In January 2021 we will introduce a 10 GB quota for project repositories. Higher limits for individual projects will be available on request. Please see https://doku.lrz.de/display/PUBLIC/GitLab for more information.

syntaxtree.py 21.4 KB
Newer Older
1
"""syntaxtree.py - syntax tree classes for DHParser
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Copyright 2016  by Eckhart Arnold (arnold@badw.de)
                Bavarian Academy of Sciences an Humanities (badw.de)

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

    http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied.  See the License for the specific language governing
permissions and limitations under the License.
"""

Eckhart Arnold's avatar
Eckhart Arnold committed
19
import collections.abc
20
import copy
21
import os
22 23
from functools import partial

24 25 26 27
try:
    import regex as re
except ImportError:
    import re
28 29 30 31 32 33
try:
    from typing import AbstractSet, Any, ByteString, Callable, cast, Container, Dict, \
        Iterator, List, NamedTuple, Sequence, Union, Text, Tuple
except ImportError:
    from .typing34 import AbstractSet, Any, ByteString, Callable, cast, Container, Dict, \
        Iterator, List, NamedTuple, Sequence, Union, Text, Tuple
34

35
from DHParser.toolkit import is_logging, log_dir, StringView, linebreaks, line_col, identity
36

37
__all__ = ('WHITESPACE_PTYPE',
38
           'MockParser',
39
           'TOKEN_PTYPE',
Eckhart Arnold's avatar
Eckhart Arnold committed
40
           'ZOMBIE_PARSER',
Eckhart Arnold's avatar
Eckhart Arnold committed
41
           'ParserBase',
Eckhart Arnold's avatar
Eckhart Arnold committed
42 43
           'Error',
           'Node',
44 45
           'mock_syntax_tree',
           'TransformationFunc')
Eckhart Arnold's avatar
Eckhart Arnold committed
46 47


Eckhart Arnold's avatar
Eckhart Arnold committed
48 49 50 51 52 53 54 55 56 57
class ParserBase:
    """
    ParserBase is the base class for all real and mock parser classes.
    It is defined here, because Node objects require a parser object
    for instantiation.
    """
    def __init__(self, name=''):  # , pbases=frozenset()):
        self.name = name  # type: str
        self._ptype = ':' + self.__class__.__name__  # type: str

58 59 60
    def __repr__(self):
        return self.name + self.ptype

Eckhart Arnold's avatar
Eckhart Arnold committed
61
    def __str__(self):
62
        return self.name + (' = ' if self.name else '') + repr(self)
Eckhart Arnold's avatar
Eckhart Arnold committed
63 64 65 66 67

    @property
    def ptype(self) -> str:
        return self._ptype

68 69 70
    @property
    def repr(self) -> str:
        return self.name if self.name else repr(self)
Eckhart Arnold's avatar
Eckhart Arnold committed
71

72

73 74 75 76
WHITESPACE_PTYPE = ':Whitespace'
TOKEN_PTYPE = ':Token'


Eckhart Arnold's avatar
Eckhart Arnold committed
77
class MockParser(ParserBase):
78 79
    """
    MockParser objects can be used to reconstruct syntax trees from a
Eckhart Arnold's avatar
Eckhart Arnold committed
80 81
    serialized form like S-expressions or XML. Mock objects can mimic
    different parser types by assigning them a ptype on initialization.
82 83 84 85 86
    
    Mock objects should not be used for anything other than 
    syntax tree (re-)construction. In all other cases where a parser
    object substitute is needed, chose the singleton ZOMBIE_PARSER.
    """
Eckhart Arnold's avatar
Eckhart Arnold committed
87
    def __init__(self, name='', ptype=''):  # , pbases=frozenset()):
88
        assert not ptype or ptype[0] == ':'
Eckhart Arnold's avatar
Eckhart Arnold committed
89
        super(MockParser, self).__init__(name)
90
        self.name = name
Eckhart Arnold's avatar
Eckhart Arnold committed
91
        self._ptype = ptype or ':' + self.__class__.__name__
92 93 94


class ZombieParser(MockParser):
95 96 97 98 99 100 101 102 103 104 105 106
    """
    Serves as a substitute for a Parser instance.

    ``ZombieParser`` is the class of the singelton object
    ``ZOMBIE_PARSER``. The  ``ZOMBIE_PARSER`` has a name and can be
    called, but it never matches. It serves as a substitute where only
    these (or one of these properties) is needed, but no real Parser-
    object is instantiated.
    """
    alive = False

    def __init__(self):
107
        super(ZombieParser, self).__init__("__ZOMBIE__")
108 109 110 111
        assert not self.__class__.alive, "There can be only one!"
        assert self.__class__ == ZombieParser, "No derivatives, please!"
        self.__class__.alive = True

Eckhart Arnold's avatar
Eckhart Arnold committed
112 113 114 115 116 117
    def __copy__(self):
        return self

    def __deepcopy__(self, memo):
        return self

118 119 120 121
    def __call__(self, text):
        """Better call Saul ;-)"""
        return None, text

122 123 124 125

ZOMBIE_PARSER = ZombieParser()


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
class Error:
    __slots__ = ['message', 'category', 'code', 'pos', 'line', 'column']

    ERROR = "error"
    WARNING = "warning"

    def __init__(self, message: str, category: str='', code: str=''):
        self.message = message
        self.category = category or Error.ERROR
        self.code = code
        self.pos = -1
        self.line = -1
        self.column = -1

    def __str__(self):
        return ("line: %3i, column: %2i" % (self.line, self.column)
                + ", %s: %s" % (self.category, self.message))

    @staticmethod
    def from_template(template: str, category: str='', content: Union[tuple, dict]=()):
        if isinstance(content, tuple):
            return Error(template % content, category, template)
        else:
            return Error(template.format(**content), category, template)

151

152
ChildrenType = Tuple['Node', ...]
153 154
StrictResultType = Union[ChildrenType, StringView, str]
ResultType = Union[ChildrenType, 'Node', StringView, str, None]
155 156


Eckhart Arnold's avatar
Eckhart Arnold committed
157 158
def flatten_sxpr(sxpr: str) -> str:
    """Returns S-expression `sxpr` as a one-liner without unnecessary
Eckhart Arnold's avatar
Eckhart Arnold committed
159 160 161
    whitespace.

    Example:
Eckhart Arnold's avatar
Eckhart Arnold committed
162
    >>> flatten_sxpr('(a\\n    (b\\n        c\\n    )\\n)\\n')
Eckhart Arnold's avatar
Eckhart Arnold committed
163 164 165 166 167
    '(a (b c))'
    """
    return re.sub('\s(?=\))', '', re.sub('\s+', ' ', sxpr)).strip()


Eckhart Arnold's avatar
Eckhart Arnold committed
168
class Node(collections.abc.Sized):
169 170 171 172 173 174 175 176 177 178 179
    """
    Represents a node in the concrete or abstract syntax tree.

    Attributes:
        tag_name (str):  The name of the node, which is either its
            parser's name or, if that is empty, the parser's class name
        result (str or tuple):  The result of the parser which
            generated this node, which can be either a string or a
            tuple of child nodes.
        children (tuple):  The tuple of child nodes or an empty tuple
            if there are no child nodes. READ ONLY!
180 181 182 183 184 185
        parser (Parser):  The parser which generated this node. 
            WARNING: In case you use mock syntax trees for testing or
            parser replacement during the AST-transformation: DO NOT
            rely on this being a real parser object in any phase after 
            parsing (i.e. AST-transformation and compiling), for 
            example by calling ``isinstance(node.parer, ...)``.
186 187
        errors (list):  A list of parser- or compiler-errors:
            tuple(position, string) attached to this node
Eckhart Arnold's avatar
Eckhart Arnold committed
188 189
        error_flag (bool):  True, if either the node or any of its
            descendants has errors.
190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209
        len (int):  The full length of the node's string result if the
            node is a leaf node or, otherwise, the concatenated string
            result's of its descendants. The figure always represents
            the length before AST-transformation ans will never change
            through AST-transformation. READ ONLY!
        pos (int):  the position of the node within the parsed text.

            The value of ``pos`` is -1 meaning invalid by default. 
            Setting this value will set the positions of all child
            nodes relative to this value.  

            To set the pos values of all nodes in a syntax tree, the
            pos value of the root node should be set to 0 right 
            after parsing.

            Other than that, this value should be considered READ ONLY. 
            At any rate, it should only be reassigned only during
            parsing stage and never during or after the
            AST-transformation.
    """
210

211 212
    __slots__ = ['_result', 'children', '_errors', '_len', '_pos', 'parser', 'error_flag']

213

Eckhart Arnold's avatar
Eckhart Arnold committed
214
    def __init__(self, parser, result: ResultType) -> None:
215 216 217
        """Initializes the ``Node``-object with the ``Parser``-Instance
        that generated the node and the parser's result.
        """
Eckhart Arnold's avatar
Eckhart Arnold committed
218
        # self._result = ''  # type: StrictResultType
219
        # self.children = ()  # type: ChildrenType
220
        # self.error_flag = False  # type: bool
221
        self._errors = []  # type: List[Error]
222
        self.result = result
Eckhart Arnold's avatar
Eckhart Arnold committed
223
        self._len = len(self._result) if not self.children else \
224
            sum(child._len for child in self.children)  # type: int
Eckhart Arnold's avatar
Eckhart Arnold committed
225
        # self.pos: int  = 0  # continuous updating of pos values wastes a lot of time
226
        self._pos = -1  # type: int
227 228
        self.parser = parser or ZOMBIE_PARSER

Eckhart Arnold's avatar
Eckhart Arnold committed
229

230 231
    def __str__(self):
        if self.children:
232
            return "".join(str(child) for child in self.children)
233 234
        return str(self.result)

Eckhart Arnold's avatar
Eckhart Arnold committed
235

236 237 238 239 240 241
    def __repr__(self):
        mpargs = {'name': self.parser.name, 'ptype': self.parser.ptype}
        parg = "MockParser({name}, {ptype})".format(**mpargs)
        rarg = str(self) if not self.children else \
               "(" + ", ".join(repr(child) for child in self.children) + ")"
        return "Node(%s, %s)" % (parg, rarg)
242

Eckhart Arnold's avatar
Eckhart Arnold committed
243

244 245 246 247 248 249 250 251 252
    def __len__(self):
        return self._len


    def __bool__(self):
        # A node that is not None is always True, even if it's empty
        return True


253
    def __eq__(self, other):
254 255
        # return str(self.parser) == str(other.parser) and self.result == other.result
        return self.tag_name == other.tag_name and self.result == other.result
256

Eckhart Arnold's avatar
Eckhart Arnold committed
257

258
    def __hash__(self):
259
        return hash(self.tag_name)
260

Eckhart Arnold's avatar
Eckhart Arnold committed
261

262 263 264 265 266 267
    def __deepcopy__(self, memodict={}):
        result = copy.deepcopy(self.result)
        other = Node(self.parser, result)
        other._pos = self._pos
        return other

Eckhart Arnold's avatar
Eckhart Arnold committed
268

269
    @property   # this needs to be a (dynamic) property, in case sef.parser gets updated
270
    def tag_name(self) -> str:
271
        return self.parser.name or self.parser.ptype
272

Eckhart Arnold's avatar
Eckhart Arnold committed
273

274
    @property
Eckhart Arnold's avatar
Eckhart Arnold committed
275
    def result(self) -> StrictResultType:
276 277 278
        return self._result

    @result.setter
Eckhart Arnold's avatar
Eckhart Arnold committed
279
    def result(self, result: ResultType):
Eckhart Arnold's avatar
Eckhart Arnold committed
280
        # # made obsolete by static type checking with mypy
281 282 283
        # assert ((isinstance(result, tuple) and all(isinstance(child, Node) for child in result))
        #         or isinstance(result, Node)
        #         or isinstance(result, str)), str(result)
Eckhart Arnold's avatar
Eckhart Arnold committed
284 285
        # Possible optimization: Do not allow single nodes as argument:
        # assert not isinstance(result, Node)
286
        self._result = (result,) if isinstance(result, Node) else str(result) \
287
            if isinstance(result, StringView) else result or ''  # type: StrictResultType
288
        self.children = cast(ChildrenType, self._result) \
Eckhart Arnold's avatar
Eckhart Arnold committed
289
            if isinstance(self._result, tuple) else cast(ChildrenType, ())  # type: ChildrenType
290
        self.error_flag = any(r.error_flag for r in self.children)  # type: bool
291 292

    @property
293
    def pos(self) -> int:
294 295 296 297
        assert self._pos >= 0, "position value not initialized!"
        return self._pos

    @pos.setter
298
    def pos(self, pos: int):
299 300
        self._pos = pos
        offset = 0
301
        # recursively adjust pos-values of all children
302 303
        for child in self.children:
            child.pos = pos + offset
304
            offset += len(child)
305 306 307
        # add pos-values to Error-objects
        for err in self._errors:
            err.pos = pos
308

Eckhart Arnold's avatar
Eckhart Arnold committed
309

310
    @property
311
    def errors(self) -> List[Error]:
312 313
        return self._errors.copy()

314

315 316 317 318 319
    # def add_error(self, error_str: str) -> 'Node':
    #     assert isinstance(error_str, str)
    #     self._errors.append(error_str)
    #     self.error_flag = True
    #     return self
Eckhart Arnold's avatar
Eckhart Arnold committed
320

321 322 323 324 325 326 327 328 329 330

    def add_error(self: 'Node',
                  template: Union[str, Error],
                  category: str='',
                  content: Union[tuple, dict]=()) -> 'Node':
        if isinstance(template, Error):
            assert not (bool(category) or bool(content))
            self._errors.append(template)
        else:
            self._errors.append(Error.from_template(template, category, content))
Eckhart Arnold's avatar
Eckhart Arnold committed
331 332 333
        self.error_flag = True
        return self

Eckhart Arnold's avatar
Eckhart Arnold committed
334

335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351
    def _finalize_errors(self, lbreaks: List[int]):
        if self.error_flag:
            for err in self._errors:
                assert err.pos >= 0
                err.line, err.column = line_col(lbreaks, err.pos)
            for child in self.children:
                child._finalize_errors(lbreaks)


    def finalize_errors(self, source_text: Union[StringView, str]):
        """Recursively adds line- and column-numbers to all error objects.
        """
        if self.error_flag:
            lbreaks = linebreaks(source_text)
            self._finalize_errors(lbreaks)


Eckhart Arnold's avatar
Eckhart Arnold committed
352 353 354 355 356 357 358 359 360 361 362 363 364 365
    def collect_errors(self, clear_errors=False) -> List[Error]:
        """
        Returns all errors of this node or any child node in the form
        of a set of tuples (position, error_message), where position
        is always relative to this node.
        """
        errors = self.errors
        if clear_errors:
            self._errors = []
            self.error_flag = False
        if self.children:
            for child in self.children:
                errors.extend(child.collect_errors(clear_errors))
        return errors
366

Eckhart Arnold's avatar
Eckhart Arnold committed
367

368
    def _tree_repr(self, tab, openF, closeF, dataF=identity, density=0) -> str:
369 370 371 372 373 374 375 376
        """
        Generates a tree representation of this node and its children
        in string from.

        The kind ot tree-representation that is determined by several
        function parameters. This could be an XML-representation or a
        lisp-like S-expression.

377
        Args:
378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393
            tab (str):  The indentation string, e.g. '\t' or '    '
            openF:  (Node->str) A function that returns an opening
                string (e.g. an XML-tag_name) for a given node
            closeF:  (Node->str) A function that returns a closeF
                string (e.g. an XML-tag_name) for a given node.
            dataF:  (str->str) A function that filters the data string
                before printing, e.g. to add quotation marks

        Returns (str):
            A string that contains a (serialized) tree representation
            of the node and its children.
        """
        head = openF(self)
        tail = closeF(self)

        if not self.result:
394
            return head.rstrip() + tail.lstrip()
395

396
        D = None if density & 2 else ''
397 398 399

        if self.children:
            content = []
400
            for child in self.children:
401
                subtree = child._tree_repr(tab, openF, closeF, dataF, density).split('\n')
402
                content.append('\n'.join((tab + s) for s in subtree))
403
            return head + '\n'.join(content) + tail.lstrip(D)
404

405
        res = cast(str, self.result)  # safe, because if there are no children, result is a string
406 407 408 409
        if density & 1 and res.find('\n') < 0:  # and head[0] == "<":
            # except for XML, add a gap between opening statement and content
            gap = ' ' if head.rstrip()[-1] != '>' else ''
            return head.rstrip() + gap + dataF(self.result) + tail.lstrip()
410
        else:
411
            return head + '\n'.join([tab + dataF(s) for s in res.split('\n')]) + tail.lstrip(D)
412

Eckhart Arnold's avatar
Eckhart Arnold committed
413

414
    def as_sxpr(self, src: str=None, compact: bool=False) -> str:
415 416 417
        """
        Returns content as S-expression, i.e. in lisp-like form.

418
        Args:
419 420 421 422 423
            src:  The source text or `None`. In case the source text is
                given the position of the element in the text will be
                reported as line and column.
        """

424 425
        lB, rB, D = ('', '', 1) if compact else ('(', '\n)', 0)

426
        def opening(node) -> str:
427
            s = lB + node.tag_name
428 429
            # s += " '(pos %i)" % node.pos
            if src:
Eckhart Arnold's avatar
Eckhart Arnold committed
430
                s += " '(pos %i " % node.pos + " %i %i)" % line_col(src, node.pos)
431 432 433
            if node.errors:
                s += " '(err '(%s))" % ' '.join(str(err).replace('"', r'\"')
                                                for err in node.errors)
434
            return s + '\n'
435 436 437 438 439 440

        def pretty(s):
            return '"%s"' % s if s.find('"') < 0 \
                else "'%s'" % s if s.find("'") < 0 \
                else '"%s"' % s.replace('"', r'\"')

441
        return self._tree_repr('    ', opening, lambda node: rB, pretty, density=D)
442

Eckhart Arnold's avatar
Eckhart Arnold committed
443

444
    def as_xml(self, src: str=None) -> str:
445 446 447
        """
        Returns content as XML-tree.

448
        Args:
449 450 451 452 453
            src:  The source text or `None`. In case the source text is
                given the position will also be reported as line and
                column.
        """

454
        def opening(node) -> str:
455 456 457 458 459 460
            s = '<' + node.tag_name
            # s += ' pos="%i"' % node.pos
            if src:
                s += ' line="%i" col="%i"' % line_col(src, node.pos)
            if node.errors:
                s += ' err="%s"' % ''.join(str(err).replace('"', r'\"') for err in node.errors)
461
            return s + ">\n"
462 463

        def closing(node):
464
            return '\n</' + node.tag_name + '>'
465

466
        return self._tree_repr('    ', opening, closing, density=1)
467

Eckhart Arnold's avatar
Eckhart Arnold committed
468

Eckhart Arnold's avatar
Eckhart Arnold committed
469 470 471
    def structure(self) -> str:
        """Return structure (and content) as S-expression on a single line
        without any line breaks."""
Eckhart Arnold's avatar
Eckhart Arnold committed
472 473
        return flatten_sxpr(self.as_sxpr())

474

Eckhart Arnold's avatar
Eckhart Arnold committed
475
    def content(self) -> str:
476
        """
Eckhart Arnold's avatar
Eckhart Arnold committed
477 478
        Returns content as string, inserting error messages where
        errors occurred.
479
        """
Eckhart Arnold's avatar
Eckhart Arnold committed
480 481 482 483
        s = "".join(child.content() for child in self.children) if self.children \
            else str(self.result)
        return (
        ' <<< Error on "%s" | %s >>> ' % (s, '; '.join(self._errors))) if self._errors else s
484

Eckhart Arnold's avatar
Eckhart Arnold committed
485

486
    def find(self, match_function: Callable) -> Iterator['Node']:
487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506
        """Finds nodes in the tree that match a specific criterion.
        
        ``find`` is a generator that yields all nodes for which the
        given ``match_function`` evaluates to True. The tree is 
        traversed pre-order.
        
        Args:
            match_function (function): A function  that takes as Node
                object as argument and returns True or False
        Yields:
            Node: all nodes of the tree for which 
            ``match_function(node)`` returns True
        """
        if match_function(self):
            yield self
        else:
            for child in self.children:
                for nd in child.find(match_function):
                    yield nd

Eckhart Arnold's avatar
Eckhart Arnold committed
507

508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524
    # def range(self, match_first, match_last):
    #     """Iterates over the range of nodes, starting from the first
    #     node for which ``match_first`` becomes True until the first node
    #     after this one for which ``match_last`` becomes true or until
    #     the end if it never does.
    #
    #     Args:
    #         match_first (function): A function  that takes as Node
    #             object as argument and returns True or False
    #         match_last (function): A function  that takes as Node
    #             object as argument and returns True or False
    #     Yields:
    #         Node: all nodes of the tree for which
    #         ``match_function(node)`` returns True
    #     """


525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546
    # def navigate(self, path):
    #     """Yields the results of all descendant elements matched by
    #     ``path``, e.g.
    #     'd/s' yields 'l' from (d (s l)(e (r x1) (r x2))
    #     'e/r' yields 'x1', then 'x2'
    #     'e'   yields (r x1)(r x2)
    #
    #     Args:
    #         path (str):  The path of the object, e.g. 'a/b/c'. The
    #             components of ``path`` can be regular expressions
    #
    #     Returns:
    #         The object at the path, either a string or a Node or
    #         ``None``, if the path did not match.
    #     """
    #     def nav(node, pl):
    #         if pl:
    #             return itertools.chain(nav(child, pl[1:]) for child in node.children
    #                                    if re.match(pl[0], child.tag_name))
    #         else:
    #             return self.result,
    #     return nav(path.split('/'))
547

Eckhart Arnold's avatar
Eckhart Arnold committed
548

549 550 551 552 553
    def tree_size(self) -> int:
        """Recursively counts the number of nodes in the tree including the root node."""
        return sum(child.tree_size() for child in self.children) + 1


Eckhart Arnold's avatar
Eckhart Arnold committed
554
    def log(self, log_file_name):
555
        if is_logging():
Eckhart Arnold's avatar
Eckhart Arnold committed
556 557 558 559
            path = os.path.join(log_dir(), log_file_name)
            if os.path.exists(path):
                print('WARNING: Log-file "%s" already exists and will be overwritten!' % path)
            with open(path, "w", encoding="utf-8") as f:
560
                f.write(self.as_sxpr())
Eckhart Arnold's avatar
Eckhart Arnold committed
561

562

563
def mock_syntax_tree(sxpr):
564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587
    """
    Generates a tree of nodes from an S-expression.

    Example:
    >>> mock_syntax_tree("(a (b c))").as_sxpr()
    '(a\\n    (b\\n        "c"\\n    )\\n)'
    """

    def next_block(s):
        s = s.strip()
        while s[0] != ')':
            if s[0] != '(': raise ValueError('"(" expected, not ' + s[:10])
            # assert s[0] == '(', s
            level = 1
            i = 1
            while level > 0:
                if s[i] == '(':
                    level += 1
                elif s[i] == ')':
                    level -= 1
                i += 1
            yield s[:i]
            s = s[i:].strip()

588 589 590 591 592 593 594 595 596
    sxpr = sxpr.strip()
    if sxpr[0] != '(': raise ValueError('"(" expected, not ' + sxpr[:10])
    # assert sxpr[0] == '(', sxpr
    sxpr = sxpr[1:].strip()
    m = re.match('[\w:]+', sxpr)
    name, class_name = (sxpr[:m.end()].split(':') + [''])[:2]
    sxpr = sxpr[m.end():].strip()
    if sxpr[0] == '(':
        result = tuple(mock_syntax_tree(block) for block in next_block(sxpr))
597 598
    else:
        lines = []
599
        while sxpr and sxpr[0] != ')':
600
            for qm in ['"""', "'''", '"', "'"]:
601
                m = re.match(qm + r'.*?' + qm, sxpr, re.DOTALL)
602 603
                if m:
                    i = len(qm)
604 605
                    lines.append(sxpr[i:m.end() - i])
                    sxpr = sxpr[m.end():].strip()
606 607
                    break
            else:
608 609 610
                m = re.match(r'(?:(?!\)).)*', sxpr, re.DOTALL)
                lines.append(sxpr[:m.end()])
                sxpr = sxpr[m.end():]
611 612 613 614
        result = "\n".join(lines)
    return Node(MockParser(name, ':' + class_name), result)


615
TransformationFunc = Union[Callable[[Node], Any], partial]
616

617 618 619 620 621

# if __name__ == "__main__":
#     st = mock_syntax_tree("(alpha (beta (gamma i\nj\nk) (delta y)) (epsilon z))")
#     print(st.as_sxpr())
#     print(st.as_xml())