test_parse.py 34.3 KB
Newer Older
1 2
#!/usr/bin/python3

3
"""test_parse.py - tests of the parsers-module of DHParser
4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

Author: Eckhart Arnold <arnold@badw.de>

Copyright 2017 Bavarian Academy of Sciences and Humanities

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.
"""

import sys
23
from functools import partial
24

Eckhart Arnold's avatar
Eckhart Arnold committed
25
sys.path.extend(['../', './'])
26

27
from DHParser.toolkit import compile_python_object, get_config_value, set_config_value
28
from DHParser.log import logging, is_logging, log_ST, log_parsing_history
29
from DHParser.error import Error, is_error
30
from DHParser.parse import ParserError, Parser, Grammar, Forward, TKN, ZeroOrMore, RE, \
eckhart's avatar
eckhart committed
31
    RegExp, Lookbehind, NegativeLookahead, OneOrMore, Series, Alternative, AllOf, SomeOf, \
32
    UnknownParserError, MetaParser, GrammarError, EMPTY_NODE
33
from DHParser import compile_source
34 35
from DHParser.ebnf import get_ebnf_grammar, get_ebnf_transformer, get_ebnf_compiler, DHPARSER_IMPORTS
from DHParser.dsl import grammar_provider, CompilationError
Eckhart Arnold's avatar
Eckhart Arnold committed
36
from DHParser.syntaxtree import Node
37
from DHParser.stringview import StringView
38 39


40 41 42 43 44
class TestParserError:
    def test_parser_error_str(self):
        pe = ParserError(Node('TAG', 'test').with_pos(0), StringView('Beispiel'), True)
        assert str(pe).find('Beispiel') >= 0 and str(pe).find('TAG') >= 0

eckhart's avatar
eckhart committed
45 46 47 48 49 50 51 52 53 54
class TestParserClass:
    def test_apply(self):
        minilang ="""
            expr = expr ("+"|"-") term | term
            term = term ("*"|"/") factor | factor
            factor = /[0-9]+/~
            """
        gr = grammar_provider(minilang)()
        l = []
        def visitor(p: Parser):
di68kap's avatar
di68kap committed
55
            l.append(p.pname + p.ptype)
eckhart's avatar
eckhart committed
56 57 58 59 60 61 62 63 64 65 66
        gr.root__.apply(visitor)
        s1 = ", ".join(l)
        l = []
        gr.root__.apply(visitor)
        s2 = ", ".join(l)
        l = []
        gr.root__.apply(visitor)
        s3 = ", ".join(l)
        assert s1 == s2 == s3


67
class TestInfiLoopsAndRecursion:
68
    def test_direct_left_recursion1(self):
eckhart's avatar
eckhart committed
69
        minilang ="""
70 71 72 73
            expr = expr ("+"|"-") term | term
            term = term ("*"|"/") factor | factor
            factor = /[0-9]+/~
            """
74
        snippet = "9 + 8 + 7 + 6 + 5 + 3 * 4"
75
        parser = grammar_provider(minilang)()
76
        assert parser
Eckhart Arnold's avatar
Eckhart Arnold committed
77
        syntax_tree = parser(snippet)
78 79
        assert not is_error(syntax_tree.error_flag), str(syntax_tree.errors_sorted)
        assert snippet == syntax_tree.content, str(syntax_tree)
Eckhart Arnold's avatar
Eckhart Arnold committed
80
        if is_logging():
81
            log_ST(syntax_tree, "test_LeftRecursion_direct.cst")
82
            log_parsing_history(parser, "test_LeftRecursion_direct")
83

84
    def test_direct_left_recursion2(self):
85 86 87 88 89 90
        minilang = """
            expr = ex
            ex   = expr ("+"|"-") term | term
            term = term ("*"|"/") factor | factor
            factor = /[0-9]+/~
            """
91
        snippet = "9 + 8 + 7 + 6 + 5 + 3 * 4"
92
        parser = grammar_provider(minilang)()
93
        assert parser
94 95
        with logging():
            syntax_tree = parser(snippet)
96 97
        assert not is_error(syntax_tree.error_flag), syntax_tree.errors_sorted
        assert snippet == syntax_tree.content
98

99
    def test_indirect_left_recursion1(self):
100 101
        minilang = """
            Expr    = //~ (Product | Sum | Value)
102 103
            Product = Expr { ('*' | '/') Expr }+
            Sum     = Expr { ('+' | '-') Expr }+
104
            Value   = /[0-9.]+/~ | '(' §Expr ')'
105
            """
106
        parser = grammar_provider(minilang)()
107
        assert parser
108
        snippet = "8 * 4"
109
        syntax_tree = parser(snippet)
110
        assert not is_error(syntax_tree.error_flag), syntax_tree.errors_sorted
111 112
        snippet = "7 + 8 * 4"
        syntax_tree = parser(snippet)
113
        assert not is_error(syntax_tree.error_flag), syntax_tree.errors_sorted
114 115
        snippet = "9 + 8 * (4 + 3)"
        syntax_tree = parser(snippet)
116 117
        assert not is_error(syntax_tree.error_flag), syntax_tree.errors_sorted
        assert snippet == syntax_tree.content
118
        if is_logging():
119
            log_ST(syntax_tree, "test_LeftRecursion_indirect.cst")
120
            log_parsing_history(parser, "test_LeftRecursion_indirect")
121

122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141
    # # BEWARE: EXPERIMENTAL TEST can be long running
    # def test_indirect_left_recursion2(self):
    #     arithmetic_syntax = """
    #         expression     = addition | subtraction
    #         addition       = (expression | term) "+" (expression | term)
    #         subtraction    = (expression | term) "-" (expression | term)
    #         term           = multiplication | division
    #         multiplication = (term | factor) "*" (term | factor)
    #         division       = (term | factor) "/" (term | factor)
    #         factor         = [SIGN] ( NUMBER | VARIABLE | group ) { VARIABLE | group }
    #         group          = "(" expression ")"
    #         SIGN           = /[+-]/
    #         NUMBER         = /(?:0|(?:[1-9]\d*))(?:\.\d+)?/~
    #         VARIABLE       = /[A-Za-z]/~
    #         """
    #     arithmetic = grammar_provider(arithmetic_syntax)()
    #     arithmetic.left_recursion_depth__ = 2
    #     assert arithmetic
    #     syntax_tree = arithmetic("(a + b) * (a - b)")
    #     assert syntax_tree.errors
142

143 144 145 146 147 148 149 150 151 152
    def test_break_inifnite_loop_ZeroOrMore(self):
        forever = ZeroOrMore(RegExp(''))
        result = Grammar(forever)('')  # infinite loops will automatically be broken
        assert repr(result) == "Node(:EMPTY__, )", repr(result)

    def test_break_inifnite_loop_OneOrMore(self):
        forever = OneOrMore(RegExp(''))
        result = Grammar(forever)('')  # infinite loops will automatically be broken
        assert repr(result) == "Node(:EMPTY__, )", str(result)

153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177
    # def test_infinite_loops(self):
    #     minilang = """forever = { // } \n"""
    #     try:
    #         parser_class = grammar_provider(minilang)
    #     except CompilationError as error:
    #         assert all(e.code == Error.INFINITE_LOOP for e in error.errors)
    #     save = get_config_value('static_analysis')
    #     set_config_value('static_analysis', 'late')
    #     provider = grammar_provider(minilang)
    #     try:
    #         parser = provider()
    #     except GrammarError as error:
    #         assert error.errors[0][2].code == Error.INFINITE_LOOP
    #     set_config_value('static_analysis', 'none')
    #     parser = provider()
    #     snippet = " "
    #     syntax_tree = parser(snippet)
    #     assert any(e.code == Error.INFINITE_LOOP for e in syntax_tree.errors)
    #     res = parser.static_analysis()
    #     assert res and res[0][2].code == Error.INFINITE_LOOP
    #     minilang = """not_forever = { / / } \n"""
    #     parser = grammar_provider(minilang)()
    #     res = parser.static_analysis()
    #     assert not res
    #     set_config_value('static_analysis', save)
178

Eckhart Arnold's avatar
Eckhart Arnold committed
179
class TestFlowControl:
180 181
    def setup(self):
        self.t1 = """
eckhart's avatar
eckhart committed
182
        All work and no play
183 184 185 186 187
        makes Jack a dull boy
        END
        """
        self.t2 = "All word and not play makes Jack a dull boy END\n"

Eckhart Arnold's avatar
Eckhart Arnold committed
188
    def test_lookbehind(self):
189
        ws = RegExp(r'\s*')
Eckhart Arnold's avatar
Eckhart Arnold committed
190
        end = RegExp("END")
191
        doc_end = Lookbehind(RegExp('\\s*?\\n')) + end
192
        word = RegExp(r'\w+')
Eckhart Arnold's avatar
Eckhart Arnold committed
193 194 195 196
        sequence = OneOrMore(NegativeLookahead(end) + word + ws)
        document = ws + sequence + doc_end + ws

        parser = Grammar(document)
197 198 199 200 201 202 203
        cst = parser(self.t1)
        assert not cst.error_flag, cst.as_sxpr()
        cst = parser(self.t2)
        assert cst.error_flag, cst.as_sxpr()

    def test_lookbehind_indirect(self):
        class LookbehindTestGrammar(Grammar):
eckhart's avatar
eckhart committed
204
            parser_initialization__ = ["upon instantiation"]
205
            ws = RegExp(r'\s*')
206
            end = RegExp('END')
207
            SUCC_LB = RegExp('\\s*?\\n')
208
            doc_end = Series(Lookbehind(SUCC_LB), end)
209
            word = RegExp(r'\w+')
210 211 212 213 214 215
            sequence = OneOrMore(Series(NegativeLookahead(end), word, ws))
            document = Series(ws, sequence, doc_end, ws)
            root__ = document

        parser = LookbehindTestGrammar()
        cst = parser(self.t1)
Eckhart Arnold's avatar
Eckhart Arnold committed
216
        assert not cst.error_flag, cst.as_sxpr()
217
        cst = parser(self.t2)
Eckhart Arnold's avatar
Eckhart Arnold committed
218 219 220
        assert cst.error_flag, cst.as_sxpr()


221 222 223 224 225 226 227
class TestRegex:
    def test_multilineRegex(self):
        mlregex = r"""
        regex =  /\w+    # one or more alphabetical characters including the underscore
                  [+]    # followed by a plus sign
                  \w*    # possibly followed by more alpha chracters/
        """
Eckhart Arnold's avatar
Eckhart Arnold committed
228 229
        result, messages, syntax_tree = compile_source(mlregex, None, get_ebnf_grammar(),
                        get_ebnf_transformer(), get_ebnf_compiler('MultilineRegexTest'))
230
        assert result
231
        assert not messages, str(messages)
232
        parser = compile_python_object(DHPARSER_IMPORTS + result, r'\w+Grammar$')()
eckhart's avatar
eckhart committed
233 234
        node = parser('abc+def', parser.regex)
        assert not node.error_flag
235
        assert node.tag_name == "regex"
236 237
        assert str(node) == 'abc+def'

238 239 240 241 242 243 244 245 246 247
    def test_multilineRegex_wo_Comments(self):
        mlregex = r"""
        regex =  /\w+ 
                  [+]  
                  \w* /
        """
        result, messages, syntax_tree = compile_source(mlregex, None, get_ebnf_grammar(),
                        get_ebnf_transformer(), get_ebnf_compiler('MultilineRegexTest'))
        assert result
        assert not messages, str(messages)
248
        parser = compile_python_object(DHPARSER_IMPORTS + result, r'\w+Grammar$')()
249 250
        node = parser('abc+def', parser.regex)
        assert not node.error_flag
251
        assert node.tag_name == "regex"
252 253
        assert str(node) == 'abc+def'

254 255 256 257 258 259 260 261 262
    def text_ignore_case(self):
        mlregex = r"""
        @ ignorecase = True
        regex = /alpha/
        """
        result, messages, syntax_tree = compile_source(mlregex, None, get_ebnf_grammar(),
                        get_ebnf_transformer(), get_ebnf_compiler('MultilineRegexTest'))
        assert result
        assert not messages
263
        parser = compile_python_object(DHPARSER_IMPORTS + result, r'\w+Grammar$')()
264 265 266 267
        node, rest = parser.regex('Alpha')
        assert node
        assert not node.error_flag
        assert rest == ''
268
        assert node.tag_name == "regex"
269 270 271 272 273 274 275 276 277 278
        assert str(node) == 'Alpha'

        mlregex = r"""
        @ ignorecase = False
        regex = /alpha/
        """
        result, messages, syntax_tree = compile_source(mlregex, None, get_ebnf_grammar(),
                        get_ebnf_transformer(), get_ebnf_compiler('MultilineRegexTest'))
        assert result
        assert not messages
279
        parser = compile_python_object(DHPARSER_IMPORTS + result, r'\w+Grammar$')()
280 281 282 283
        node, rest = parser.regex('Alpha')
        assert node.error_flag


284 285 286 287 288 289 290 291 292 293 294 295
    def test_token(self):
        tokenlang = r"""
            @whitespace = linefeed
            lang        = "" begin_token {/\w+/ ""} end_token
            begin_token = "\begin{document}"
            end_token   = "\end{document}"
            """
        testdoc = r"""
            \begin{document}
            test
            \end{document}
            """
Eckhart Arnold's avatar
Eckhart Arnold committed
296 297
        result, messages, syntax_tree = compile_source(tokenlang, None, get_ebnf_grammar(),
                                    get_ebnf_transformer(), get_ebnf_compiler("TokenTest"))
298
        assert result
eckhart's avatar
eckhart committed
299
        assert not messages, str(messages)
300
        parser = compile_python_object(DHPARSER_IMPORTS + result, r'\w+Grammar$')()
Eckhart Arnold's avatar
Eckhart Arnold committed
301
        result = parser(testdoc)
302
        # log_parsing_history(parser, "test.log")
303 304
        assert not result.error_flag

305

306
class TestGrammar:
307
    def setup(self):
308 309 310 311 312 313
        grammar = r"""@whitespace = horizontal
        haupt        = textzeile LEERZEILE
        textzeile    = { WORT }+
        WORT         = /[^ \t]+/~
        LEERZEILE    = /\n[ \t]*(?=\n)/~
        """
314 315 316
        self.pyparser, messages, syntax_tree = compile_source(grammar, None, get_ebnf_grammar(),
                                                              get_ebnf_transformer(), get_ebnf_compiler("PosTest"))
        assert self.pyparser
317
        assert not messages
318 319 320 321

    def test_pos_values_initialized(self):
        # checks whether pos values in the parsing result and in the
        # history record have been initialized
322
        with logging("LOGS"):
323
            grammar = compile_python_object(DHPARSER_IMPORTS + self.pyparser, r'\w+Grammar$')()
324 325
            grammar("no_file_name*")
        for record in grammar.history__:
326 327
            assert not record.node or record.node.pos >= 0

328
    def test_select_parsing(self):
329
        grammar = compile_python_object(DHPARSER_IMPORTS + self.pyparser, r'\w+Grammar$')()
330 331 332 333
        grammar("wort", "WORT")
        grammar("eine Zeile", "textzeile")
        grammar("kein Haupt", "haupt")
        grammar("so ist es richtig", "haupt")
334

di68kap's avatar
di68kap committed
335 336
    def test_grammar_subclassing(self):
        class Arithmetic(Grammar):
337
            r'''
di68kap's avatar
di68kap committed
338 339 340 341 342 343
            expression =  term  { ("+" | "-") term }
            term       =  factor  { ("*" | "/") factor }
            factor     =  INTEGER | "("  expression  ")"
            INTEGER    =  /\d+/~
            '''
            expression = Forward()
344 345 346 347
            INTEGER = RE('\\d+')
            factor = INTEGER | TKN("(") + expression + TKN(")")
            term = factor + ZeroOrMore((TKN("*") | TKN("/")) + factor)
            expression.set(term + ZeroOrMore((TKN("+") | TKN("-")) + term))
di68kap's avatar
di68kap committed
348 349 350 351 352 353
            root__ = expression

        grammar = Arithmetic()
        CST = grammar('3+4')
        assert not CST.error_flag, CST.as_sxpr()

354

355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377
class TestSeries:
    def test_non_mandatory(self):
        lang = """
        document = series | /.*/
        series = "A" "B" "C" "D"
        """
        parser = grammar_provider(lang)()
        st = parser("ABCD");
        assert not st.error_flag
        st = parser("A_CD");
        assert not st.error_flag
        st = parser("AB_D");
        assert not st.error_flag

    def test_mandatory(self):
        """Test for the §-operator. The Series-parser should raise an
        error for any non-match that occurs after the mandatory-operator.
        """
        lang = """
        document = series | /.*/
        series = "A" "B" §"C" "D"
        """
        parser = grammar_provider(lang)()
378 379 380
        st = parser("ABCD");  assert not st.error_flag
        st = parser("A_CD");  assert not st.error_flag
        st = parser("AB_D");  assert st.error_flag
381
        assert st.errors_sorted[0].code == Error.MANDATORY_CONTINUATION
382
        # transitivity of mandatory-operator
383
        st = parser("ABC_");  assert st.error_flag
384
        assert st.errors_sorted[0].code == Error.MANDATORY_CONTINUATION
385 386

    def test_series_composition(self):
387
        TA, TB, TC, TD, TE = (TKN(b) for b in "ABCDE")
388 389 390 391 392
        s1 = Series(TA, TB, TC, mandatory=2)
        s2 = Series(TD, TE)

        combined = Alternative(s1 + s2, RegExp('.*'))
        parser = Grammar(combined)
393 394 395
        st = parser("ABCDE");  assert not st.error_flag
        st = parser("A_CDE");  assert not st.error_flag
        st = parser("AB_DE");  assert st.error_flag
396
        assert st.errors_sorted[0].code == Error.MANDATORY_CONTINUATION
397
        st = parser("ABC_E");  assert st.error_flag
398
        assert st.errors_sorted[0].code == Error.MANDATORY_CONTINUATION
399 400 401

        combined = Alternative(s2 + s1, RegExp('.*'))
        parser = Grammar(combined)
402 403 404 405 406 407
        st = parser("DEABC");  assert not st.error_flag
        st = parser("_EABC");  assert not st.error_flag
        st = parser("D_ABC");  assert not st.error_flag
        st = parser("DE_BC");  assert not st.error_flag
        st = parser("DEA_C");  assert not st.error_flag
        st = parser("DEAB_");  assert st.error_flag
408
        assert st.errors_sorted[0].code == Error.MANDATORY_CONTINUATION
409

eckhart's avatar
eckhart committed
410 411 412 413 414 415 416 417 418 419
    # def test_boundary_cases(self):
    #     lang = """
    #     document = series | §!single | /.*/
    #     series = "A" "B" §"C" "D"
    #     single = "E"
    #     """
    #     parser_class = grammar_provider(lang)
    #     parser = parser_class()
    #     print(parser.python_src__)
    #     print(parser_class.python_src__)
eckhart's avatar
eckhart committed
420

421

422 423 424
class TestAllOfSomeOf:
    def test_allOf_order(self):
        """Test that parsers of an AllOf-List can match in arbitrary order."""
425
        prefixes = AllOf(TKN("A"), TKN("B"))
426 427
        assert Grammar(prefixes)('A B').content == 'A B'
        assert Grammar(prefixes)('B A').content == 'B A'
428
        # aternative Form
429
        prefixes = AllOf(Series(TKN("B"), TKN("A")))
430
        assert Grammar(prefixes)('A B').content == 'A B'
431 432 433 434

    def test_allOf_completeness(self):
        """Test that an error is raised if not  all parsers of an AllOf-List
        match."""
435
        prefixes = AllOf(TKN("A"), TKN("B"))
436 437 438 439 440
        assert Grammar(prefixes)('B').error_flag

    def test_allOf_redundance(self):
        """Test that one and the same parser may be listed several times
        and must be matched several times accordingly."""
441
        prefixes = AllOf(TKN("A"), TKN("B"), TKN("A"))
442 443 444
        assert Grammar(prefixes)('A A B').content == 'A A B'
        assert Grammar(prefixes)('A B A').content == 'A B A'
        assert Grammar(prefixes)('B A A').content == 'B A A'
445 446 447 448
        assert Grammar(prefixes)('A B B').error_flag

    def test_someOf_order(self):
        """Test that parsers of an AllOf-List can match in arbitrary order."""
449
        prefixes = SomeOf(TKN("A"), TKN("B"))
450 451
        assert Grammar(prefixes)('A B').content == 'A B'
        assert Grammar(prefixes)('B A').content == 'B A'
452
        # aternative Form
453
        prefixes = SomeOf(Alternative(TKN("B"), TKN("A")))
454 455
        assert Grammar(prefixes)('A B').content == 'A B'
        assert Grammar(prefixes)('B').content == 'B'
456 457 458 459

    def test_someOf_redundance(self):
        """Test that one and the same parser may be listed several times
        and must be matched several times accordingly."""
460
        prefixes = SomeOf(TKN("A"), TKN("B"), TKN("A"))
461 462 463
        assert Grammar(prefixes)('A A B').content == 'A A B'
        assert Grammar(prefixes)('A B A').content == 'A B A'
        assert Grammar(prefixes)('B A A').content == 'B A A'
464 465 466
        assert Grammar(prefixes)('A B B').error_flag


467
class TestPopRetrieve:
468
    mini_language = r"""
469 470
        document       = { text | codeblock }
        codeblock      = delimiter { text | (!:delimiter delimiter_sign) } ::delimiter
471
        delimiter      = delimiter_sign  # never use delimiter between capture and pop except for retrival!
472
        delimiter_sign = /`+/
eckhart's avatar
eckhart committed
473
        text           = /[^`]+/
474
        """
475
    mini_lang2 = r"""
476 477 478 479 480 481 482 483
        @braces_filter=counterpart
        document       = { text | codeblock }
        codeblock      = braces { text | opening_braces | (!:braces closing_braces) } ::braces
        braces         = opening_braces
        opening_braces = /\{+/
        closing_braces = /\}+/
        text           = /[^{}]+/
        """
484
    mini_lang3 = r"""
485 486 487 488
        document       = { text | env }
        env            = (specialtag | opentag) text [closespecial | closetag]
        opentag        = "<" name ">"
        specialtag     = "<" /ABC/ !name ">"
eckhart's avatar
eckhart committed
489
        closetag       = close_slash | close_star
490 491 492 493 494 495 496 497
        close_slash    = "<" ::name "/>"
        close_star     = "<" ::name "*>"
        closespecial   = "<" /ABC/~ ">"
        name           = /\w+/~
        text           = /[^<>]+/
        """

    def setup(self):
498 499 500
        self.minilang_parser = grammar_provider(self.mini_language)()
        self.minilang_parser2 = grammar_provider(self.mini_lang2)()
        self.minilang_parser3 = grammar_provider(self.mini_lang3)()
501 502 503

    @staticmethod
    def opening_delimiter(node, name):
504
        return node.tag_name == name # and not isinstance(node.parser, Retrieve)
505 506 507

    @staticmethod
    def closing_delimiter(node):
508 509
        return node.tag_name in {':Pop', ':Retrieve'}
        # return isinstance(node.parser, Retrieve)
510 511 512 513 514 515 516 517 518

    def test_compile_mini_language(self):
        assert self.minilang_parser
        assert self.minilang_parser2
        assert self.minilang_parser3

    def test_stackhandling(self):
        ambigous_opening = "<ABCnormal> normal tag <ABCnormal*>"
        syntax_tree = self.minilang_parser3(ambigous_opening)
519
        assert not syntax_tree.error_flag, str(syntax_tree.errors_sorted)
520 521 522

        ambigous_opening = "<ABCnormal> normal tag <ABCnormal/>"
        syntax_tree = self.minilang_parser3(ambigous_opening)
523
        assert not syntax_tree.error_flag, str(syntax_tree.errors_sorted)
524 525 526

        forgot_closing_tag = "<em> where is the closing tag?"
        syntax_tree = self.minilang_parser3(forgot_closing_tag)
527
        assert syntax_tree.error_flag, str(syntax_tree.errors_sorted)
528 529 530

        proper = "<em> has closing tag <em/>"
        syntax_tree = self.minilang_parser3(proper)
531
        assert not syntax_tree.error_flag, str(syntax_tree.errors_sorted)
532 533 534

        proper = "<em> has closing tag <em*>"
        syntax_tree = self.minilang_parser3(proper)
535
        assert not syntax_tree.error_flag, str(syntax_tree.errors_sorted)
536

537
    def test_cache_neutrality(self):
538 539
        """Test that packrat-caching does not interfere with the variable-
        changing parsers: Capture and Retrieve."""
540
        lang = r"""
541 542 543 544 545 546 547 548 549
            text = opening closing
            opening = (unmarked_package | marked_package)
            closing = ::variable
            unmarked_package = package "."
            marked_package = package "*" "."
            package = "(" variable ")"
            variable = /\w+/~
            """
        case = "(secret)*. secret"
550
        gr = grammar_provider(lang)()
551
        st = gr(case)
552
        assert not st.error_flag, str(st.errors_sorted)
553 554 555 556

    def test_single_line(self):
        teststr = "Anfang ```code block `` <- keine Ende-Zeichen ! ``` Ende"
        syntax_tree = self.minilang_parser(teststr)
557
        assert not syntax_tree.errors_sorted
558 559
        delim = str(next(syntax_tree.select(partial(self.opening_delimiter, name="delimiter"))))
        pop = str(next(syntax_tree.select(self.closing_delimiter)))
560 561
        assert delim == pop
        if is_logging():
562
            log_ST(syntax_tree, "test_PopRetrieve_single_line.cst")
563 564 565 566 567 568 569 570 571 572 573

    def test_multi_line(self):
        teststr = """
            Anfang ```code block `` <- keine Ende-Zeichen ! ``` Ebde

            Absatz ohne ``` codeblock, aber
            das stellt sich erst am Ende herause...

            Mehrzeliger ```code block
            """
        syntax_tree = self.minilang_parser(teststr)
574
        assert not syntax_tree.errors_sorted
575 576
        delim = str(next(syntax_tree.select(partial(self.opening_delimiter, name="delimiter"))))
        pop = str(next(syntax_tree.select(self.closing_delimiter)))
577 578
        assert delim == pop
        if is_logging():
579
            log_ST(syntax_tree, "test_PopRetrieve_multi_line.cst")
580 581 582 583

    def test_single_line_complement(self):
        teststr = "Anfang {{{code block }} <- keine Ende-Zeichen ! }}} Ende"
        syntax_tree = self.minilang_parser2(teststr)
584
        assert not syntax_tree.errors_sorted
585 586
        delim = str(next(syntax_tree.select(partial(self.opening_delimiter, name="braces"))))
        pop = str(next(syntax_tree.select(self.closing_delimiter)))
587 588
        assert len(delim) == len(pop) and delim != pop
        if is_logging():
589
            log_ST(syntax_tree, "test_PopRetrieve_single_line.cst")
590 591 592 593 594 595 596 597 598 599 600

    def test_multi_line_complement(self):
        teststr = """
            Anfang {{{code block {{ <- keine Ende-Zeichen ! }}} Ende

            Absatz ohne {{{ codeblock, aber
            das stellt sich erst am Ende heraus...

            Mehrzeliger }}}code block
            """
        syntax_tree = self.minilang_parser2(teststr)
601
        assert not syntax_tree.errors_sorted
602 603
        delim = str(next(syntax_tree.select(partial(self.opening_delimiter, name="braces"))))
        pop = str(next(syntax_tree.select(self.closing_delimiter)))
604 605
        assert len(delim) == len(pop) and delim != pop
        if is_logging():
606
            log_ST(syntax_tree, "test_PopRetrieve_multi_line.cst")
607 608


609
class TestWhitespaceHandling:
Eckhart Arnold's avatar
Eckhart Arnold committed
610
    minilang = """
611 612 613 614 615 616 617 618 619
        doc = A B
        A = "A"
        B = "B"
        Rdoc = ar br
        ar = /A/
        br = /B/
        """

    def setup(self):
620
        self.gr = grammar_provider(self.minilang)()
621 622 623 624 625 626 627 628 629 630 631 632

    def test_token_whitespace(self):
        st = self.gr("AB", 'doc')
        assert not st.error_flag
        st = self.gr("A B", 'doc')
        assert not st.error_flag

    def test_regexp_whitespace(self):
        st = self.gr("AB", 'Rdoc')
        assert not st.error_flag
        st = self.gr("A B", 'Rdoc')
        assert st.error_flag
di68kap's avatar
di68kap committed
633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652


class TestErrorReporting:
    grammar = """
        root      = series alpha | anything
        series    = subseries &alpha 
        subseries = alpha §beta
        alpha     = /[a-z]+/
        beta      = /[A-Z]+/
        anything  = /.*/
        """

    def setup(self):
        self.parser = grammar_provider(self.grammar)()

    def test_error_propagation(self):
        testcode1 = "halloB"
        testcode2 = "XYZ"
        testcode3 = "hallo "
        cst = self.parser(testcode1)
653
        assert not cst.error_flag, str(cst.errors_sorted)
di68kap's avatar
di68kap committed
654 655 656 657
        cst = self.parser(testcode2)
        assert not cst.error_flag
        cst = self.parser(testcode3)
        assert cst.error_flag
658 659


di68kap's avatar
di68kap committed
660 661 662 663 664 665 666
class TestBorderlineCases:
    def test_not_matching(self):
        minilang = """parser = /X/"""
        gr = grammar_provider(minilang)()
        cst = gr('X', 'parser')
        assert not cst.error_flag
        cst = gr(' ', 'parser')
667
        assert cst.error_flag and cst.errors_sorted[0].code == Error.PARSER_DID_NOT_MATCH
di68kap's avatar
di68kap committed
668
        cst = gr('', 'parser')
669
        assert cst.error_flag and cst.errors_sorted[0].code == Error.PARSER_DID_NOT_MATCH
di68kap's avatar
di68kap committed
670 671 672 673 674 675 676

    def test_matching(self):
        minilang = """parser = /.?/"""
        gr = grammar_provider(minilang)()
        cst = gr(' ', 'parser')
        assert not cst.error_flag
        cst = gr('  ', 'parser')
677
        assert cst.error_flag and cst.errors_sorted[0].code == Error.PARSER_STOPPED_BEFORE_END
di68kap's avatar
di68kap committed
678 679 680 681
        cst = gr('', 'parser')
        assert not cst.error_flag


Eckhart Arnold's avatar
Eckhart Arnold committed
682
class TestReentryAfterError:
683
    def setup(self):
Eckhart Arnold's avatar
Eckhart Arnold committed
684 685 686 687 688
        lang = """
        document = alpha [beta] gamma "."
          alpha = "ALPHA" abc
            abc = §"a" "b" "c"
          beta = "BETA" (bac | bca)
689 690
            bac = "b" "a" §"c"
            bca = "b" "c" §"a"
Eckhart Arnold's avatar
Eckhart Arnold committed
691 692 693 694
          gamma = "GAMMA" §(cab | cba)
            cab = "c" "a" §"b"
            cba = "c" "b" §"a"
        """
695
        self.gr = grammar_provider(lang)()
696

697 698
    def test_no_resume_rules(self):
        gr = self.gr;  gr.resume_rules = dict()
699 700
        content = 'ALPHA acb BETA bac GAMMA cab .'
        cst = gr(content)
701 702 703 704
        assert cst.error_flag
        assert cst.content == content
        assert cst.pick('alpha').content.startswith('ALPHA')

eckhart's avatar
eckhart committed
705 706 707 708 709 710 711 712
    def test_no_resume_rules_partial_parsing(self):
        gr = self.gr;  gr.resume_rules = dict()
        content = 'ALPHA acb'
        cst = gr(content, 'alpha')
        assert cst.error_flag
        assert cst.content == content
        assert cst.pick('alpha').content.startswith('ALPHA')

713 714
    def test_simple_resume_rule(self):
        gr = self.gr;  gr.resume_rules = dict()
715
        gr.resume_rules__['alpha'] = ['BETA']
716
        content = 'ALPHA acb BETA bac GAMMA cab .'
717 718 719 720 721
        cst = gr(content)
        assert cst.error_flag
        assert cst.content == content
        assert cst.pick('alpha').content.startswith('ALPHA')
        # because of resuming, there should be only on error message
722
        assert len(cst.errors_sorted) == 1
723

724 725
    def test_failing_resume_rule(self):
        gr = self.gr;  gr.resume_rules = dict()
726
        gr.resume_rules__['alpha'] = ['XXX']
727
        content = 'ALPHA acb BETA bac GAMMA cab .'
728 729 730 731 732 733
        cst = gr(content)
        assert cst.error_flag
        assert cst.content == content
        # assert cst.pick('alpha').content.startswith('ALPHA')
        # because of resuming, there should be only on error message

734 735
    def test_severl_reentry_points(self):
        gr = self.gr;  gr.resume_rules = dict()
736
        gr.resume_rules__['alpha'] = ['BETA', 'GAMMA']
737
        content = 'ALPHA acb BETA bac GAMMA cab .'
738 739 740 741 742
        cst = gr(content)
        assert cst.error_flag
        assert cst.content == content
        assert cst.pick('alpha').content.startswith('ALPHA')
        # because of resuming, there should be only on error message
743
        assert len(cst.errors_sorted) == 1
744

745 746
    def test_several_reentry_points_second_point_matching(self):
        gr = self.gr;  gr.resume_rules = dict()
747 748 749
        gr.resume_rules__['alpha'] = ['BETA', 'GAMMA']
        content = 'ALPHA acb GAMMA cab .'
        cst = gr(content)
Eckhart Arnold's avatar
Eckhart Arnold committed
750
        assert cst.error_flag
751
        assert cst.content == content
752 753
        assert cst.pick('alpha').content.startswith('ALPHA')
        # because of resuming, there should be only on error message
754
        assert len(cst.errors_sorted) == 1
755

756 757 758 759 760 761 762 763 764 765 766
    def test_several_resume_rules_innermost_rule_matching(self):
        gr = self.gr;  gr.resume_rules = dict()
        gr.resume_rules__['alpha'] = ['BETA', 'GAMMA']
        gr.resume_rules__['beta'] = ['GAMMA']
        gr.resume_rules__['bac'] = ['GAMMA']
        content = 'ALPHA abc BETA bad GAMMA cab .'
        cst = gr(content)
        assert cst.error_flag
        assert cst.content == content
        assert cst.pick('alpha').content.startswith('ALPHA')
        # because of resuming, there should be only on error message
767
        assert len(cst.errors_sorted) == 1
768 769 770 771 772 773 774
        # multiple failures
        content = 'ALPHA acb BETA bad GAMMA cab .'
        cst = gr(content)
        assert cst.error_flag
        assert cst.content == content
        assert cst.pick('alpha').content.startswith('ALPHA')
        # because of resuming, there should be only on error message
775
        assert len(cst.errors_sorted) == 2
776

777

778 779 780 781 782 783 784 785 786 787
class TestConfiguredErrorMessages:
    def test_(self):
        lang = """
            document = series | /.*/
            @series_error = "a badly configured error message {5}"
            series = "X" | head §"C" "D"
            head = "A" "B"
            """
        parser = grammar_provider(lang)()
        st = parser("AB_D");  assert st.error_flag
788 789
        assert st.errors_sorted[0].code == Error.MALFORMED_ERROR_STRING
        assert st.errors_sorted[1].code == Error.MANDATORY_CONTINUATION
790 791


792 793 794 795 796 797 798 799 800 801
class TestUnknownParserError:
    def test_unknown_parser_error(self):
        gr = Grammar()
        try:
            gr("", "NonExistantParser")
            assert False, "UnknownParserError expected!"
        except UnknownParserError:
            pass


802 803
class TestEarlyTokenWhitespaceDrop:
    def setup(self):
804
        self.lang = r"""
805 806 807
            @ drop = token, whitespace
            expression = term  { ("+" | "-") term}
            term       = factor  { ("*"|"/") factor}
808 809
            factor     = number | variable | "("  expression  ")" 
                       | constant | fixed
810
            variable   = /[a-z]/~
811 812 813
            number     = /\d+/~
            constant   = "A" | "B"
            fixed      = "X"   
814 815 816
            """
        self.gr = grammar_provider(self.lang)()

817
    def test_drop(self):
818 819 820
        cst = self.gr('4 + 3 * 5')
        assert not cst.pick(':Token')
        assert not cst.pick(':Whitespace')
821 822 823 824 825 826 827 828
        cst = self.gr('A + B')
        try:
            _ = next(cst.select(lambda node: node.content == 'A'))
            assert False, "Tokens in compound expressions should be dropped!"
        except StopIteration:
            pass
        cst = self.gr('X * y')
        assert next(cst.select(lambda node: node.content == 'X'))
829

830

Eckhart Arnold's avatar
Eckhart Arnold committed
831
class TestMetaParser:
832 833 834 835 836 837 838
    def setup(self):
        self.mp = MetaParser()
        self.mp.grammar = Grammar()  # override placeholder warning
        self.mp.pname = "named"
        self.mp.tag_name = self.mp.pname

    def test_return_value(self):
839 840
        save = get_config_value('flatten_tree_while_parsing')
        set_config_value('flatten_tree_while_parsing', True)
841
        nd = self.mp._return_value(Node('tagged', 'non-empty'))
Eckhart Arnold's avatar
Eckhart Arnold committed
842 843 844 845
        assert nd.tag_name == 'named', nd.as_sxpr()
        assert len(nd.children) == 1
        assert nd.children[0].tag_name == 'tagged'
        assert nd.children[0].result == "non-empty"
846
        nd = self.mp._return_value(Node('tagged', ''))
Eckhart Arnold's avatar
Eckhart Arnold committed
847 848 849 850
        assert nd.tag_name == 'named', nd.as_sxpr()
        assert len(nd.children) == 1
        assert nd.children[0].tag_name == 'tagged'
        assert not nd.children[0].result
851
        nd = self.mp._return_value(Node(':anonymous', 'content'))
Eckhart Arnold's avatar
Eckhart Arnold committed
852 853 854
        assert nd.tag_name == 'named', nd.as_sxpr()
        assert not nd.children
        assert nd.result == 'content'
855
        nd = self.mp._return_value(Node(':anonymous', ''))
Eckhart Arnold's avatar
Eckhart Arnold committed
856 857 858
        assert nd.tag_name == 'named', nd.as_sxpr()
        assert not nd.children
        assert not nd.content
859 860 861 862 863
        nd = self.mp._return_value(EMPTY_NODE)
        assert nd.tag_name == 'named' and not nd.children, nd.as_sxpr()
        self.mp.pname = ''
        self.mp.tag_name = ':unnamed'
        nd = self.mp._return_value(Node('tagged', 'non-empty'))
Eckhart Arnold's avatar
Eckhart Arnold committed
864 865 866
        assert nd.tag_name == 'tagged', nd.as_sxpr()
        assert len(nd.children) == 0
        assert nd.content == 'non-empty'
867
        nd = self.mp._return_value(Node('tagged', ''))
Eckhart Arnold's avatar
Eckhart Arnold committed
868 869 870
        assert nd.tag_name == 'tagged', nd.as_sxpr()
        assert len(nd.children) == 0
        assert not nd.content
871
        nd = self.mp._return_value(Node(':anonymous', 'content'))
Eckhart Arnold's avatar
Eckhart Arnold committed
872 873 874
        assert nd.tag_name == ':anonymous', nd.as_sxpr()
        assert not nd.children
        assert nd.result == 'content'
875
        nd = self.mp._return_value(Node('', ''))
Eckhart Arnold's avatar
Eckhart Arnold committed
876 877 878
        assert nd.tag_name == '', nd.as_sxpr()
        assert not nd.children
        assert not nd.content
879 880
        assert self.mp._return_value(None) == EMPTY_NODE
        assert self.mp._return_value(EMPTY_NODE) == EMPTY_NODE
881
        set_config_value('flatten_tree_while_parsing', save)
Eckhart Arnold's avatar
Eckhart Arnold committed
882

883 884 885 886 887 888
    def test_return_values(self):
        self.mp.pname = "named"
        self.mp.tag_name = self.mp.pname
        rv = self.mp._return_values((Node('tag', 'content'), EMPTY_NODE))
        assert rv[-1].tag_name != EMPTY_NODE.tag_name, rv[-1].tag_name

889 890 891 892 893 894 895 896 897 898 899 900
    def test_in_context(self):
        minilang = """
            term       = factor  { (DIV|MUL) factor}
            factor     = NUMBER | VARIABLE
            MUL        = "*" | &factor
            DIV        = "/"            
            NUMBER     = /(?:0|(?:[1-9]\d*))(?:\.\d+)?/~
            VARIABLE   = /[A-Za-z]/~
            """
        gr = grammar_provider(minilang)()
        cst = gr("2x")
        assert bool(cst.pick('MUL'))
Eckhart Arnold's avatar
Eckhart Arnold committed
901

902
if __name__ == "__main__":
903
    from DHParser.testing import runner
904 905
    with logging(False):
        runner("", globals())