodepaths: if found: break found = self.cached_find_resource(n, filename) if found: self.nodes.append(found) if filename[-4:] != '.moc': self.addlines(found) else: if not filename in self.names: self.names.append(filename) return found def addlines(self, node): self.currentnode_stack.append(node.parent) filepath = node.abspath(self.env) self.count_files += 1 if self.count_files > recursion_limit: raise PreprocError("recursion limit exceeded") pc = self.parse_cache debug('preproc: reading file %r', filepath) try: lns = pc[filepath] except KeyError: pass else: self.lines.extend(lns) return try: lines = filter_comments(filepath) lines.append((POPFILE, '')) lines.reverse() pc[filepath] = lines # cache the lines filtered self.lines.extend(lines) except IOError: raise PreprocError("could not read the file %s" % filepath) except Exception: if Logs.verbose > 0: error("parsing %s failed" % filepath) traceback.print_exc() def start(self, node, env): debug('preproc: scanning %s (in %s)', node.name, node.parent.name) self.env = env variant = node.variant(env) bld = node.__class__.bld try: self.parse_cache = bld.parse_cache except AttributeError: bld.parse_cache = {} self.parse_cache = bld.parse_cache self.addlines(node) if env['DEFLINES']: lst = [('define', x) for x in env['DEFLINES']] lst.reverse() self.lines.extend(lst) while self.lines: (kind, line) = self.lines.pop() if kind == POPFILE: self.currentnode_stack.pop() continue try: self.process_line(kind, line) except Exception, e: if Logs.verbose: debug('preproc: line parsing failed (%s): %s %s', e, line, Utils.ex_stack()) def process_line(self, token, line): """ WARNING: a new state must be added for if* because the endif """ ve = Logs.verbose if ve: debug('preproc: line is %s - %s state is %s', token, line, self.state) state = self.state # make certain we define the state if we are about to enter in an if block if token in ['ifdef', 'ifndef', 'if']: state.append(undefined) elif token == 'endif': state.pop() # skip lines when in a dead 'if' branch, wait for the endif if not token in ['else', 'elif', 'endif']: if skipped in self.state or ignored in self.state: return if token == 'if': ret = eval_macro(tokenize(line), self.defs) if ret: state[-1] = accepted else: state[-1] = ignored elif token == 'ifdef': m = re_mac.match(line) if m and m.group(0) in self.defs: state[-1] = accepted else: state[-1] = ignored elif token == 'ifndef': m = re_mac.match(line) if m and m.group(0) in self.defs: state[-1] = ignored else: state[-1] = accepted elif token == 'include' or token == 'import': (kind, inc) = extract_include(line, self.defs) if inc in self.ban_includes: return if token == 'import': self.ban_includes.add(inc) if ve: debug('preproc: include found %s (%s) ', inc, kind) if kind == '"' or not strict_quotes: self.tryfind(inc) elif token == 'elif': if state[-1] == accepted: state[-1] = skipped elif state[-1] == ignored: if eval_macro(tokenize(line), self.defs): state[-1] = accepted elif token == 'else': if state[-1] == accepted: state[-1] = skipped elif state[-1] == ignored: state[-1] = accepted elif token == 'define': try: self.defs[define_name(line)] = line except: raise PreprocError("invalid define line %s" % line) elif token == 'undef': m = re_mac.match(line) if m and m.group(0) in self.defs: self.defs.__delitem__(m.group(0)) #print "undef %s" % name elif token == 'pragma': if re_pragma_once.match(line.lower()): self.ban_includes.add(self.curfile) def get_deps(node, env, nodepaths=[]): """ Get the dependencies using a c/c++ preprocessor, this is required for finding dependencies of the kind #include some_macro() """ gruik = c_parser(nodepaths) gruik.start(node, env) return (gruik.nodes, gruik.names) #################### dumb dependency scanner re_inc = re.compile(\ '^[ \t]*(#|%:)[ \t]*(include)[ \t]*(.*)\r*$', re.IGNORECASE | re.MULTILINE) def lines_includes(filename): code = Utils.readf(filename) if use_trigraphs: for (a, b) in trig_def: code = code.split(a).join(b) code = re_nl.sub('', code) code = re_cpp.sub(repl, code) return [(m.group(2), m.group(3)) for m in re.finditer(re_inc, code)] def get_deps_simple(node, env, nodepaths=[], defines={}): """ Get the dependencies by just looking recursively at the #include statements """ nodes = [] names = [] def find_deps(node): lst = lines_includes(node.abspath(env)) for (_, line) in lst: (t, filename) = extract_include(line, defines) if filename in names: continue if filename.endswith('.moc'): names.append(filename) found = None for n in nodepaths: if found: break found = n.find_resource(filename) if not found: if not filename in names: names.append(filename) elif not found in nodes: nodes.append(found) find_deps(node) find_deps(node) return (nodes, names)