1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477
|
#!/usr/bin/env python
# -*- Mode: Python; py-indent-offset: 4 -*-
# Search through a header file looking for function prototypes.
# For each prototype, generate a scheme style definition.
# GPL'ed
# Toby D. Reeves <toby@max.rl.plh.af.mil>
# Modified by James Henstridge <james@daa.com.au> to output stuff in
# Havoc's new defs format. Info on this format can be seen at:
# http://www.gnome.org/mailing-lists/archives/gtk-devel-list/2000-January/0085.shtml
import string, sys, re, types
# ------------------ Create typecodes from typenames ---------
_upperstr_pat1 = re.compile(r'([^A-Z])([A-Z])')
_upperstr_pat2 = re.compile(r'([A-Z][A-Z])([A-Z][0-9a-z])')
_upperstr_pat3 = re.compile(r'^([A-Z])([A-Z])')
def to_upper_str(name):
"""Converts a typename to the equivalent upercase and underscores
name. This is used to form the type conversion macros and enum/flag
name variables"""
name = _upperstr_pat1.sub(r'\1_\2', name)
name = _upperstr_pat2.sub(r'\1_\2', name)
name = _upperstr_pat3.sub(r'\1_\2', name, count=1)
return string.upper(name)
def typecode_prefix(typename):
"""create a typecode (eg. GTK_TYPE_WIDGET) from a typename"""
return string.replace(to_upper_str(typename), '_', '_TYPE_', 1)
def typecode_postfix(typename):
"""create a typecode (eg. GST_MEDIA_INFO_TYPE) from a typename"""
return to_upper_str(typename) + '_TYPE'
def typecode_re(regex):
"""create a typecode from a regex"""
assert regex[0] == 's'
delim = regex[1]
s, pat, sub, null = regex.split(delim)
assert null == ''
def typecode(typename):
return re.sub(pat, sub, to_upper_str(typename))
return typecode
# ------------------ Find object definitions -----------------
def strip_comments(buf):
parts = []
lastpos = 0
while 1:
pos = string.find(buf, '/*', lastpos)
if pos >= 0:
parts.append(buf[lastpos:pos])
pos = string.find(buf, '*/', pos)
if pos >= 0:
lastpos = pos + 2
else:
break
else:
parts.append(buf[lastpos:])
break
return string.join(parts, '')
obj_name_pat = "[A-Z][a-z]*[A-Z][A-Za-z0-9]*"
split_prefix_pat = re.compile('([A-Z][a-z]*)([A-Za-z0-9]+)')
def find_obj_defs(buf, objdefs=[]):
"""
Try to find object definitions in header files.
"""
# filter out comments from buffer.
buf = strip_comments(buf)
maybeobjdefs = [] # contains all possible objects from file
# first find all structures that look like they may represent a GtkObject
pat = re.compile("struct _(" + obj_name_pat + ")\s*{\s*" +
"(" + obj_name_pat + ")\s+", re.MULTILINE)
pos = 0
while pos < len(buf):
m = pat.search(buf, pos)
if not m: break
maybeobjdefs.append((m.group(1), m.group(2)))
pos = m.end()
# handle typedef struct { ... } style struct defs.
pat = re.compile("typedef struct\s+[_\w]*\s*{\s*" +
"(" + obj_name_pat + ")\s+[^}]*}\s*" +
"(" + obj_name_pat + ")\s*;", re.MULTILINE)
pos = 0
while pos < len(buf):
m = pat.search(buf, pos)
if not m: break
maybeobjdefs.append((m.group(2), m.group(2)))
pos = m.end()
# now find all structures that look like they might represent a class:
pat = re.compile("struct _(" + obj_name_pat + ")Class\s*{\s*" +
"(" + obj_name_pat + ")Class\s+", re.MULTILINE)
pos = 0
while pos < len(buf):
m = pat.search(buf, pos)
if not m: break
t = (m.group(1), m.group(2))
# if we find an object structure together with a corresponding
# class structure, then we have probably found a GtkObject subclass.
if t in maybeobjdefs:
objdefs.append(t)
pos = m.end()
pat = re.compile("typedef struct\s+[_\w]*\s*{\s*" +
"(" + obj_name_pat + ")Class\s+[^}]*}\s*" +
"(" + obj_name_pat + ")Class\s*;", re.MULTILINE)
pos = 0
while pos < len(buf):
m = pat.search(buf, pos)
if not m: break
t = (m.group(2), m.group(1))
# if we find an object structure together with a corresponding
# class structure, then we have probably found a GtkObject subclass.
if t in maybeobjdefs:
objdefs.append(t)
pos = m.end()
def sort_obj_defs(objdefs):
objdefs.sort() # not strictly needed, but looks nice
pos = 0
while pos < len(objdefs):
klass,parent = objdefs[pos]
for i in range(pos+1, len(objdefs)):
# parent below subclass ... reorder
if objdefs[i][0] == parent:
objdefs.insert(i+1, objdefs[pos])
del objdefs[pos]
break
else:
pos = pos + 1
return objdefs
def write_obj_defs(objdefs, output):
if type(output)==types.StringType:
fp=open(output,'w')
elif type(output)==types.FileType:
fp=output
else:
fp=sys.stdout
fp.write(';; -*- scheme -*-\n')
fp.write('; object definitions ...\n')
for klass, parent in objdefs:
m = split_prefix_pat.match(klass)
cmodule = None
cname = klass
if m:
cmodule = m.group(1)
cname = m.group(2)
fp.write('(define-object ' + cname + '\n')
if cmodule:
fp.write(' (in-module "' + cmodule + '")\n')
if parent:
fp.write(' (parent "' + parent + '")\n')
fp.write(' (c-name "' + klass + '")\n')
fp.write(' (gtype-id "' + typecode(klass) + '")\n')
# should do something about accessible fields
fp.write(')\n\n')
# ------------------ Find enum definitions -----------------
def find_enum_defs(buf, enums=[]):
# strip comments
# bulk comments
buf = strip_comments(buf)
buf = re.sub('\n', ' ', buf)
enum_pat = re.compile(r'enum\s*{([^}]*)}\s*([A-Z][A-Za-z0-9]*)(\s|;)')
splitter = re.compile(r'\s*,\s', re.MULTILINE)
pos = 0
while pos < len(buf):
m = enum_pat.search(buf, pos)
if not m: break
name = m.group(2)
vals = m.group(1)
isflags = string.find(vals, '<<') >= 0
entries = []
for val in splitter.split(vals):
if not string.strip(val): continue
entries.append(string.split(val)[0])
if name != 'GdkCursorType':
enums.append((name, isflags, entries))
pos = m.end()
def write_enum_defs(enums, output=None, without_gtype=0):
if type(output)==types.StringType:
fp=open(output,'w')
elif type(output)==types.FileType:
fp=output
else:
fp=sys.stdout
fp.write(';; Enumerations and flags ...\n\n')
trans = string.maketrans(string.uppercase + '_', string.lowercase + '-')
for cname, isflags, entries in enums:
name = cname
module = None
m = split_prefix_pat.match(cname)
if m:
module = m.group(1)
name = m.group(2)
if isflags:
fp.write('(define-flags ' + name + '\n')
else:
fp.write('(define-enum ' + name + '\n')
if module:
fp.write(' (in-module "' + module + '")\n')
fp.write(' (c-name "' + cname + '")\n')
if not without_gtype:
fp.write(' (gtype-id "' + typecode(cname) + '")\n')
prefix = entries[0]
for ent in entries:
# shorten prefix til we get a match ...
# and handle GDK_FONT_FONT, GDK_FONT_FONTSET case
while ent[:len(prefix)] != prefix or len(prefix) >= len(ent):
prefix = prefix[:-1]
prefix_len = len(prefix)
fp.write(' (values\n')
for ent in entries:
fp.write(' \'("%s" "%s")\n' %
(string.translate(ent[prefix_len:], trans), ent))
fp.write(' )\n')
fp.write(')\n\n')
# ------------------ Find function definitions -----------------
def clean_func(buf):
"""
Ideally would make buf have a single prototype on each line.
Actually just cuts out a good deal of junk, but leaves lines
where a regex can figure prototypes out.
"""
# bulk comments
buf = strip_comments(buf)
# compact continued lines
pat = re.compile(r"""\\\n""", re.MULTILINE)
buf=pat.sub('',buf)
# Preprocess directives
pat = re.compile(r"""^[#].*?$""", re.MULTILINE)
buf=pat.sub('',buf)
# GLib declaration braces
pat = re.compile(r"""^\s*G_(BEGIN|END)_DECLS\s*$""", re.MULTILINE)
buf=pat.sub('',buf)
#typedefs, stucts, and enums
pat = re.compile(r"""^(typedef|struct|enum)(\s|.|\n)*?;\s*""", re.MULTILINE)
buf=pat.sub('',buf)
#multiple whitespace
pat = re.compile(r"""\s+""", re.MULTILINE)
buf=pat.sub(' ',buf)
#clean up line ends
pat = re.compile(r""";\s*""", re.MULTILINE)
buf=pat.sub('\n',buf)
buf = buf.lstrip()
#associate *, &, and [] with type instead of variable
#pat=re.compile(r'\s+([*|&]+)\s*(\w+)')
pat=re.compile(r' \s+ ([*|&]+) \s* (\w+)',re.VERBOSE)
buf=pat.sub(r'\1 \2', buf)
pat=re.compile(r'\s+ (\w+) \[ \s* \]',re.VERBOSE)
buf=pat.sub(r'[] \1', buf)
# make return types that are const work.
buf = string.replace(buf, 'G_CONST_RETURN ', 'const-')
buf = string.replace(buf, 'const ', 'const-')
return buf
proto_pat=re.compile(r"""
(?P<ret>(\s|-|\w|\&|\*)+\s*) # return type
\s+ # skip whitespace
(?P<func>\w+)\s*[(] # match the function name until the opening (
\s* # skip any whitespace
(?P<args>.*?)\s*[)] # group the function arguments
""", re.IGNORECASE|re.VERBOSE)
#"""
arg_split_pat = re.compile("\s*,\s*")
def define_func(buf, fp, detect_methods):
buf=clean_func(buf)
buf=string.split(buf,'\n')
for p in buf:
if len(p)==0: continue
m=proto_pat.match(p)
if m==None:
if verbose:
sys.stderr.write('No match:|%s|\n'%p)
continue
func = m.group('func')
ret = string.join(m.group('ret').split() ,"-")
args=m.group('args')
args=arg_split_pat.split(args)
for i in range(len(args)):
spaces = string.count(args[i], ' ')
if spaces > 1:
args[i] = string.replace(args[i], ' ', '-', spaces - 1)
write_func(fp, func, ret, args, detect_methods)
get_type_pat = re.compile(r'(const-)?([A-Za-z0-9]+)\*\s+')
pointer_pat = re.compile('.*\*$')
func_new_pat = re.compile('(\w+)_new$')
def write_func(fp, name, ret, args, detect_methods):
if detect_methods and len(args) >= 1:
# methods must have at least one argument
munged_name = string.replace(name, '_', '')
m = get_type_pat.match(args[0])
if m:
obj = m.group(2)
if munged_name[:len(obj)] == string.lower(obj):
regex = string.join(map(lambda x: x+'_?',string.lower(obj)),'')
mname = re.sub(regex, '', name)
fp.write('(define-method ' + mname + '\n')
fp.write(' (of-object "' + obj + '")\n')
fp.write(' (c-name "' + name + '")\n')
if ret != 'void':
fp.write(' (return-type "' + ret + '")\n')
else:
fp.write(' (return-type "none")\n')
is_varargs = 0
has_args = len(args) > 1
for arg in args[1:]:
if arg == '...':
is_varargs = 1
elif arg in ('void', 'void '):
has_args = 0
if has_args:
fp.write(' (parameters\n')
for arg in args[1:]:
if arg != '...':
tupleArg = tuple(string.split(arg))
if len(tupleArg) == 2:
fp.write(' \'("%s" "%s")\n' % tupleArg)
fp.write(' )\n')
if is_varargs:
fp.write(' (varargs #t)\n')
fp.write(')\n\n')
return
# it is either a constructor or normal function
fp.write('(define-function ' + name + '\n')
fp.write(' (c-name "' + name + '")\n')
# Hmmm... Let's asume that a constructor function name
# ends with '_new' and it returns a pointer.
m = func_new_pat.match(name)
if pointer_pat.match(ret) and m:
cname = ''
for s in m.group(1).split ('_'):
cname += s.title()
if cname != '':
fp.write(' (is-constructor-of "' + cname + '")\n')
if ret != 'void':
fp.write(' (return-type "' + ret + '")\n')
else:
fp.write(' (return-type "none")\n')
is_varargs = 0
has_args = len(args) > 0
for arg in args:
if arg == '...':
is_varargs = 1
elif arg in ('void', 'void '):
has_args = 0
if has_args:
fp.write(' (parameters\n')
for arg in args:
if arg != '...':
tupleArg = tuple(string.split(arg))
if len(tupleArg) == 2:
fp.write(' \'("%s" "%s")\n' % tupleArg)
fp.write(' )\n')
if is_varargs:
fp.write(' (varargs #t)\n')
fp.write(')\n\n')
def write_def(input, output, detect_methods):
fp = open(input)
buf = fp.read()
fp.close()
if type(output) == types.StringType:
fp = open(output,'w')
elif type(output) == types.FileType:
fp = output
else:
fp = sys.stdout
fp.write('\n;; From %s\n\n' % input)
buf = define_func(buf, fp, detect_methods)
fp.write('\n')
# ------------------ Main function -----------------
verbose=0
typecode = typecode_prefix # default
if __name__ == '__main__':
import getopt
do_types = 0
with_c_enums = 0
do_procs = 0
header = 0
detect_methods = True
opts, args = getopt.getopt(sys.argv[1:], 'v',
['types', 'c-enums', 'procs',
'type-postfix', 'type-re=', 'all',
'with-header=', 'no-methods'])
for o, v in opts:
if o == '-v':
verbose = 1
elif o == '--all':
do_types = do_procs = 1
elif o == '--types':
do_types = 1
elif o == '--c-enums':
with_c_enums = 1
elif o == '--procs':
do_procs = 1
elif o == '--type-postfix':
typecode = typecode_postfix
elif o == '--type-re':
typecode = typecode_re(v)
elif o == '--with-header':
header = v
elif o == '--no-methods':
detect_methods = False
if not args[0:1]:
print 'Must specify at least one input file name'
sys.exit(-1)
if not (do_types or do_procs):
print 'Must say --types, --procs, or --all'
sys.exit(-1)
print ';; -*- scheme -*-'
if header: print header
# read all the object definitions in
objdefs = []
enums = []
for filename in args:
buf = open(filename).read()
find_obj_defs(buf, objdefs)
find_enum_defs(buf, enums)
objdefs = sort_obj_defs(objdefs)
if do_types:
write_enum_defs(enums,None, without_gtype = with_c_enums)
write_obj_defs(objdefs,None)
if do_procs:
for filename in args:
write_def(filename,None, detect_methods)
|