Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 1 | # A parser for SGML, using the derived class as static DTD. |
| 2 | |
| 3 | # XXX This only supports those SGML features used by HTML. |
| 4 | |
| 5 | # XXX There should be a way to distinguish between PCDATA (parsed |
| 6 | # character data -- the normal case), RCDATA (replaceable character |
| 7 | # data -- only char and entity references and end tags are special) |
| 8 | # and CDATA (character data -- only end tags are special). |
| 9 | |
| 10 | |
| 11 | import regex |
| 12 | import string |
| 13 | |
| 14 | |
| 15 | # Regular expressions used for parsing |
| 16 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 17 | interesting = regex.compile('[&<]') |
| 18 | incomplete = regex.compile('&\([a-zA-Z][a-zA-Z0-9]*\|#[0-9]*\)?\|' |
| 19 | '<\([a-zA-Z][^<>]*\|' |
| 20 | '/\([a-zA-Z][^<>]*\)?\|' |
| 21 | '![^<>]*\)?') |
| 22 | |
| 23 | entityref = regex.compile('&\([a-zA-Z][a-zA-Z0-9]*\)[^a-zA-Z0-9]') |
| 24 | charref = regex.compile('&#\([0-9]+\)[^0-9]') |
| 25 | |
| 26 | starttagopen = regex.compile('<[>a-zA-Z]') |
| 27 | shorttagopen = regex.compile('<[a-zA-Z][a-zA-Z0-9]*/') |
| 28 | shorttag = regex.compile('<\([a-zA-Z][a-zA-Z0-9]*\)/\([^/]*\)/') |
| 29 | endtagopen = regex.compile('</[<>a-zA-Z]') |
| 30 | endbracket = regex.compile('[<>]') |
Guido van Rossum | 145b2e0 | 1995-08-04 04:22:39 +0000 | [diff] [blame] | 31 | special = regex.compile('<![^<>]*>') |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 32 | commentopen = regex.compile('<!--') |
Guido van Rossum | 145b2e0 | 1995-08-04 04:22:39 +0000 | [diff] [blame] | 33 | commentclose = regex.compile('--[ \t\n]*>') |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 34 | tagfind = regex.compile('[a-zA-Z][a-zA-Z0-9]*') |
| 35 | attrfind = regex.compile( |
Fred Drake | 09bcf8c | 1996-12-16 21:56:27 +0000 | [diff] [blame] | 36 | '[ \t\n]+\([a-zA-Z_][-.a-zA-Z_0-9]*\)' |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 37 | '\([ \t\n]*=[ \t\n]*' |
| 38 | '\(\'[^\']*\'\|"[^"]*"\|[-a-zA-Z0-9./:+*%?!()_#=~]*\)\)?') |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 39 | |
| 40 | |
| 41 | # SGML parser base class -- find tags and call handler functions. |
| 42 | # Usage: p = SGMLParser(); p.feed(data); ...; p.close(). |
| 43 | # The dtd is defined by deriving a class which defines methods |
| 44 | # with special names to handle tags: start_foo and end_foo to handle |
| 45 | # <foo> and </foo>, respectively, or do_foo to handle <foo> by itself. |
| 46 | # (Tags are converted to lower case for this purpose.) The data |
| 47 | # between tags is passed to the parser by calling self.handle_data() |
| 48 | # with some data as argument (the data may be split up in arbutrary |
| 49 | # chunks). Entity references are passed by calling |
| 50 | # self.handle_entityref() with the entity reference as argument. |
| 51 | |
| 52 | class SGMLParser: |
| 53 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 54 | # Interface -- initialize and reset this instance |
| 55 | def __init__(self, verbose=0): |
| 56 | self.verbose = verbose |
| 57 | self.reset() |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 58 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 59 | # Interface -- reset this instance. Loses all unprocessed data |
| 60 | def reset(self): |
| 61 | self.rawdata = '' |
| 62 | self.stack = [] |
| 63 | self.lasttag = '???' |
| 64 | self.nomoretags = 0 |
| 65 | self.literal = 0 |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 66 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 67 | # For derived classes only -- enter literal mode (CDATA) till EOF |
| 68 | def setnomoretags(self): |
| 69 | self.nomoretags = self.literal = 1 |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 70 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 71 | # For derived classes only -- enter literal mode (CDATA) |
| 72 | def setliteral(self, *args): |
| 73 | self.literal = 1 |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 74 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 75 | # Interface -- feed some data to the parser. Call this as |
| 76 | # often as you want, with as little or as much text as you |
| 77 | # want (may include '\n'). (This just saves the text, all the |
| 78 | # processing is done by goahead().) |
| 79 | def feed(self, data): |
| 80 | self.rawdata = self.rawdata + data |
| 81 | self.goahead(0) |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 82 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 83 | # Interface -- handle the remaining data |
| 84 | def close(self): |
| 85 | self.goahead(1) |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 86 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 87 | # Internal -- handle data as far as reasonable. May leave state |
| 88 | # and data to be processed by a subsequent call. If 'end' is |
| 89 | # true, force handling all data as if followed by EOF marker. |
| 90 | def goahead(self, end): |
| 91 | rawdata = self.rawdata |
| 92 | i = 0 |
| 93 | n = len(rawdata) |
| 94 | while i < n: |
| 95 | if self.nomoretags: |
| 96 | self.handle_data(rawdata[i:n]) |
| 97 | i = n |
| 98 | break |
| 99 | j = interesting.search(rawdata, i) |
| 100 | if j < 0: j = n |
| 101 | if i < j: self.handle_data(rawdata[i:j]) |
| 102 | i = j |
| 103 | if i == n: break |
| 104 | if rawdata[i] == '<': |
| 105 | if starttagopen.match(rawdata, i) >= 0: |
| 106 | if self.literal: |
| 107 | self.handle_data(rawdata[i]) |
| 108 | i = i+1 |
| 109 | continue |
| 110 | k = self.parse_starttag(i) |
| 111 | if k < 0: break |
| 112 | i = k |
| 113 | continue |
| 114 | if endtagopen.match(rawdata, i) >= 0: |
| 115 | k = self.parse_endtag(i) |
| 116 | if k < 0: break |
| 117 | i = k |
| 118 | self.literal = 0 |
| 119 | continue |
| 120 | if commentopen.match(rawdata, i) >= 0: |
| 121 | if self.literal: |
| 122 | self.handle_data(rawdata[i]) |
| 123 | i = i+1 |
| 124 | continue |
| 125 | k = self.parse_comment(i) |
| 126 | if k < 0: break |
| 127 | i = i+k |
| 128 | continue |
| 129 | k = special.match(rawdata, i) |
| 130 | if k >= 0: |
| 131 | if self.literal: |
| 132 | self.handle_data(rawdata[i]) |
| 133 | i = i+1 |
| 134 | continue |
| 135 | i = i+k |
| 136 | continue |
| 137 | elif rawdata[i] == '&': |
| 138 | k = charref.match(rawdata, i) |
| 139 | if k >= 0: |
| 140 | k = i+k |
| 141 | if rawdata[k-1] != ';': k = k-1 |
| 142 | name = charref.group(1) |
| 143 | self.handle_charref(name) |
| 144 | i = k |
| 145 | continue |
| 146 | k = entityref.match(rawdata, i) |
| 147 | if k >= 0: |
| 148 | k = i+k |
| 149 | if rawdata[k-1] != ';': k = k-1 |
| 150 | name = entityref.group(1) |
| 151 | self.handle_entityref(name) |
| 152 | i = k |
| 153 | continue |
| 154 | else: |
| 155 | raise RuntimeError, 'neither < nor & ??' |
| 156 | # We get here only if incomplete matches but |
| 157 | # nothing else |
| 158 | k = incomplete.match(rawdata, i) |
| 159 | if k < 0: |
| 160 | self.handle_data(rawdata[i]) |
| 161 | i = i+1 |
| 162 | continue |
| 163 | j = i+k |
| 164 | if j == n: |
| 165 | break # Really incomplete |
| 166 | self.handle_data(rawdata[i:j]) |
| 167 | i = j |
| 168 | # end while |
| 169 | if end and i < n: |
| 170 | self.handle_data(rawdata[i:n]) |
| 171 | i = n |
| 172 | self.rawdata = rawdata[i:] |
| 173 | # XXX if end: check for empty stack |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 174 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 175 | # Internal -- parse comment, return length or -1 if not terminated |
| 176 | def parse_comment(self, i): |
| 177 | rawdata = self.rawdata |
| 178 | if rawdata[i:i+4] <> '<!--': |
| 179 | raise RuntimeError, 'unexpected call to handle_comment' |
| 180 | j = commentclose.search(rawdata, i+4) |
| 181 | if j < 0: |
| 182 | return -1 |
| 183 | self.handle_comment(rawdata[i+4: j]) |
| 184 | j = j+commentclose.match(rawdata, j) |
| 185 | return j-i |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 186 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 187 | # Internal -- handle starttag, return length or -1 if not terminated |
| 188 | def parse_starttag(self, i): |
| 189 | rawdata = self.rawdata |
| 190 | if shorttagopen.match(rawdata, i) >= 0: |
| 191 | # SGML shorthand: <tag/data/ == <tag>data</tag> |
| 192 | # XXX Can data contain &... (entity or char refs)? |
| 193 | # XXX Can data contain < or > (tag characters)? |
| 194 | # XXX Can there be whitespace before the first /? |
| 195 | j = shorttag.match(rawdata, i) |
| 196 | if j < 0: |
| 197 | return -1 |
| 198 | tag, data = shorttag.group(1, 2) |
| 199 | tag = string.lower(tag) |
| 200 | self.finish_shorttag(tag, data) |
| 201 | k = i+j |
| 202 | if rawdata[k-1] == '<': |
| 203 | k = k-1 |
| 204 | return k |
| 205 | # XXX The following should skip matching quotes (' or ") |
| 206 | j = endbracket.search(rawdata, i+1) |
| 207 | if j < 0: |
| 208 | return -1 |
| 209 | # Now parse the data between i+1 and j into a tag and attrs |
| 210 | attrs = [] |
| 211 | if rawdata[i:i+2] == '<>': |
| 212 | # SGML shorthand: <> == <last open tag seen> |
| 213 | k = j |
| 214 | tag = self.lasttag |
| 215 | else: |
| 216 | k = tagfind.match(rawdata, i+1) |
| 217 | if k < 0: |
| 218 | raise RuntimeError, 'unexpected call to parse_starttag' |
| 219 | k = i+1+k |
| 220 | tag = string.lower(rawdata[i+1:k]) |
| 221 | self.lasttag = tag |
| 222 | while k < j: |
| 223 | l = attrfind.match(rawdata, k) |
| 224 | if l < 0: break |
| 225 | attrname, rest, attrvalue = attrfind.group(1, 2, 3) |
| 226 | if not rest: |
| 227 | attrvalue = attrname |
| 228 | elif attrvalue[:1] == '\'' == attrvalue[-1:] or \ |
| 229 | attrvalue[:1] == '"' == attrvalue[-1:]: |
| 230 | attrvalue = attrvalue[1:-1] |
| 231 | attrs.append((string.lower(attrname), attrvalue)) |
| 232 | k = k + l |
| 233 | if rawdata[j] == '>': |
| 234 | j = j+1 |
| 235 | self.finish_starttag(tag, attrs) |
| 236 | return j |
| 237 | |
| 238 | # Internal -- parse endtag |
| 239 | def parse_endtag(self, i): |
| 240 | rawdata = self.rawdata |
| 241 | j = endbracket.search(rawdata, i+1) |
| 242 | if j < 0: |
| 243 | return -1 |
| 244 | tag = string.lower(string.strip(rawdata[i+2:j])) |
| 245 | if rawdata[j] == '>': |
| 246 | j = j+1 |
| 247 | self.finish_endtag(tag) |
| 248 | return j |
| 249 | |
| 250 | # Internal -- finish parsing of <tag/data/ (same as <tag>data</tag>) |
| 251 | def finish_shorttag(self, tag, data): |
| 252 | self.finish_starttag(tag, []) |
| 253 | self.handle_data(data) |
| 254 | self.finish_endtag(tag) |
| 255 | |
| 256 | # Internal -- finish processing of start tag |
| 257 | # Return -1 for unknown tag, 0 for open-only tag, 1 for balanced tag |
| 258 | def finish_starttag(self, tag, attrs): |
| 259 | try: |
| 260 | method = getattr(self, 'start_' + tag) |
| 261 | except AttributeError: |
| 262 | try: |
| 263 | method = getattr(self, 'do_' + tag) |
| 264 | except AttributeError: |
| 265 | self.unknown_starttag(tag, attrs) |
| 266 | return -1 |
| 267 | else: |
| 268 | self.handle_starttag(tag, method, attrs) |
| 269 | return 0 |
| 270 | else: |
| 271 | self.stack.append(tag) |
| 272 | self.handle_starttag(tag, method, attrs) |
| 273 | return 1 |
| 274 | |
| 275 | # Internal -- finish processing of end tag |
| 276 | def finish_endtag(self, tag): |
| 277 | if not tag: |
| 278 | found = len(self.stack) - 1 |
| 279 | if found < 0: |
| 280 | self.unknown_endtag(tag) |
| 281 | return |
| 282 | else: |
| 283 | if tag not in self.stack: |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 284 | try: |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 285 | method = getattr(self, 'end_' + tag) |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 286 | except AttributeError: |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 287 | self.unknown_endtag(tag) |
| 288 | return |
| 289 | found = len(self.stack) |
| 290 | for i in range(found): |
| 291 | if self.stack[i] == tag: found = i |
| 292 | while len(self.stack) > found: |
| 293 | tag = self.stack[-1] |
| 294 | try: |
| 295 | method = getattr(self, 'end_' + tag) |
| 296 | except AttributeError: |
| 297 | method = None |
| 298 | if method: |
| 299 | self.handle_endtag(tag, method) |
| 300 | else: |
| 301 | self.unknown_endtag(tag) |
| 302 | del self.stack[-1] |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 303 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 304 | # Overridable -- handle start tag |
| 305 | def handle_starttag(self, tag, method, attrs): |
| 306 | method(attrs) |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 307 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 308 | # Overridable -- handle end tag |
| 309 | def handle_endtag(self, tag, method): |
| 310 | method() |
Guido van Rossum | efe5ac4 | 1995-06-22 18:56:36 +0000 | [diff] [blame] | 311 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 312 | # Example -- report an unbalanced </...> tag. |
| 313 | def report_unbalanced(self, tag): |
| 314 | if self.verbose: |
| 315 | print '*** Unbalanced </' + tag + '>' |
| 316 | print '*** Stack:', self.stack |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 317 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 318 | # Example -- handle character reference, no need to override |
| 319 | def handle_charref(self, name): |
| 320 | try: |
| 321 | n = string.atoi(name) |
| 322 | except string.atoi_error: |
| 323 | self.unknown_charref(name) |
| 324 | return |
| 325 | if not 0 <= n <= 255: |
| 326 | self.unknown_charref(name) |
| 327 | return |
| 328 | self.handle_data(chr(n)) |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 329 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 330 | # Definition of entities -- derived classes may override |
| 331 | entitydefs = \ |
| 332 | {'lt': '<', 'gt': '>', 'amp': '&', 'quot': '"', 'apos': '\''} |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 333 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 334 | # Example -- handle entity reference, no need to override |
| 335 | def handle_entityref(self, name): |
| 336 | table = self.entitydefs |
| 337 | if table.has_key(name): |
| 338 | self.handle_data(table[name]) |
| 339 | else: |
| 340 | self.unknown_entityref(name) |
| 341 | return |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 342 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 343 | # Example -- handle data, should be overridden |
| 344 | def handle_data(self, data): |
| 345 | pass |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 346 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 347 | # Example -- handle comment, could be overridden |
| 348 | def handle_comment(self, data): |
| 349 | pass |
| 350 | |
| 351 | # To be overridden -- handlers for unknown objects |
| 352 | def unknown_starttag(self, tag, attrs): pass |
| 353 | def unknown_endtag(self, tag): pass |
| 354 | def unknown_charref(self, ref): pass |
| 355 | def unknown_entityref(self, ref): pass |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 356 | |
| 357 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 358 | class TestSGMLParser(SGMLParser): |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 359 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 360 | def __init__(self, verbose=0): |
| 361 | self.testdata = "" |
| 362 | SGMLParser.__init__(self, verbose) |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 363 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 364 | def handle_data(self, data): |
| 365 | self.testdata = self.testdata + data |
| 366 | if len(`self.testdata`) >= 70: |
| 367 | self.flush() |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 368 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 369 | def flush(self): |
| 370 | data = self.testdata |
| 371 | if data: |
| 372 | self.testdata = "" |
| 373 | print 'data:', `data` |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 374 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 375 | def handle_comment(self, data): |
| 376 | self.flush() |
| 377 | r = `data` |
| 378 | if len(r) > 68: |
| 379 | r = r[:32] + '...' + r[-32:] |
| 380 | print 'comment:', r |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 381 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 382 | def unknown_starttag(self, tag, attrs): |
| 383 | self.flush() |
| 384 | if not attrs: |
| 385 | print 'start tag: <' + tag + '>' |
| 386 | else: |
| 387 | print 'start tag: <' + tag, |
| 388 | for name, value in attrs: |
| 389 | print name + '=' + '"' + value + '"', |
| 390 | print '>' |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 391 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 392 | def unknown_endtag(self, tag): |
| 393 | self.flush() |
| 394 | print 'end tag: </' + tag + '>' |
| 395 | |
| 396 | def unknown_entityref(self, ref): |
| 397 | self.flush() |
| 398 | print '*** unknown entity ref: &' + ref + ';' |
| 399 | |
| 400 | def unknown_charref(self, ref): |
| 401 | self.flush() |
| 402 | print '*** unknown char ref: &#' + ref + ';' |
| 403 | |
| 404 | def close(self): |
| 405 | SGMLParser.close(self) |
| 406 | self.flush() |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 407 | |
| 408 | |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 409 | def test(args = None): |
| 410 | import sys |
| 411 | |
| 412 | if not args: |
| 413 | args = sys.argv[1:] |
| 414 | |
| 415 | if args and args[0] == '-s': |
| 416 | args = args[1:] |
| 417 | klass = SGMLParser |
| 418 | else: |
| 419 | klass = TestSGMLParser |
| 420 | |
| 421 | if args: |
| 422 | file = args[0] |
| 423 | else: |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 424 | file = 'test.html' |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 425 | |
| 426 | if file == '-': |
| 427 | f = sys.stdin |
| 428 | else: |
| 429 | try: |
| 430 | f = open(file, 'r') |
| 431 | except IOError, msg: |
| 432 | print file, ":", msg |
| 433 | sys.exit(1) |
| 434 | |
| 435 | data = f.read() |
| 436 | if f is not sys.stdin: |
| 437 | f.close() |
| 438 | |
| 439 | x = klass() |
| 440 | for c in data: |
| 441 | x.feed(c) |
| 442 | x.close() |
Guido van Rossum | 7c750e1 | 1995-02-27 13:16:55 +0000 | [diff] [blame] | 443 | |
| 444 | |
Guido van Rossum | 1dba24e | 1995-03-04 22:28:49 +0000 | [diff] [blame] | 445 | if __name__ == '__main__': |
Guido van Rossum | 4876651 | 1996-03-28 18:45:04 +0000 | [diff] [blame] | 446 | test() |