OLD | NEW |
(Empty) | |
| 1 # |
| 2 # ElementTree |
| 3 # $Id: ElementTree.py 3440 2008-07-18 14:45:01Z fredrik $ |
| 4 # |
| 5 # light-weight XML support for Python 2.3 and later. |
| 6 # |
| 7 # history (since 1.2.6): |
| 8 # 2005-11-12 fl added tostringlist/fromstringlist helpers |
| 9 # 2006-07-05 fl merged in selected changes from the 1.3 sandbox |
| 10 # 2006-07-05 fl removed support for 2.1 and earlier |
| 11 # 2007-06-21 fl added deprecation/future warnings |
| 12 # 2007-08-25 fl added doctype hook, added parser version attribute etc |
| 13 # 2007-08-26 fl added new serializer code (better namespace handling, etc) |
| 14 # 2007-08-27 fl warn for broken /tag searches on tree level |
| 15 # 2007-09-02 fl added html/text methods to serializer (experimental) |
| 16 # 2007-09-05 fl added method argument to tostring/tostringlist |
| 17 # 2007-09-06 fl improved error handling |
| 18 # 2007-09-13 fl added itertext, iterfind; assorted cleanups |
| 19 # 2007-12-15 fl added C14N hooks, copy method (experimental) |
| 20 # |
| 21 # Copyright (c) 1999-2008 by Fredrik Lundh. All rights reserved. |
| 22 # |
| 23 # fredrik@pythonware.com |
| 24 # http://www.pythonware.com |
| 25 # |
| 26 # -------------------------------------------------------------------- |
| 27 # The ElementTree toolkit is |
| 28 # |
| 29 # Copyright (c) 1999-2008 by Fredrik Lundh |
| 30 # |
| 31 # By obtaining, using, and/or copying this software and/or its |
| 32 # associated documentation, you agree that you have read, understood, |
| 33 # and will comply with the following terms and conditions: |
| 34 # |
| 35 # Permission to use, copy, modify, and distribute this software and |
| 36 # its associated documentation for any purpose and without fee is |
| 37 # hereby granted, provided that the above copyright notice appears in |
| 38 # all copies, and that both that copyright notice and this permission |
| 39 # notice appear in supporting documentation, and that the name of |
| 40 # Secret Labs AB or the author not be used in advertising or publicity |
| 41 # pertaining to distribution of the software without specific, written |
| 42 # prior permission. |
| 43 # |
| 44 # SECRET LABS AB AND THE AUTHOR DISCLAIMS ALL WARRANTIES WITH REGARD |
| 45 # TO THIS SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANT- |
| 46 # ABILITY AND FITNESS. IN NO EVENT SHALL SECRET LABS AB OR THE AUTHOR |
| 47 # BE LIABLE FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY |
| 48 # DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, |
| 49 # WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS |
| 50 # ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE |
| 51 # OF THIS SOFTWARE. |
| 52 # -------------------------------------------------------------------- |
| 53 |
| 54 # Licensed to PSF under a Contributor Agreement. |
| 55 # See http://www.python.org/psf/license for licensing details. |
| 56 |
| 57 __all__ = [ |
| 58 # public symbols |
| 59 "Comment", |
| 60 "dump", |
| 61 "Element", "ElementTree", |
| 62 "fromstring", "fromstringlist", |
| 63 "iselement", "iterparse", |
| 64 "parse", "ParseError", |
| 65 "PI", "ProcessingInstruction", |
| 66 "QName", |
| 67 "SubElement", |
| 68 "tostring", "tostringlist", |
| 69 "TreeBuilder", |
| 70 "VERSION", |
| 71 "XML", |
| 72 "XMLParser", "XMLTreeBuilder", |
| 73 ] |
| 74 |
| 75 VERSION = "1.3.0" |
| 76 |
| 77 ## |
| 78 # The <b>Element</b> type is a flexible container object, designed to |
| 79 # store hierarchical data structures in memory. The type can be |
| 80 # described as a cross between a list and a dictionary. |
| 81 # <p> |
| 82 # Each element has a number of properties associated with it: |
| 83 # <ul> |
| 84 # <li>a <i>tag</i>. This is a string identifying what kind of data |
| 85 # this element represents (the element type, in other words).</li> |
| 86 # <li>a number of <i>attributes</i>, stored in a Python dictionary.</li> |
| 87 # <li>a <i>text</i> string.</li> |
| 88 # <li>an optional <i>tail</i> string.</li> |
| 89 # <li>a number of <i>child elements</i>, stored in a Python sequence</li> |
| 90 # </ul> |
| 91 # |
| 92 # To create an element instance, use the {@link #Element} constructor |
| 93 # or the {@link #SubElement} factory function. |
| 94 # <p> |
| 95 # The {@link #ElementTree} class can be used to wrap an element |
| 96 # structure, and convert it from and to XML. |
| 97 ## |
| 98 |
| 99 import sys |
| 100 import re |
| 101 import warnings |
| 102 |
| 103 |
| 104 class _SimpleElementPath(object): |
| 105 # emulate pre-1.2 find/findtext/findall behaviour |
| 106 def find(self, element, tag, namespaces=None): |
| 107 for elem in element: |
| 108 if elem.tag == tag: |
| 109 return elem |
| 110 return None |
| 111 def findtext(self, element, tag, default=None, namespaces=None): |
| 112 elem = self.find(element, tag) |
| 113 if elem is None: |
| 114 return default |
| 115 return elem.text or "" |
| 116 def iterfind(self, element, tag, namespaces=None): |
| 117 if tag[:3] == ".//": |
| 118 for elem in element.iter(tag[3:]): |
| 119 yield elem |
| 120 for elem in element: |
| 121 if elem.tag == tag: |
| 122 yield elem |
| 123 def findall(self, element, tag, namespaces=None): |
| 124 return list(self.iterfind(element, tag, namespaces)) |
| 125 |
| 126 try: |
| 127 from . import ElementPath |
| 128 except ImportError: |
| 129 ElementPath = _SimpleElementPath() |
| 130 |
| 131 ## |
| 132 # Parser error. This is a subclass of <b>SyntaxError</b>. |
| 133 # <p> |
| 134 # In addition to the exception value, an exception instance contains a |
| 135 # specific exception code in the <b>code</b> attribute, and the line and |
| 136 # column of the error in the <b>position</b> attribute. |
| 137 |
| 138 class ParseError(SyntaxError): |
| 139 pass |
| 140 |
| 141 # -------------------------------------------------------------------- |
| 142 |
| 143 ## |
| 144 # Checks if an object appears to be a valid element object. |
| 145 # |
| 146 # @param An element instance. |
| 147 # @return A true value if this is an element object. |
| 148 # @defreturn flag |
| 149 |
| 150 def iselement(element): |
| 151 # FIXME: not sure about this; might be a better idea to look |
| 152 # for tag/attrib/text attributes |
| 153 return isinstance(element, Element) or hasattr(element, "tag") |
| 154 |
| 155 ## |
| 156 # Element class. This class defines the Element interface, and |
| 157 # provides a reference implementation of this interface. |
| 158 # <p> |
| 159 # The element name, attribute names, and attribute values can be |
| 160 # either ASCII strings (ordinary Python strings containing only 7-bit |
| 161 # ASCII characters) or Unicode strings. |
| 162 # |
| 163 # @param tag The element name. |
| 164 # @param attrib An optional dictionary, containing element attributes. |
| 165 # @param **extra Additional attributes, given as keyword arguments. |
| 166 # @see Element |
| 167 # @see SubElement |
| 168 # @see Comment |
| 169 # @see ProcessingInstruction |
| 170 |
| 171 class Element(object): |
| 172 # <tag attrib>text<child/>...</tag>tail |
| 173 |
| 174 ## |
| 175 # (Attribute) Element tag. |
| 176 |
| 177 tag = None |
| 178 |
| 179 ## |
| 180 # (Attribute) Element attribute dictionary. Where possible, use |
| 181 # {@link #Element.get}, |
| 182 # {@link #Element.set}, |
| 183 # {@link #Element.keys}, and |
| 184 # {@link #Element.items} to access |
| 185 # element attributes. |
| 186 |
| 187 attrib = None |
| 188 |
| 189 ## |
| 190 # (Attribute) Text before first subelement. This is either a |
| 191 # string or the value None. Note that if there was no text, this |
| 192 # attribute may be either None or an empty string, depending on |
| 193 # the parser. |
| 194 |
| 195 text = None |
| 196 |
| 197 ## |
| 198 # (Attribute) Text after this element's end tag, but before the |
| 199 # next sibling element's start tag. This is either a string or |
| 200 # the value None. Note that if there was no text, this attribute |
| 201 # may be either None or an empty string, depending on the parser. |
| 202 |
| 203 tail = None # text after end tag, if any |
| 204 |
| 205 # constructor |
| 206 |
| 207 def __init__(self, tag, attrib={}, **extra): |
| 208 attrib = attrib.copy() |
| 209 attrib.update(extra) |
| 210 self.tag = tag |
| 211 self.attrib = attrib |
| 212 self._children = [] |
| 213 |
| 214 def __repr__(self): |
| 215 return "<Element %s at 0x%x>" % (repr(self.tag), id(self)) |
| 216 |
| 217 ## |
| 218 # Creates a new element object of the same type as this element. |
| 219 # |
| 220 # @param tag Element tag. |
| 221 # @param attrib Element attributes, given as a dictionary. |
| 222 # @return A new element instance. |
| 223 |
| 224 def makeelement(self, tag, attrib): |
| 225 return self.__class__(tag, attrib) |
| 226 |
| 227 ## |
| 228 # (Experimental) Copies the current element. This creates a |
| 229 # shallow copy; subelements will be shared with the original tree. |
| 230 # |
| 231 # @return A new element instance. |
| 232 |
| 233 def copy(self): |
| 234 elem = self.makeelement(self.tag, self.attrib) |
| 235 elem.text = self.text |
| 236 elem.tail = self.tail |
| 237 elem[:] = self |
| 238 return elem |
| 239 |
| 240 ## |
| 241 # Returns the number of subelements. Note that this only counts |
| 242 # full elements; to check if there's any content in an element, you |
| 243 # have to check both the length and the <b>text</b> attribute. |
| 244 # |
| 245 # @return The number of subelements. |
| 246 |
| 247 def __len__(self): |
| 248 return len(self._children) |
| 249 |
| 250 def __nonzero__(self): |
| 251 warnings.warn( |
| 252 "The behavior of this method will change in future versions. " |
| 253 "Use specific 'len(elem)' or 'elem is not None' test instead.", |
| 254 FutureWarning, stacklevel=2 |
| 255 ) |
| 256 return len(self._children) != 0 # emulate old behaviour, for now |
| 257 |
| 258 ## |
| 259 # Returns the given subelement, by index. |
| 260 # |
| 261 # @param index What subelement to return. |
| 262 # @return The given subelement. |
| 263 # @exception IndexError If the given element does not exist. |
| 264 |
| 265 def __getitem__(self, index): |
| 266 return self._children[index] |
| 267 |
| 268 ## |
| 269 # Replaces the given subelement, by index. |
| 270 # |
| 271 # @param index What subelement to replace. |
| 272 # @param element The new element value. |
| 273 # @exception IndexError If the given element does not exist. |
| 274 |
| 275 def __setitem__(self, index, element): |
| 276 # if isinstance(index, slice): |
| 277 # for elt in element: |
| 278 # assert iselement(elt) |
| 279 # else: |
| 280 # assert iselement(element) |
| 281 self._children[index] = element |
| 282 |
| 283 ## |
| 284 # Deletes the given subelement, by index. |
| 285 # |
| 286 # @param index What subelement to delete. |
| 287 # @exception IndexError If the given element does not exist. |
| 288 |
| 289 def __delitem__(self, index): |
| 290 del self._children[index] |
| 291 |
| 292 ## |
| 293 # Adds a subelement to the end of this element. In document order, |
| 294 # the new element will appear after the last existing subelement (or |
| 295 # directly after the text, if it's the first subelement), but before |
| 296 # the end tag for this element. |
| 297 # |
| 298 # @param element The element to add. |
| 299 |
| 300 def append(self, element): |
| 301 # assert iselement(element) |
| 302 self._children.append(element) |
| 303 |
| 304 ## |
| 305 # Appends subelements from a sequence. |
| 306 # |
| 307 # @param elements A sequence object with zero or more elements. |
| 308 # @since 1.3 |
| 309 |
| 310 def extend(self, elements): |
| 311 # for element in elements: |
| 312 # assert iselement(element) |
| 313 self._children.extend(elements) |
| 314 |
| 315 ## |
| 316 # Inserts a subelement at the given position in this element. |
| 317 # |
| 318 # @param index Where to insert the new subelement. |
| 319 |
| 320 def insert(self, index, element): |
| 321 # assert iselement(element) |
| 322 self._children.insert(index, element) |
| 323 |
| 324 ## |
| 325 # Removes a matching subelement. Unlike the <b>find</b> methods, |
| 326 # this method compares elements based on identity, not on tag |
| 327 # value or contents. To remove subelements by other means, the |
| 328 # easiest way is often to use a list comprehension to select what |
| 329 # elements to keep, and use slice assignment to update the parent |
| 330 # element. |
| 331 # |
| 332 # @param element What element to remove. |
| 333 # @exception ValueError If a matching element could not be found. |
| 334 |
| 335 def remove(self, element): |
| 336 # assert iselement(element) |
| 337 self._children.remove(element) |
| 338 |
| 339 ## |
| 340 # (Deprecated) Returns all subelements. The elements are returned |
| 341 # in document order. |
| 342 # |
| 343 # @return A list of subelements. |
| 344 # @defreturn list of Element instances |
| 345 |
| 346 def getchildren(self): |
| 347 warnings.warn( |
| 348 "This method will be removed in future versions. " |
| 349 "Use 'list(elem)' or iteration over elem instead.", |
| 350 DeprecationWarning, stacklevel=2 |
| 351 ) |
| 352 return self._children |
| 353 |
| 354 ## |
| 355 # Finds the first matching subelement, by tag name or path. |
| 356 # |
| 357 # @param path What element to look for. |
| 358 # @keyparam namespaces Optional namespace prefix map. |
| 359 # @return The first matching element, or None if no element was found. |
| 360 # @defreturn Element or None |
| 361 |
| 362 def find(self, path, namespaces=None): |
| 363 return ElementPath.find(self, path, namespaces) |
| 364 |
| 365 ## |
| 366 # Finds text for the first matching subelement, by tag name or path. |
| 367 # |
| 368 # @param path What element to look for. |
| 369 # @param default What to return if the element was not found. |
| 370 # @keyparam namespaces Optional namespace prefix map. |
| 371 # @return The text content of the first matching element, or the |
| 372 # default value no element was found. Note that if the element |
| 373 # is found, but has no text content, this method returns an |
| 374 # empty string. |
| 375 # @defreturn string |
| 376 |
| 377 def findtext(self, path, default=None, namespaces=None): |
| 378 return ElementPath.findtext(self, path, default, namespaces) |
| 379 |
| 380 ## |
| 381 # Finds all matching subelements, by tag name or path. |
| 382 # |
| 383 # @param path What element to look for. |
| 384 # @keyparam namespaces Optional namespace prefix map. |
| 385 # @return A list or other sequence containing all matching elements, |
| 386 # in document order. |
| 387 # @defreturn list of Element instances |
| 388 |
| 389 def findall(self, path, namespaces=None): |
| 390 return ElementPath.findall(self, path, namespaces) |
| 391 |
| 392 ## |
| 393 # Finds all matching subelements, by tag name or path. |
| 394 # |
| 395 # @param path What element to look for. |
| 396 # @keyparam namespaces Optional namespace prefix map. |
| 397 # @return An iterator or sequence containing all matching elements, |
| 398 # in document order. |
| 399 # @defreturn a generated sequence of Element instances |
| 400 |
| 401 def iterfind(self, path, namespaces=None): |
| 402 return ElementPath.iterfind(self, path, namespaces) |
| 403 |
| 404 ## |
| 405 # Resets an element. This function removes all subelements, clears |
| 406 # all attributes, and sets the <b>text</b> and <b>tail</b> attributes |
| 407 # to None. |
| 408 |
| 409 def clear(self): |
| 410 self.attrib.clear() |
| 411 self._children = [] |
| 412 self.text = self.tail = None |
| 413 |
| 414 ## |
| 415 # Gets an element attribute. Equivalent to <b>attrib.get</b>, but |
| 416 # some implementations may handle this a bit more efficiently. |
| 417 # |
| 418 # @param key What attribute to look for. |
| 419 # @param default What to return if the attribute was not found. |
| 420 # @return The attribute value, or the default value, if the |
| 421 # attribute was not found. |
| 422 # @defreturn string or None |
| 423 |
| 424 def get(self, key, default=None): |
| 425 return self.attrib.get(key, default) |
| 426 |
| 427 ## |
| 428 # Sets an element attribute. Equivalent to <b>attrib[key] = value</b>, |
| 429 # but some implementations may handle this a bit more efficiently. |
| 430 # |
| 431 # @param key What attribute to set. |
| 432 # @param value The attribute value. |
| 433 |
| 434 def set(self, key, value): |
| 435 self.attrib[key] = value |
| 436 |
| 437 ## |
| 438 # Gets a list of attribute names. The names are returned in an |
| 439 # arbitrary order (just like for an ordinary Python dictionary). |
| 440 # Equivalent to <b>attrib.keys()</b>. |
| 441 # |
| 442 # @return A list of element attribute names. |
| 443 # @defreturn list of strings |
| 444 |
| 445 def keys(self): |
| 446 return self.attrib.keys() |
| 447 |
| 448 ## |
| 449 # Gets element attributes, as a sequence. The attributes are |
| 450 # returned in an arbitrary order. Equivalent to <b>attrib.items()</b>. |
| 451 # |
| 452 # @return A list of (name, value) tuples for all attributes. |
| 453 # @defreturn list of (string, string) tuples |
| 454 |
| 455 def items(self): |
| 456 return self.attrib.items() |
| 457 |
| 458 ## |
| 459 # Creates a tree iterator. The iterator loops over this element |
| 460 # and all subelements, in document order, and returns all elements |
| 461 # with a matching tag. |
| 462 # <p> |
| 463 # If the tree structure is modified during iteration, new or removed |
| 464 # elements may or may not be included. To get a stable set, use the |
| 465 # list() function on the iterator, and loop over the resulting list. |
| 466 # |
| 467 # @param tag What tags to look for (default is to return all elements). |
| 468 # @return An iterator containing all the matching elements. |
| 469 # @defreturn iterator |
| 470 |
| 471 def iter(self, tag=None): |
| 472 if tag == "*": |
| 473 tag = None |
| 474 if tag is None or self.tag == tag: |
| 475 yield self |
| 476 for e in self._children: |
| 477 for e in e.iter(tag): |
| 478 yield e |
| 479 |
| 480 # compatibility |
| 481 def getiterator(self, tag=None): |
| 482 # Change for a DeprecationWarning in 1.4 |
| 483 warnings.warn( |
| 484 "This method will be removed in future versions. " |
| 485 "Use 'elem.iter()' or 'list(elem.iter())' instead.", |
| 486 PendingDeprecationWarning, stacklevel=2 |
| 487 ) |
| 488 return list(self.iter(tag)) |
| 489 |
| 490 ## |
| 491 # Creates a text iterator. The iterator loops over this element |
| 492 # and all subelements, in document order, and returns all inner |
| 493 # text. |
| 494 # |
| 495 # @return An iterator containing all inner text. |
| 496 # @defreturn iterator |
| 497 |
| 498 def itertext(self): |
| 499 tag = self.tag |
| 500 if not isinstance(tag, basestring) and tag is not None: |
| 501 return |
| 502 if self.text: |
| 503 yield self.text |
| 504 for e in self: |
| 505 for s in e.itertext(): |
| 506 yield s |
| 507 if e.tail: |
| 508 yield e.tail |
| 509 |
| 510 # compatibility |
| 511 _Element = _ElementInterface = Element |
| 512 |
| 513 ## |
| 514 # Subelement factory. This function creates an element instance, and |
| 515 # appends it to an existing element. |
| 516 # <p> |
| 517 # The element name, attribute names, and attribute values can be |
| 518 # either 8-bit ASCII strings or Unicode strings. |
| 519 # |
| 520 # @param parent The parent element. |
| 521 # @param tag The subelement name. |
| 522 # @param attrib An optional dictionary, containing element attributes. |
| 523 # @param **extra Additional attributes, given as keyword arguments. |
| 524 # @return An element instance. |
| 525 # @defreturn Element |
| 526 |
| 527 def SubElement(parent, tag, attrib={}, **extra): |
| 528 attrib = attrib.copy() |
| 529 attrib.update(extra) |
| 530 element = parent.makeelement(tag, attrib) |
| 531 parent.append(element) |
| 532 return element |
| 533 |
| 534 ## |
| 535 # Comment element factory. This factory function creates a special |
| 536 # element that will be serialized as an XML comment by the standard |
| 537 # serializer. |
| 538 # <p> |
| 539 # The comment string can be either an 8-bit ASCII string or a Unicode |
| 540 # string. |
| 541 # |
| 542 # @param text A string containing the comment string. |
| 543 # @return An element instance, representing a comment. |
| 544 # @defreturn Element |
| 545 |
| 546 def Comment(text=None): |
| 547 element = Element(Comment) |
| 548 element.text = text |
| 549 return element |
| 550 |
| 551 ## |
| 552 # PI element factory. This factory function creates a special element |
| 553 # that will be serialized as an XML processing instruction by the standard |
| 554 # serializer. |
| 555 # |
| 556 # @param target A string containing the PI target. |
| 557 # @param text A string containing the PI contents, if any. |
| 558 # @return An element instance, representing a PI. |
| 559 # @defreturn Element |
| 560 |
| 561 def ProcessingInstruction(target, text=None): |
| 562 element = Element(ProcessingInstruction) |
| 563 element.text = target |
| 564 if text: |
| 565 element.text = element.text + " " + text |
| 566 return element |
| 567 |
| 568 PI = ProcessingInstruction |
| 569 |
| 570 ## |
| 571 # QName wrapper. This can be used to wrap a QName attribute value, in |
| 572 # order to get proper namespace handling on output. |
| 573 # |
| 574 # @param text A string containing the QName value, in the form {uri}local, |
| 575 # or, if the tag argument is given, the URI part of a QName. |
| 576 # @param tag Optional tag. If given, the first argument is interpreted as |
| 577 # an URI, and this argument is interpreted as a local name. |
| 578 # @return An opaque object, representing the QName. |
| 579 |
| 580 class QName(object): |
| 581 def __init__(self, text_or_uri, tag=None): |
| 582 if tag: |
| 583 text_or_uri = "{%s}%s" % (text_or_uri, tag) |
| 584 self.text = text_or_uri |
| 585 def __str__(self): |
| 586 return self.text |
| 587 def __hash__(self): |
| 588 return hash(self.text) |
| 589 def __cmp__(self, other): |
| 590 if isinstance(other, QName): |
| 591 return cmp(self.text, other.text) |
| 592 return cmp(self.text, other) |
| 593 |
| 594 # -------------------------------------------------------------------- |
| 595 |
| 596 ## |
| 597 # ElementTree wrapper class. This class represents an entire element |
| 598 # hierarchy, and adds some extra support for serialization to and from |
| 599 # standard XML. |
| 600 # |
| 601 # @param element Optional root element. |
| 602 # @keyparam file Optional file handle or file name. If given, the |
| 603 # tree is initialized with the contents of this XML file. |
| 604 |
| 605 class ElementTree(object): |
| 606 |
| 607 def __init__(self, element=None, file=None): |
| 608 # assert element is None or iselement(element) |
| 609 self._root = element # first node |
| 610 if file: |
| 611 self.parse(file) |
| 612 |
| 613 ## |
| 614 # Gets the root element for this tree. |
| 615 # |
| 616 # @return An element instance. |
| 617 # @defreturn Element |
| 618 |
| 619 def getroot(self): |
| 620 return self._root |
| 621 |
| 622 ## |
| 623 # Replaces the root element for this tree. This discards the |
| 624 # current contents of the tree, and replaces it with the given |
| 625 # element. Use with care. |
| 626 # |
| 627 # @param element An element instance. |
| 628 |
| 629 def _setroot(self, element): |
| 630 # assert iselement(element) |
| 631 self._root = element |
| 632 |
| 633 ## |
| 634 # Loads an external XML document into this element tree. |
| 635 # |
| 636 # @param source A file name or file object. If a file object is |
| 637 # given, it only has to implement a <b>read(n)</b> method. |
| 638 # @keyparam parser An optional parser instance. If not given, the |
| 639 # standard {@link XMLParser} parser is used. |
| 640 # @return The document root element. |
| 641 # @defreturn Element |
| 642 # @exception ParseError If the parser fails to parse the document. |
| 643 |
| 644 def parse(self, source, parser=None): |
| 645 close_source = False |
| 646 if not hasattr(source, "read"): |
| 647 source = open(source, "rb") |
| 648 close_source = True |
| 649 try: |
| 650 if not parser: |
| 651 parser = XMLParser(target=TreeBuilder()) |
| 652 while 1: |
| 653 data = source.read(65536) |
| 654 if not data: |
| 655 break |
| 656 parser.feed(data) |
| 657 self._root = parser.close() |
| 658 return self._root |
| 659 finally: |
| 660 if close_source: |
| 661 source.close() |
| 662 |
| 663 ## |
| 664 # Creates a tree iterator for the root element. The iterator loops |
| 665 # over all elements in this tree, in document order. |
| 666 # |
| 667 # @param tag What tags to look for (default is to return all elements) |
| 668 # @return An iterator. |
| 669 # @defreturn iterator |
| 670 |
| 671 def iter(self, tag=None): |
| 672 # assert self._root is not None |
| 673 return self._root.iter(tag) |
| 674 |
| 675 # compatibility |
| 676 def getiterator(self, tag=None): |
| 677 # Change for a DeprecationWarning in 1.4 |
| 678 warnings.warn( |
| 679 "This method will be removed in future versions. " |
| 680 "Use 'tree.iter()' or 'list(tree.iter())' instead.", |
| 681 PendingDeprecationWarning, stacklevel=2 |
| 682 ) |
| 683 return list(self.iter(tag)) |
| 684 |
| 685 ## |
| 686 # Finds the first toplevel element with given tag. |
| 687 # Same as getroot().find(path). |
| 688 # |
| 689 # @param path What element to look for. |
| 690 # @keyparam namespaces Optional namespace prefix map. |
| 691 # @return The first matching element, or None if no element was found. |
| 692 # @defreturn Element or None |
| 693 |
| 694 def find(self, path, namespaces=None): |
| 695 # assert self._root is not None |
| 696 if path[:1] == "/": |
| 697 path = "." + path |
| 698 warnings.warn( |
| 699 "This search is broken in 1.3 and earlier, and will be " |
| 700 "fixed in a future version. If you rely on the current " |
| 701 "behaviour, change it to %r" % path, |
| 702 FutureWarning, stacklevel=2 |
| 703 ) |
| 704 return self._root.find(path, namespaces) |
| 705 |
| 706 ## |
| 707 # Finds the element text for the first toplevel element with given |
| 708 # tag. Same as getroot().findtext(path). |
| 709 # |
| 710 # @param path What toplevel element to look for. |
| 711 # @param default What to return if the element was not found. |
| 712 # @keyparam namespaces Optional namespace prefix map. |
| 713 # @return The text content of the first matching element, or the |
| 714 # default value no element was found. Note that if the element |
| 715 # is found, but has no text content, this method returns an |
| 716 # empty string. |
| 717 # @defreturn string |
| 718 |
| 719 def findtext(self, path, default=None, namespaces=None): |
| 720 # assert self._root is not None |
| 721 if path[:1] == "/": |
| 722 path = "." + path |
| 723 warnings.warn( |
| 724 "This search is broken in 1.3 and earlier, and will be " |
| 725 "fixed in a future version. If you rely on the current " |
| 726 "behaviour, change it to %r" % path, |
| 727 FutureWarning, stacklevel=2 |
| 728 ) |
| 729 return self._root.findtext(path, default, namespaces) |
| 730 |
| 731 ## |
| 732 # Finds all toplevel elements with the given tag. |
| 733 # Same as getroot().findall(path). |
| 734 # |
| 735 # @param path What element to look for. |
| 736 # @keyparam namespaces Optional namespace prefix map. |
| 737 # @return A list or iterator containing all matching elements, |
| 738 # in document order. |
| 739 # @defreturn list of Element instances |
| 740 |
| 741 def findall(self, path, namespaces=None): |
| 742 # assert self._root is not None |
| 743 if path[:1] == "/": |
| 744 path = "." + path |
| 745 warnings.warn( |
| 746 "This search is broken in 1.3 and earlier, and will be " |
| 747 "fixed in a future version. If you rely on the current " |
| 748 "behaviour, change it to %r" % path, |
| 749 FutureWarning, stacklevel=2 |
| 750 ) |
| 751 return self._root.findall(path, namespaces) |
| 752 |
| 753 ## |
| 754 # Finds all matching subelements, by tag name or path. |
| 755 # Same as getroot().iterfind(path). |
| 756 # |
| 757 # @param path What element to look for. |
| 758 # @keyparam namespaces Optional namespace prefix map. |
| 759 # @return An iterator or sequence containing all matching elements, |
| 760 # in document order. |
| 761 # @defreturn a generated sequence of Element instances |
| 762 |
| 763 def iterfind(self, path, namespaces=None): |
| 764 # assert self._root is not None |
| 765 if path[:1] == "/": |
| 766 path = "." + path |
| 767 warnings.warn( |
| 768 "This search is broken in 1.3 and earlier, and will be " |
| 769 "fixed in a future version. If you rely on the current " |
| 770 "behaviour, change it to %r" % path, |
| 771 FutureWarning, stacklevel=2 |
| 772 ) |
| 773 return self._root.iterfind(path, namespaces) |
| 774 |
| 775 ## |
| 776 # Writes the element tree to a file, as XML. |
| 777 # |
| 778 # @def write(file, **options) |
| 779 # @param file A file name, or a file object opened for writing. |
| 780 # @param **options Options, given as keyword arguments. |
| 781 # @keyparam encoding Optional output encoding (default is US-ASCII). |
| 782 # @keyparam method Optional output method ("xml", "html", "text" or |
| 783 # "c14n"; default is "xml"). |
| 784 # @keyparam xml_declaration Controls if an XML declaration should |
| 785 # be added to the file. Use False for never, True for always, |
| 786 # None for only if not US-ASCII or UTF-8. None is default. |
| 787 |
| 788 def write(self, file_or_filename, |
| 789 # keyword arguments |
| 790 encoding=None, |
| 791 xml_declaration=None, |
| 792 default_namespace=None, |
| 793 method=None): |
| 794 # assert self._root is not None |
| 795 if not method: |
| 796 method = "xml" |
| 797 elif method not in _serialize: |
| 798 # FIXME: raise an ImportError for c14n if ElementC14N is missing? |
| 799 raise ValueError("unknown method %r" % method) |
| 800 if hasattr(file_or_filename, "write"): |
| 801 file = file_or_filename |
| 802 else: |
| 803 file = open(file_or_filename, "wb") |
| 804 write = file.write |
| 805 if not encoding: |
| 806 if method == "c14n": |
| 807 encoding = "utf-8" |
| 808 else: |
| 809 encoding = "us-ascii" |
| 810 elif xml_declaration or (xml_declaration is None and |
| 811 encoding not in ("utf-8", "us-ascii")): |
| 812 if method == "xml": |
| 813 write("<?xml version='1.0' encoding='%s'?>\n" % encoding) |
| 814 if method == "text": |
| 815 _serialize_text(write, self._root, encoding) |
| 816 else: |
| 817 qnames, namespaces = _namespaces( |
| 818 self._root, encoding, default_namespace |
| 819 ) |
| 820 serialize = _serialize[method] |
| 821 serialize(write, self._root, encoding, qnames, namespaces) |
| 822 if file_or_filename is not file: |
| 823 file.close() |
| 824 |
| 825 def write_c14n(self, file): |
| 826 # lxml.etree compatibility. use output method instead |
| 827 return self.write(file, method="c14n") |
| 828 |
| 829 # -------------------------------------------------------------------- |
| 830 # serialization support |
| 831 |
| 832 def _namespaces(elem, encoding, default_namespace=None): |
| 833 # identify namespaces used in this tree |
| 834 |
| 835 # maps qnames to *encoded* prefix:local names |
| 836 qnames = {None: None} |
| 837 |
| 838 # maps uri:s to prefixes |
| 839 namespaces = {} |
| 840 if default_namespace: |
| 841 namespaces[default_namespace] = "" |
| 842 |
| 843 def encode(text): |
| 844 return text.encode(encoding) |
| 845 |
| 846 def add_qname(qname): |
| 847 # calculate serialized qname representation |
| 848 try: |
| 849 if qname[:1] == "{": |
| 850 uri, tag = qname[1:].rsplit("}", 1) |
| 851 prefix = namespaces.get(uri) |
| 852 if prefix is None: |
| 853 prefix = _namespace_map.get(uri) |
| 854 if prefix is None: |
| 855 prefix = "ns%d" % len(namespaces) |
| 856 if prefix != "xml": |
| 857 namespaces[uri] = prefix |
| 858 if prefix: |
| 859 qnames[qname] = encode("%s:%s" % (prefix, tag)) |
| 860 else: |
| 861 qnames[qname] = encode(tag) # default element |
| 862 else: |
| 863 if default_namespace: |
| 864 # FIXME: can this be handled in XML 1.0? |
| 865 raise ValueError( |
| 866 "cannot use non-qualified names with " |
| 867 "default_namespace option" |
| 868 ) |
| 869 qnames[qname] = encode(qname) |
| 870 except TypeError: |
| 871 _raise_serialization_error(qname) |
| 872 |
| 873 # populate qname and namespaces table |
| 874 try: |
| 875 iterate = elem.iter |
| 876 except AttributeError: |
| 877 iterate = elem.getiterator # cET compatibility |
| 878 for elem in iterate(): |
| 879 tag = elem.tag |
| 880 if isinstance(tag, QName): |
| 881 if tag.text not in qnames: |
| 882 add_qname(tag.text) |
| 883 elif isinstance(tag, basestring): |
| 884 if tag not in qnames: |
| 885 add_qname(tag) |
| 886 elif tag is not None and tag is not Comment and tag is not PI: |
| 887 _raise_serialization_error(tag) |
| 888 for key, value in elem.items(): |
| 889 if isinstance(key, QName): |
| 890 key = key.text |
| 891 if key not in qnames: |
| 892 add_qname(key) |
| 893 if isinstance(value, QName) and value.text not in qnames: |
| 894 add_qname(value.text) |
| 895 text = elem.text |
| 896 if isinstance(text, QName) and text.text not in qnames: |
| 897 add_qname(text.text) |
| 898 return qnames, namespaces |
| 899 |
| 900 def _serialize_xml(write, elem, encoding, qnames, namespaces): |
| 901 tag = elem.tag |
| 902 text = elem.text |
| 903 if tag is Comment: |
| 904 write("<!--%s-->" % _encode(text, encoding)) |
| 905 elif tag is ProcessingInstruction: |
| 906 write("<?%s?>" % _encode(text, encoding)) |
| 907 else: |
| 908 tag = qnames[tag] |
| 909 if tag is None: |
| 910 if text: |
| 911 write(_escape_cdata(text, encoding)) |
| 912 for e in elem: |
| 913 _serialize_xml(write, e, encoding, qnames, None) |
| 914 else: |
| 915 write("<" + tag) |
| 916 items = elem.items() |
| 917 if items or namespaces: |
| 918 if namespaces: |
| 919 for v, k in sorted(namespaces.items(), |
| 920 key=lambda x: x[1]): # sort on prefix |
| 921 if k: |
| 922 k = ":" + k |
| 923 write(" xmlns%s=\"%s\"" % ( |
| 924 k.encode(encoding), |
| 925 _escape_attrib(v, encoding) |
| 926 )) |
| 927 for k, v in sorted(items): # lexical order |
| 928 if isinstance(k, QName): |
| 929 k = k.text |
| 930 if isinstance(v, QName): |
| 931 v = qnames[v.text] |
| 932 else: |
| 933 v = _escape_attrib(v, encoding) |
| 934 write(" %s=\"%s\"" % (qnames[k], v)) |
| 935 if text or len(elem): |
| 936 write(">") |
| 937 if text: |
| 938 write(_escape_cdata(text, encoding)) |
| 939 for e in elem: |
| 940 _serialize_xml(write, e, encoding, qnames, None) |
| 941 write("</" + tag + ">") |
| 942 else: |
| 943 write(" />") |
| 944 if elem.tail: |
| 945 write(_escape_cdata(elem.tail, encoding)) |
| 946 |
| 947 HTML_EMPTY = ("area", "base", "basefont", "br", "col", "frame", "hr", |
| 948 "img", "input", "isindex", "link", "meta" "param") |
| 949 |
| 950 try: |
| 951 HTML_EMPTY = set(HTML_EMPTY) |
| 952 except NameError: |
| 953 pass |
| 954 |
| 955 def _serialize_html(write, elem, encoding, qnames, namespaces): |
| 956 tag = elem.tag |
| 957 text = elem.text |
| 958 if tag is Comment: |
| 959 write("<!--%s-->" % _escape_cdata(text, encoding)) |
| 960 elif tag is ProcessingInstruction: |
| 961 write("<?%s?>" % _escape_cdata(text, encoding)) |
| 962 else: |
| 963 tag = qnames[tag] |
| 964 if tag is None: |
| 965 if text: |
| 966 write(_escape_cdata(text, encoding)) |
| 967 for e in elem: |
| 968 _serialize_html(write, e, encoding, qnames, None) |
| 969 else: |
| 970 write("<" + tag) |
| 971 items = elem.items() |
| 972 if items or namespaces: |
| 973 if namespaces: |
| 974 for v, k in sorted(namespaces.items(), |
| 975 key=lambda x: x[1]): # sort on prefix |
| 976 if k: |
| 977 k = ":" + k |
| 978 write(" xmlns%s=\"%s\"" % ( |
| 979 k.encode(encoding), |
| 980 _escape_attrib(v, encoding) |
| 981 )) |
| 982 for k, v in sorted(items): # lexical order |
| 983 if isinstance(k, QName): |
| 984 k = k.text |
| 985 if isinstance(v, QName): |
| 986 v = qnames[v.text] |
| 987 else: |
| 988 v = _escape_attrib_html(v, encoding) |
| 989 # FIXME: handle boolean attributes |
| 990 write(" %s=\"%s\"" % (qnames[k], v)) |
| 991 write(">") |
| 992 tag = tag.lower() |
| 993 if text: |
| 994 if tag == "script" or tag == "style": |
| 995 write(_encode(text, encoding)) |
| 996 else: |
| 997 write(_escape_cdata(text, encoding)) |
| 998 for e in elem: |
| 999 _serialize_html(write, e, encoding, qnames, None) |
| 1000 if tag not in HTML_EMPTY: |
| 1001 write("</" + tag + ">") |
| 1002 if elem.tail: |
| 1003 write(_escape_cdata(elem.tail, encoding)) |
| 1004 |
| 1005 def _serialize_text(write, elem, encoding): |
| 1006 for part in elem.itertext(): |
| 1007 write(part.encode(encoding)) |
| 1008 if elem.tail: |
| 1009 write(elem.tail.encode(encoding)) |
| 1010 |
| 1011 _serialize = { |
| 1012 "xml": _serialize_xml, |
| 1013 "html": _serialize_html, |
| 1014 "text": _serialize_text, |
| 1015 # this optional method is imported at the end of the module |
| 1016 # "c14n": _serialize_c14n, |
| 1017 } |
| 1018 |
| 1019 ## |
| 1020 # Registers a namespace prefix. The registry is global, and any |
| 1021 # existing mapping for either the given prefix or the namespace URI |
| 1022 # will be removed. |
| 1023 # |
| 1024 # @param prefix Namespace prefix. |
| 1025 # @param uri Namespace uri. Tags and attributes in this namespace |
| 1026 # will be serialized with the given prefix, if at all possible. |
| 1027 # @exception ValueError If the prefix is reserved, or is otherwise |
| 1028 # invalid. |
| 1029 |
| 1030 def register_namespace(prefix, uri): |
| 1031 if re.match("ns\d+$", prefix): |
| 1032 raise ValueError("Prefix format reserved for internal use") |
| 1033 for k, v in _namespace_map.items(): |
| 1034 if k == uri or v == prefix: |
| 1035 del _namespace_map[k] |
| 1036 _namespace_map[uri] = prefix |
| 1037 |
| 1038 _namespace_map = { |
| 1039 # "well-known" namespace prefixes |
| 1040 "http://www.w3.org/XML/1998/namespace": "xml", |
| 1041 "http://www.w3.org/1999/xhtml": "html", |
| 1042 "http://www.w3.org/1999/02/22-rdf-syntax-ns#": "rdf", |
| 1043 "http://schemas.xmlsoap.org/wsdl/": "wsdl", |
| 1044 # xml schema |
| 1045 "http://www.w3.org/2001/XMLSchema": "xs", |
| 1046 "http://www.w3.org/2001/XMLSchema-instance": "xsi", |
| 1047 # dublin core |
| 1048 "http://purl.org/dc/elements/1.1/": "dc", |
| 1049 } |
| 1050 |
| 1051 def _raise_serialization_error(text): |
| 1052 raise TypeError( |
| 1053 "cannot serialize %r (type %s)" % (text, type(text).__name__) |
| 1054 ) |
| 1055 |
| 1056 def _encode(text, encoding): |
| 1057 try: |
| 1058 return text.encode(encoding, "xmlcharrefreplace") |
| 1059 except (TypeError, AttributeError): |
| 1060 _raise_serialization_error(text) |
| 1061 |
| 1062 def _escape_cdata(text, encoding): |
| 1063 # escape character data |
| 1064 try: |
| 1065 # it's worth avoiding do-nothing calls for strings that are |
| 1066 # shorter than 500 character, or so. assume that's, by far, |
| 1067 # the most common case in most applications. |
| 1068 if "&" in text: |
| 1069 text = text.replace("&", "&") |
| 1070 if "<" in text: |
| 1071 text = text.replace("<", "<") |
| 1072 if ">" in text: |
| 1073 text = text.replace(">", ">") |
| 1074 return text.encode(encoding, "xmlcharrefreplace") |
| 1075 except (TypeError, AttributeError): |
| 1076 _raise_serialization_error(text) |
| 1077 |
| 1078 def _escape_attrib(text, encoding): |
| 1079 # escape attribute value |
| 1080 try: |
| 1081 if "&" in text: |
| 1082 text = text.replace("&", "&") |
| 1083 if "<" in text: |
| 1084 text = text.replace("<", "<") |
| 1085 if ">" in text: |
| 1086 text = text.replace(">", ">") |
| 1087 if "\"" in text: |
| 1088 text = text.replace("\"", """) |
| 1089 if "\n" in text: |
| 1090 text = text.replace("\n", " ") |
| 1091 return text.encode(encoding, "xmlcharrefreplace") |
| 1092 except (TypeError, AttributeError): |
| 1093 _raise_serialization_error(text) |
| 1094 |
| 1095 def _escape_attrib_html(text, encoding): |
| 1096 # escape attribute value |
| 1097 try: |
| 1098 if "&" in text: |
| 1099 text = text.replace("&", "&") |
| 1100 if ">" in text: |
| 1101 text = text.replace(">", ">") |
| 1102 if "\"" in text: |
| 1103 text = text.replace("\"", """) |
| 1104 return text.encode(encoding, "xmlcharrefreplace") |
| 1105 except (TypeError, AttributeError): |
| 1106 _raise_serialization_error(text) |
| 1107 |
| 1108 # -------------------------------------------------------------------- |
| 1109 |
| 1110 ## |
| 1111 # Generates a string representation of an XML element, including all |
| 1112 # subelements. |
| 1113 # |
| 1114 # @param element An Element instance. |
| 1115 # @keyparam encoding Optional output encoding (default is US-ASCII). |
| 1116 # @keyparam method Optional output method ("xml", "html", "text" or |
| 1117 # "c14n"; default is "xml"). |
| 1118 # @return An encoded string containing the XML data. |
| 1119 # @defreturn string |
| 1120 |
| 1121 def tostring(element, encoding=None, method=None): |
| 1122 class dummy: |
| 1123 pass |
| 1124 data = [] |
| 1125 file = dummy() |
| 1126 file.write = data.append |
| 1127 ElementTree(element).write(file, encoding, method=method) |
| 1128 return "".join(data) |
| 1129 |
| 1130 ## |
| 1131 # Generates a string representation of an XML element, including all |
| 1132 # subelements. The string is returned as a sequence of string fragments. |
| 1133 # |
| 1134 # @param element An Element instance. |
| 1135 # @keyparam encoding Optional output encoding (default is US-ASCII). |
| 1136 # @keyparam method Optional output method ("xml", "html", "text" or |
| 1137 # "c14n"; default is "xml"). |
| 1138 # @return A sequence object containing the XML data. |
| 1139 # @defreturn sequence |
| 1140 # @since 1.3 |
| 1141 |
| 1142 def tostringlist(element, encoding=None, method=None): |
| 1143 class dummy: |
| 1144 pass |
| 1145 data = [] |
| 1146 file = dummy() |
| 1147 file.write = data.append |
| 1148 ElementTree(element).write(file, encoding, method=method) |
| 1149 # FIXME: merge small fragments into larger parts |
| 1150 return data |
| 1151 |
| 1152 ## |
| 1153 # Writes an element tree or element structure to sys.stdout. This |
| 1154 # function should be used for debugging only. |
| 1155 # <p> |
| 1156 # The exact output format is implementation dependent. In this |
| 1157 # version, it's written as an ordinary XML file. |
| 1158 # |
| 1159 # @param elem An element tree or an individual element. |
| 1160 |
| 1161 def dump(elem): |
| 1162 # debugging |
| 1163 if not isinstance(elem, ElementTree): |
| 1164 elem = ElementTree(elem) |
| 1165 elem.write(sys.stdout) |
| 1166 tail = elem.getroot().tail |
| 1167 if not tail or tail[-1] != "\n": |
| 1168 sys.stdout.write("\n") |
| 1169 |
| 1170 # -------------------------------------------------------------------- |
| 1171 # parsing |
| 1172 |
| 1173 ## |
| 1174 # Parses an XML document into an element tree. |
| 1175 # |
| 1176 # @param source A filename or file object containing XML data. |
| 1177 # @param parser An optional parser instance. If not given, the |
| 1178 # standard {@link XMLParser} parser is used. |
| 1179 # @return An ElementTree instance |
| 1180 |
| 1181 def parse(source, parser=None): |
| 1182 tree = ElementTree() |
| 1183 tree.parse(source, parser) |
| 1184 return tree |
| 1185 |
| 1186 ## |
| 1187 # Parses an XML document into an element tree incrementally, and reports |
| 1188 # what's going on to the user. |
| 1189 # |
| 1190 # @param source A filename or file object containing XML data. |
| 1191 # @param events A list of events to report back. If omitted, only "end" |
| 1192 # events are reported. |
| 1193 # @param parser An optional parser instance. If not given, the |
| 1194 # standard {@link XMLParser} parser is used. |
| 1195 # @return A (event, elem) iterator. |
| 1196 |
| 1197 def iterparse(source, events=None, parser=None): |
| 1198 close_source = False |
| 1199 if not hasattr(source, "read"): |
| 1200 source = open(source, "rb") |
| 1201 close_source = True |
| 1202 if not parser: |
| 1203 parser = XMLParser(target=TreeBuilder()) |
| 1204 return _IterParseIterator(source, events, parser, close_source) |
| 1205 |
| 1206 class _IterParseIterator(object): |
| 1207 |
| 1208 def __init__(self, source, events, parser, close_source=False): |
| 1209 self._file = source |
| 1210 self._close_file = close_source |
| 1211 self._events = [] |
| 1212 self._index = 0 |
| 1213 self._error = None |
| 1214 self.root = self._root = None |
| 1215 self._parser = parser |
| 1216 # wire up the parser for event reporting |
| 1217 parser = self._parser._parser |
| 1218 append = self._events.append |
| 1219 if events is None: |
| 1220 events = ["end"] |
| 1221 for event in events: |
| 1222 if event == "start": |
| 1223 try: |
| 1224 parser.ordered_attributes = 1 |
| 1225 parser.specified_attributes = 1 |
| 1226 def handler(tag, attrib_in, event=event, append=append, |
| 1227 start=self._parser._start_list): |
| 1228 append((event, start(tag, attrib_in))) |
| 1229 parser.StartElementHandler = handler |
| 1230 except AttributeError: |
| 1231 def handler(tag, attrib_in, event=event, append=append, |
| 1232 start=self._parser._start): |
| 1233 append((event, start(tag, attrib_in))) |
| 1234 parser.StartElementHandler = handler |
| 1235 elif event == "end": |
| 1236 def handler(tag, event=event, append=append, |
| 1237 end=self._parser._end): |
| 1238 append((event, end(tag))) |
| 1239 parser.EndElementHandler = handler |
| 1240 elif event == "start-ns": |
| 1241 def handler(prefix, uri, event=event, append=append): |
| 1242 try: |
| 1243 uri = (uri or "").encode("ascii") |
| 1244 except UnicodeError: |
| 1245 pass |
| 1246 append((event, (prefix or "", uri or ""))) |
| 1247 parser.StartNamespaceDeclHandler = handler |
| 1248 elif event == "end-ns": |
| 1249 def handler(prefix, event=event, append=append): |
| 1250 append((event, None)) |
| 1251 parser.EndNamespaceDeclHandler = handler |
| 1252 else: |
| 1253 raise ValueError("unknown event %r" % event) |
| 1254 |
| 1255 def next(self): |
| 1256 while 1: |
| 1257 try: |
| 1258 item = self._events[self._index] |
| 1259 self._index += 1 |
| 1260 return item |
| 1261 except IndexError: |
| 1262 pass |
| 1263 if self._error: |
| 1264 e = self._error |
| 1265 self._error = None |
| 1266 raise e |
| 1267 if self._parser is None: |
| 1268 self.root = self._root |
| 1269 if self._close_file: |
| 1270 self._file.close() |
| 1271 raise StopIteration |
| 1272 # load event buffer |
| 1273 del self._events[:] |
| 1274 self._index = 0 |
| 1275 data = self._file.read(16384) |
| 1276 if data: |
| 1277 try: |
| 1278 self._parser.feed(data) |
| 1279 except SyntaxError as exc: |
| 1280 self._error = exc |
| 1281 else: |
| 1282 self._root = self._parser.close() |
| 1283 self._parser = None |
| 1284 |
| 1285 def __iter__(self): |
| 1286 return self |
| 1287 |
| 1288 ## |
| 1289 # Parses an XML document from a string constant. This function can |
| 1290 # be used to embed "XML literals" in Python code. |
| 1291 # |
| 1292 # @param source A string containing XML data. |
| 1293 # @param parser An optional parser instance. If not given, the |
| 1294 # standard {@link XMLParser} parser is used. |
| 1295 # @return An Element instance. |
| 1296 # @defreturn Element |
| 1297 |
| 1298 def XML(text, parser=None): |
| 1299 if not parser: |
| 1300 parser = XMLParser(target=TreeBuilder()) |
| 1301 parser.feed(text) |
| 1302 return parser.close() |
| 1303 |
| 1304 ## |
| 1305 # Parses an XML document from a string constant, and also returns |
| 1306 # a dictionary which maps from element id:s to elements. |
| 1307 # |
| 1308 # @param source A string containing XML data. |
| 1309 # @param parser An optional parser instance. If not given, the |
| 1310 # standard {@link XMLParser} parser is used. |
| 1311 # @return A tuple containing an Element instance and a dictionary. |
| 1312 # @defreturn (Element, dictionary) |
| 1313 |
| 1314 def XMLID(text, parser=None): |
| 1315 if not parser: |
| 1316 parser = XMLParser(target=TreeBuilder()) |
| 1317 parser.feed(text) |
| 1318 tree = parser.close() |
| 1319 ids = {} |
| 1320 for elem in tree.iter(): |
| 1321 id = elem.get("id") |
| 1322 if id: |
| 1323 ids[id] = elem |
| 1324 return tree, ids |
| 1325 |
| 1326 ## |
| 1327 # Parses an XML document from a string constant. Same as {@link #XML}. |
| 1328 # |
| 1329 # @def fromstring(text) |
| 1330 # @param source A string containing XML data. |
| 1331 # @return An Element instance. |
| 1332 # @defreturn Element |
| 1333 |
| 1334 fromstring = XML |
| 1335 |
| 1336 ## |
| 1337 # Parses an XML document from a sequence of string fragments. |
| 1338 # |
| 1339 # @param sequence A list or other sequence containing XML data fragments. |
| 1340 # @param parser An optional parser instance. If not given, the |
| 1341 # standard {@link XMLParser} parser is used. |
| 1342 # @return An Element instance. |
| 1343 # @defreturn Element |
| 1344 # @since 1.3 |
| 1345 |
| 1346 def fromstringlist(sequence, parser=None): |
| 1347 if not parser: |
| 1348 parser = XMLParser(target=TreeBuilder()) |
| 1349 for text in sequence: |
| 1350 parser.feed(text) |
| 1351 return parser.close() |
| 1352 |
| 1353 # -------------------------------------------------------------------- |
| 1354 |
| 1355 ## |
| 1356 # Generic element structure builder. This builder converts a sequence |
| 1357 # of {@link #TreeBuilder.start}, {@link #TreeBuilder.data}, and {@link |
| 1358 # #TreeBuilder.end} method calls to a well-formed element structure. |
| 1359 # <p> |
| 1360 # You can use this class to build an element structure using a custom XML |
| 1361 # parser, or a parser for some other XML-like format. |
| 1362 # |
| 1363 # @param element_factory Optional element factory. This factory |
| 1364 # is called to create new Element instances, as necessary. |
| 1365 |
| 1366 class TreeBuilder(object): |
| 1367 |
| 1368 def __init__(self, element_factory=None): |
| 1369 self._data = [] # data collector |
| 1370 self._elem = [] # element stack |
| 1371 self._last = None # last element |
| 1372 self._tail = None # true if we're after an end tag |
| 1373 if element_factory is None: |
| 1374 element_factory = Element |
| 1375 self._factory = element_factory |
| 1376 |
| 1377 ## |
| 1378 # Flushes the builder buffers, and returns the toplevel document |
| 1379 # element. |
| 1380 # |
| 1381 # @return An Element instance. |
| 1382 # @defreturn Element |
| 1383 |
| 1384 def close(self): |
| 1385 assert len(self._elem) == 0, "missing end tags" |
| 1386 assert self._last is not None, "missing toplevel element" |
| 1387 return self._last |
| 1388 |
| 1389 def _flush(self): |
| 1390 if self._data: |
| 1391 if self._last is not None: |
| 1392 text = "".join(self._data) |
| 1393 if self._tail: |
| 1394 assert self._last.tail is None, "internal error (tail)" |
| 1395 self._last.tail = text |
| 1396 else: |
| 1397 assert self._last.text is None, "internal error (text)" |
| 1398 self._last.text = text |
| 1399 self._data = [] |
| 1400 |
| 1401 ## |
| 1402 # Adds text to the current element. |
| 1403 # |
| 1404 # @param data A string. This should be either an 8-bit string |
| 1405 # containing ASCII text, or a Unicode string. |
| 1406 |
| 1407 def data(self, data): |
| 1408 self._data.append(data) |
| 1409 |
| 1410 ## |
| 1411 # Opens a new element. |
| 1412 # |
| 1413 # @param tag The element name. |
| 1414 # @param attrib A dictionary containing element attributes. |
| 1415 # @return The opened element. |
| 1416 # @defreturn Element |
| 1417 |
| 1418 def start(self, tag, attrs): |
| 1419 self._flush() |
| 1420 self._last = elem = self._factory(tag, attrs) |
| 1421 if self._elem: |
| 1422 self._elem[-1].append(elem) |
| 1423 self._elem.append(elem) |
| 1424 self._tail = 0 |
| 1425 return elem |
| 1426 |
| 1427 ## |
| 1428 # Closes the current element. |
| 1429 # |
| 1430 # @param tag The element name. |
| 1431 # @return The closed element. |
| 1432 # @defreturn Element |
| 1433 |
| 1434 def end(self, tag): |
| 1435 self._flush() |
| 1436 self._last = self._elem.pop() |
| 1437 assert self._last.tag == tag,\ |
| 1438 "end tag mismatch (expected %s, got %s)" % ( |
| 1439 self._last.tag, tag) |
| 1440 self._tail = 1 |
| 1441 return self._last |
| 1442 |
| 1443 ## |
| 1444 # Element structure builder for XML source data, based on the |
| 1445 # <b>expat</b> parser. |
| 1446 # |
| 1447 # @keyparam target Target object. If omitted, the builder uses an |
| 1448 # instance of the standard {@link #TreeBuilder} class. |
| 1449 # @keyparam html Predefine HTML entities. This flag is not supported |
| 1450 # by the current implementation. |
| 1451 # @keyparam encoding Optional encoding. If given, the value overrides |
| 1452 # the encoding specified in the XML file. |
| 1453 # @see #ElementTree |
| 1454 # @see #TreeBuilder |
| 1455 |
| 1456 class XMLParser(object): |
| 1457 |
| 1458 def __init__(self, html=0, target=None, encoding=None): |
| 1459 try: |
| 1460 from xml.parsers import expat |
| 1461 except ImportError: |
| 1462 try: |
| 1463 import pyexpat as expat |
| 1464 except ImportError: |
| 1465 raise ImportError( |
| 1466 "No module named expat; use SimpleXMLTreeBuilder instead" |
| 1467 ) |
| 1468 parser = expat.ParserCreate(encoding, "}") |
| 1469 if target is None: |
| 1470 target = TreeBuilder() |
| 1471 # underscored names are provided for compatibility only |
| 1472 self.parser = self._parser = parser |
| 1473 self.target = self._target = target |
| 1474 self._error = expat.error |
| 1475 self._names = {} # name memo cache |
| 1476 # callbacks |
| 1477 parser.DefaultHandlerExpand = self._default |
| 1478 parser.StartElementHandler = self._start |
| 1479 parser.EndElementHandler = self._end |
| 1480 parser.CharacterDataHandler = self._data |
| 1481 # optional callbacks |
| 1482 parser.CommentHandler = self._comment |
| 1483 parser.ProcessingInstructionHandler = self._pi |
| 1484 # let expat do the buffering, if supported |
| 1485 try: |
| 1486 self._parser.buffer_text = 1 |
| 1487 except AttributeError: |
| 1488 pass |
| 1489 # use new-style attribute handling, if supported |
| 1490 try: |
| 1491 self._parser.ordered_attributes = 1 |
| 1492 self._parser.specified_attributes = 1 |
| 1493 parser.StartElementHandler = self._start_list |
| 1494 except AttributeError: |
| 1495 pass |
| 1496 self._doctype = None |
| 1497 self.entity = {} |
| 1498 try: |
| 1499 self.version = "Expat %d.%d.%d" % expat.version_info |
| 1500 except AttributeError: |
| 1501 pass # unknown |
| 1502 |
| 1503 def _raiseerror(self, value): |
| 1504 err = ParseError(value) |
| 1505 err.code = value.code |
| 1506 err.position = value.lineno, value.offset |
| 1507 raise err |
| 1508 |
| 1509 def _fixtext(self, text): |
| 1510 # convert text string to ascii, if possible |
| 1511 try: |
| 1512 return text.encode("ascii") |
| 1513 except UnicodeError: |
| 1514 return text |
| 1515 |
| 1516 def _fixname(self, key): |
| 1517 # expand qname, and convert name string to ascii, if possible |
| 1518 try: |
| 1519 name = self._names[key] |
| 1520 except KeyError: |
| 1521 name = key |
| 1522 if "}" in name: |
| 1523 name = "{" + name |
| 1524 self._names[key] = name = self._fixtext(name) |
| 1525 return name |
| 1526 |
| 1527 def _start(self, tag, attrib_in): |
| 1528 fixname = self._fixname |
| 1529 fixtext = self._fixtext |
| 1530 tag = fixname(tag) |
| 1531 attrib = {} |
| 1532 for key, value in attrib_in.items(): |
| 1533 attrib[fixname(key)] = fixtext(value) |
| 1534 return self.target.start(tag, attrib) |
| 1535 |
| 1536 def _start_list(self, tag, attrib_in): |
| 1537 fixname = self._fixname |
| 1538 fixtext = self._fixtext |
| 1539 tag = fixname(tag) |
| 1540 attrib = {} |
| 1541 if attrib_in: |
| 1542 for i in range(0, len(attrib_in), 2): |
| 1543 attrib[fixname(attrib_in[i])] = fixtext(attrib_in[i+1]) |
| 1544 return self.target.start(tag, attrib) |
| 1545 |
| 1546 def _data(self, text): |
| 1547 return self.target.data(self._fixtext(text)) |
| 1548 |
| 1549 def _end(self, tag): |
| 1550 return self.target.end(self._fixname(tag)) |
| 1551 |
| 1552 def _comment(self, data): |
| 1553 try: |
| 1554 comment = self.target.comment |
| 1555 except AttributeError: |
| 1556 pass |
| 1557 else: |
| 1558 return comment(self._fixtext(data)) |
| 1559 |
| 1560 def _pi(self, target, data): |
| 1561 try: |
| 1562 pi = self.target.pi |
| 1563 except AttributeError: |
| 1564 pass |
| 1565 else: |
| 1566 return pi(self._fixtext(target), self._fixtext(data)) |
| 1567 |
| 1568 def _default(self, text): |
| 1569 prefix = text[:1] |
| 1570 if prefix == "&": |
| 1571 # deal with undefined entities |
| 1572 try: |
| 1573 self.target.data(self.entity[text[1:-1]]) |
| 1574 except KeyError: |
| 1575 from xml.parsers import expat |
| 1576 err = expat.error( |
| 1577 "undefined entity %s: line %d, column %d" % |
| 1578 (text, self._parser.ErrorLineNumber, |
| 1579 self._parser.ErrorColumnNumber) |
| 1580 ) |
| 1581 err.code = 11 # XML_ERROR_UNDEFINED_ENTITY |
| 1582 err.lineno = self._parser.ErrorLineNumber |
| 1583 err.offset = self._parser.ErrorColumnNumber |
| 1584 raise err |
| 1585 elif prefix == "<" and text[:9] == "<!DOCTYPE": |
| 1586 self._doctype = [] # inside a doctype declaration |
| 1587 elif self._doctype is not None: |
| 1588 # parse doctype contents |
| 1589 if prefix == ">": |
| 1590 self._doctype = None |
| 1591 return |
| 1592 text = text.strip() |
| 1593 if not text: |
| 1594 return |
| 1595 self._doctype.append(text) |
| 1596 n = len(self._doctype) |
| 1597 if n > 2: |
| 1598 type = self._doctype[1] |
| 1599 if type == "PUBLIC" and n == 4: |
| 1600 name, type, pubid, system = self._doctype |
| 1601 elif type == "SYSTEM" and n == 3: |
| 1602 name, type, system = self._doctype |
| 1603 pubid = None |
| 1604 else: |
| 1605 return |
| 1606 if pubid: |
| 1607 pubid = pubid[1:-1] |
| 1608 if hasattr(self.target, "doctype"): |
| 1609 self.target.doctype(name, pubid, system[1:-1]) |
| 1610 elif self.doctype is not self._XMLParser__doctype: |
| 1611 # warn about deprecated call |
| 1612 self._XMLParser__doctype(name, pubid, system[1:-1]) |
| 1613 self.doctype(name, pubid, system[1:-1]) |
| 1614 self._doctype = None |
| 1615 |
| 1616 ## |
| 1617 # (Deprecated) Handles a doctype declaration. |
| 1618 # |
| 1619 # @param name Doctype name. |
| 1620 # @param pubid Public identifier. |
| 1621 # @param system System identifier. |
| 1622 |
| 1623 def doctype(self, name, pubid, system): |
| 1624 """This method of XMLParser is deprecated.""" |
| 1625 warnings.warn( |
| 1626 "This method of XMLParser is deprecated. Define doctype() " |
| 1627 "method on the TreeBuilder target.", |
| 1628 DeprecationWarning, |
| 1629 ) |
| 1630 |
| 1631 # sentinel, if doctype is redefined in a subclass |
| 1632 __doctype = doctype |
| 1633 |
| 1634 ## |
| 1635 # Feeds data to the parser. |
| 1636 # |
| 1637 # @param data Encoded data. |
| 1638 |
| 1639 def feed(self, data): |
| 1640 try: |
| 1641 self._parser.Parse(data, 0) |
| 1642 except self._error, v: |
| 1643 self._raiseerror(v) |
| 1644 |
| 1645 ## |
| 1646 # Finishes feeding data to the parser. |
| 1647 # |
| 1648 # @return An element structure. |
| 1649 # @defreturn Element |
| 1650 |
| 1651 def close(self): |
| 1652 try: |
| 1653 self._parser.Parse("", 1) # end of data |
| 1654 except self._error, v: |
| 1655 self._raiseerror(v) |
| 1656 tree = self.target.close() |
| 1657 del self.target, self._parser # get rid of circular references |
| 1658 return tree |
| 1659 |
| 1660 # compatibility |
| 1661 XMLTreeBuilder = XMLParser |
| 1662 |
| 1663 # workaround circular import. |
| 1664 try: |
| 1665 from ElementC14N import _serialize_c14n |
| 1666 _serialize["c14n"] = _serialize_c14n |
| 1667 except ImportError: |
| 1668 pass |
OLD | NEW |