test_parse.py 38.6 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

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.
"""

22
import os
23
import sys
24
from functools import partial
25

26
27
scriptpath = os.path.dirname(__file__) or '.'
sys.path.append(os.path.abspath(os.path.join(scriptpath, '..')))
28

eckhart's avatar
eckhart committed
29
from DHParser.configuration import get_config_value, set_config_value
30
from DHParser.toolkit import compile_python_object, re
31
from DHParser.log import is_logging, log_ST, log_parsing_history
32
from DHParser.error import Error, is_error
33
from DHParser.parse import ParserError, Parser, Grammar, Forward, TKN, ZeroOrMore, RE, \
eckhart's avatar
eckhart committed
34
    RegExp, Lookbehind, NegativeLookahead, OneOrMore, Series, Alternative, AllOf, SomeOf, \
35
    UnknownParserError, MetaParser, EMPTY_NODE
36
from DHParser import compile_source
37
from DHParser.ebnf import get_ebnf_grammar, get_ebnf_transformer, get_ebnf_compiler, DHPARSER_IMPORTS
38
from DHParser.dsl import grammar_provider
39
from DHParser.syntaxtree import Node, parse_sxpr
40
from DHParser.stringview import StringView
41
42


43
44
45
46
47
48
49
50
51
52

class TestWhitespace:
    # TODO: add test cases here
    def test_whitespace_comment_mangling(self):
        pass

    def test_non_emptify_version(self):
        pass


53
54
class TestParserError:
    def test_parser_error_str(self):
55
        pe = ParserError(Node('TAG', 'test').with_pos(0), StringView('Beispiel'), None, True)
56
57
        assert str(pe).find('Beispiel') >= 0 and str(pe).find('TAG') >= 0

58
59
60
61
62
63
64
65
66
67
68
69
70
    def test_false_lookahead_only_message(self):
        """Error.PARSER_LOOKAHEAD_*_ONLY errors must not be reported if there
        no lookahead parser in the history!"""
        lang = """
        word = letters { letters | `-` letters }
        letters = /[A-Za-z]+/
        """
        gr = grammar_provider(lang)()
        st = gr('hard-time', track_history=True)
        assert not st.errors
        st = gr('hard-', track_history=True)
        assert st.errors and not any(e.code == 1045 for e in st.errors)

eckhart's avatar
eckhart committed
71

eckhart's avatar
eckhart committed
72
73
74
75
76
77
78
79
80
81
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
82
            l.append(p.pname + p.ptype)
eckhart's avatar
eckhart committed
83
84
85
86
87
88
89
90
91
92
93
        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


94
class TestInfiLoopsAndRecursion:
95
    def test_direct_left_recursion1(self):
eckhart's avatar
eckhart committed
96
        minilang = """
97
98
99
100
            expr = expr ("+"|"-") term | term
            term = term ("*"|"/") factor | factor
            factor = /[0-9]+/~
            """
101
        snippet = "9 + 8 + 7 + 6 + 5 + 3 * 4"
102
        parser = grammar_provider(minilang)()
103
        assert parser
Eckhart Arnold's avatar
Eckhart Arnold committed
104
        syntax_tree = parser(snippet)
105
106
        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
107
        if is_logging():
108
            log_ST(syntax_tree, "test_LeftRecursion_direct.cst")
109
            log_parsing_history(parser, "test_LeftRecursion_direct")
110

111
    def test_direct_left_recursion2(self):
112
113
114
115
116
117
        minilang = """
            expr = ex
            ex   = expr ("+"|"-") term | term
            term = term ("*"|"/") factor | factor
            factor = /[0-9]+/~
            """
118
        snippet = "9 + 8 + 7 + 6 + 5 + 3 * 4"
119
        parser = grammar_provider(minilang)()
120
        assert parser
121
        syntax_tree = parser(snippet)
122
123
        assert not is_error(syntax_tree.error_flag), syntax_tree.errors_sorted
        assert snippet == syntax_tree.content
124

125
    def test_indirect_left_recursion1(self):
126
127
        minilang = """
            Expr    = //~ (Product | Sum | Value)
128
129
            Product = Expr { ('*' | '/') Expr }+
            Sum     = Expr { ('+' | '-') Expr }+
130
            Value   = /[0-9.]+/~ | '(' §Expr ')'
131
            """
132
        parser = grammar_provider(minilang)()
133
        assert parser
134
        snippet = "8 * 4"
135
        syntax_tree = parser(snippet)
136
        assert not is_error(syntax_tree.error_flag), syntax_tree.errors_sorted
137
138
        snippet = "7 + 8 * 4"
        syntax_tree = parser(snippet)
139
        assert not is_error(syntax_tree.error_flag), syntax_tree.errors_sorted
140
141
        snippet = "9 + 8 * (4 + 3)"
        syntax_tree = parser(snippet)
142
143
        assert not is_error(syntax_tree.error_flag), syntax_tree.errors_sorted
        assert snippet == syntax_tree.content
144
        if is_logging():
145
            log_ST(syntax_tree, "test_LeftRecursion_indirect.cst")
146
            log_parsing_history(parser, "test_LeftRecursion_indirect")
147

148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
    # # 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
168

169
170
171
172
173
174
175
176
177
178
    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)

179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
    # 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)
204

eckhart's avatar
eckhart committed
205

Eckhart Arnold's avatar
Eckhart Arnold committed
206
class TestFlowControl:
207
208
    def setup(self):
        self.t1 = """
eckhart's avatar
eckhart committed
209
        All work and no play
210
211
212
213
214
        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
215
    def test_lookbehind(self):
216
        ws = RegExp(r'\s*')
Eckhart Arnold's avatar
Eckhart Arnold committed
217
        end = RegExp("END")
218
        doc_end = Lookbehind(RegExp('\\s*?\\n')) + end
219
        word = RegExp(r'\w+')
Eckhart Arnold's avatar
Eckhart Arnold committed
220
221
222
223
        sequence = OneOrMore(NegativeLookahead(end) + word + ws)
        document = ws + sequence + doc_end + ws

        parser = Grammar(document)
224
225
226
227
228
229
230
        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
231
            parser_initialization__ = ["upon instantiation"]
232
            ws = RegExp(r'\s*')
233
            end = RegExp('END')
234
            SUCC_LB = RegExp('\\s*?\\n')
235
            doc_end = Series(Lookbehind(SUCC_LB), end)
236
            word = RegExp(r'\w+')
237
238
239
240
241
242
            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
243
        assert not cst.error_flag, cst.as_sxpr()
244
        cst = parser(self.t2)
Eckhart Arnold's avatar
Eckhart Arnold committed
245
246
247
        assert cst.error_flag, cst.as_sxpr()


248
249
250
251
252
253
254
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
255
256
        result, messages, syntax_tree = compile_source(mlregex, None, get_ebnf_grammar(),
                        get_ebnf_transformer(), get_ebnf_compiler('MultilineRegexTest'))
257
        assert result
258
        assert not messages, str(messages)
259
        parser = compile_python_object(DHPARSER_IMPORTS + result, r'\w+Grammar$')()
eckhart's avatar
eckhart committed
260
261
        node = parser('abc+def', parser.regex)
        assert not node.error_flag
262
        assert node.tag_name == "regex"
263
264
        assert str(node) == 'abc+def'

265
266
    def test_multilineRegex_wo_Comments(self):
        mlregex = r"""
eckhart's avatar
eckhart committed
267
268
        regex =  /\w+
                  [+]
269
270
271
272
273
274
                  \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)
275
        parser = compile_python_object(DHPARSER_IMPORTS + result, r'\w+Grammar$')()
276
277
        node = parser('abc+def', parser.regex)
        assert not node.error_flag
278
        assert node.tag_name == "regex"
279
280
        assert str(node) == 'abc+def'

281
282
283
284
285
286
287
288
289
    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
290
        parser = compile_python_object(DHPARSER_IMPORTS + result, r'\w+Grammar$')()
291
292
293
294
        node, rest = parser.regex('Alpha')
        assert node
        assert not node.error_flag
        assert rest == ''
295
        assert node.tag_name == "regex"
296
297
298
299
300
301
302
303
304
305
        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
306
        parser = compile_python_object(DHPARSER_IMPORTS + result, r'\w+Grammar$')()
307
308
309
310
        node, rest = parser.regex('Alpha')
        assert node.error_flag


311
312
313
314
315
316
317
318
319
320
321
322
    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's avatar
eckhart committed
323
324
325
        result, messages, syntax_tree = compile_source(
            tokenlang, None, get_ebnf_grammar(), get_ebnf_transformer(),
            get_ebnf_compiler("TokenTest"))
326
        assert result
eckhart's avatar
eckhart committed
327
        assert not messages, str(messages)
328
        parser = compile_python_object(DHPARSER_IMPORTS + result, r'\w+Grammar$')()
Eckhart Arnold's avatar
Eckhart Arnold committed
329
        result = parser(testdoc)
330
        # log_parsing_history(parser, "test.log")
331
332
        assert not result.error_flag

333

334
class TestGrammar:
335
    def setup(self):
336
337
338
339
340
341
        grammar = r"""@whitespace = horizontal
        haupt        = textzeile LEERZEILE
        textzeile    = { WORT }+
        WORT         = /[^ \t]+/~
        LEERZEILE    = /\n[ \t]*(?=\n)/~
        """
342
343
344
        self.pyparser, messages, syntax_tree = compile_source(grammar, None, get_ebnf_grammar(),
                                                              get_ebnf_transformer(), get_ebnf_compiler("PosTest"))
        assert self.pyparser
345
        assert not messages
346
347
348
349

    def test_pos_values_initialized(self):
        # checks whether pos values in the parsing result and in the
        # history record have been initialized
350
351
        grammar = compile_python_object(DHPARSER_IMPORTS + self.pyparser, r'\w+Grammar$')()
        grammar("no_file_name*")
352
        for record in grammar.history__:
353
354
            assert not record.node or record.node.pos >= 0

355
    def test_select_parsing(self):
356
        grammar = compile_python_object(DHPARSER_IMPORTS + self.pyparser, r'\w+Grammar$')()
357
358
359
360
        grammar("wort", "WORT")
        grammar("eine Zeile", "textzeile")
        grammar("kein Haupt", "haupt")
        grammar("so ist es richtig", "haupt")
361

di68kap's avatar
di68kap committed
362
363
    def test_grammar_subclassing(self):
        class Arithmetic(Grammar):
364
            r'''
di68kap's avatar
di68kap committed
365
366
367
368
369
370
            expression =  term  { ("+" | "-") term }
            term       =  factor  { ("*" | "/") factor }
            factor     =  INTEGER | "("  expression  ")"
            INTEGER    =  /\d+/~
            '''
            expression = Forward()
371
372
373
374
            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
375
376
377
378
379
380
            root__ = expression

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

381
382
383
384
385
386
387
388
389
390
391
    def test_incomplete_matching(self):
        """Tests whether the flag `complete_match` works as expected when
        calling a grammar object in order to parse a document."""
        gr = grammar_provider('word = ~/\\w+/\n')()
        st = gr('eins')
        assert not st.errors
        st = gr('eins zwei')
        assert st.errors[0].code == Error.PARSER_STOPPED_BEFORE_END
        st = gr('eins zwei', complete_match=False)
        assert not st.errors

392

393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
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)()
416
417
418
        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
419
        assert st.errors_sorted[0].code == Error.MANDATORY_CONTINUATION
420
        # transitivity of mandatory-operator
421
        st = parser("ABC_");  assert st.error_flag
422
        assert st.errors_sorted[0].code == Error.MANDATORY_CONTINUATION
423
424

    def test_series_composition(self):
425
        TA, TB, TC, TD, TE = (TKN(b) for b in "ABCDE")
426
427
428
429
430
        s1 = Series(TA, TB, TC, mandatory=2)
        s2 = Series(TD, TE)

        combined = Alternative(s1 + s2, RegExp('.*'))
        parser = Grammar(combined)
431
432
433
        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
434
        assert st.errors_sorted[0].code == Error.MANDATORY_CONTINUATION
435
        st = parser("ABC_E");  assert st.error_flag
436
        assert st.errors_sorted[0].code == Error.MANDATORY_CONTINUATION
437
438
439

        combined = Alternative(s2 + s1, RegExp('.*'))
        parser = Grammar(combined)
440
441
442
443
444
445
        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
446
        assert st.errors_sorted[0].code == Error.MANDATORY_CONTINUATION
447

eckhart's avatar
eckhart committed
448
449
450
451
452
453
454
455
456
457
    # 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
458

459

460
461
462
class TestAllOfSomeOf:
    def test_allOf_order(self):
        """Test that parsers of an AllOf-List can match in arbitrary order."""
463
        prefixes = AllOf(TKN("A"), TKN("B"))
464
465
        assert Grammar(prefixes)('A B').content == 'A B'
        assert Grammar(prefixes)('B A').content == 'B A'
466
        # aternative Form
467
        prefixes = AllOf(Series(TKN("B"), TKN("A")))
468
        assert Grammar(prefixes)('A B').content == 'A B'
469
470
471
472

    def test_allOf_completeness(self):
        """Test that an error is raised if not  all parsers of an AllOf-List
        match."""
473
        prefixes = AllOf(TKN("A"), TKN("B"))
474
475
476
477
478
        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."""
479
        prefixes = AllOf(TKN("A"), TKN("B"), TKN("A"))
480
481
482
        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'
483
484
485
486
        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."""
487
        prefixes = SomeOf(TKN("A"), TKN("B"))
488
489
        assert Grammar(prefixes)('A B').content == 'A B'
        assert Grammar(prefixes)('B A').content == 'B A'
490
        # aternative Form
491
        prefixes = SomeOf(Alternative(TKN("B"), TKN("A")))
492
493
        assert Grammar(prefixes)('A B').content == 'A B'
        assert Grammar(prefixes)('B').content == 'B'
494
495
496
497

    def test_someOf_redundance(self):
        """Test that one and the same parser may be listed several times
        and must be matched several times accordingly."""
498
        prefixes = SomeOf(TKN("A"), TKN("B"), TKN("A"))
499
500
501
        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'
502
503
504
        assert Grammar(prefixes)('A B B').error_flag


505
506
507
508
class TestErrorRecovery:
    def test_series_skip(self):
        lang = """
        document = series | /.*/
509
        @series_skip = /(?=[A-Z])/
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
        series = "A" "B" §"C" "D"
        """
        parser = grammar_provider(lang)()
        st = parser('AB_D')
        assert len(st.errors) == 1  # no additional "stopped before end"-error!

    def test_AllOf_skip(self):
        lang = """
        document = allof | /.*/
        @allof_skip = /[A-Z]/
        allof = < "A" §"B" "C" "D" >
        """
        parser = grammar_provider(lang)()
        st = parser('CADB')
        assert 'allof' in st and st['allof'].content == "CADB"
        st = parser('_BCD')
        assert st.equals(parse_sxpr('(document "_BCD")'))
        st = parser('_ABC')
        assert st.equals(parse_sxpr('(document "_ABC")'))
        st = parser('A_CD')
        assert st['allof'].content == "A_CD"
        st = parser('AB_D')
        assert st['allof'].content == "AB_D"
        st = parser('A__D')
        assert st['allof'].content == "A__D"
        st = parser('CA_D')
        assert st['allof'].content == "CA_D"
        st = parser('BC_A')
        assert st['allof'].content == "BC_A"


541
class TestPopRetrieve:
542
    mini_language = r"""
543
544
        document       = { text | codeblock }
        codeblock      = delimiter { text | (!:delimiter delimiter_sign) } ::delimiter
545
        delimiter      = delimiter_sign  # never use delimiter between capture and pop except for retrival!
546
        delimiter_sign = /`+/
eckhart's avatar
eckhart committed
547
        text           = /[^`]+/
548
        """
549
    mini_lang2 = r"""
550
551
552
553
554
555
556
557
        @braces_filter=counterpart
        document       = { text | codeblock }
        codeblock      = braces { text | opening_braces | (!:braces closing_braces) } ::braces
        braces         = opening_braces
        opening_braces = /\{+/
        closing_braces = /\}+/
        text           = /[^{}]+/
        """
558
    mini_lang3 = r"""
559
560
561
562
        document       = { text | env }
        env            = (specialtag | opentag) text [closespecial | closetag]
        opentag        = "<" name ">"
        specialtag     = "<" /ABC/ !name ">"
eckhart's avatar
eckhart committed
563
        closetag       = close_slash | close_star
564
565
566
567
568
569
570
571
        close_slash    = "<" ::name "/>"
        close_star     = "<" ::name "*>"
        closespecial   = "<" /ABC/~ ">"
        name           = /\w+/~
        text           = /[^<>]+/
        """

    def setup(self):
572
573
574
        self.minilang_parser = grammar_provider(self.mini_language)()
        self.minilang_parser2 = grammar_provider(self.mini_lang2)()
        self.minilang_parser3 = grammar_provider(self.mini_lang3)()
575
576
577

    @staticmethod
    def opening_delimiter(node, name):
578
        return node.tag_name == name # and not isinstance(node.parser, Retrieve)
579
580
581

    @staticmethod
    def closing_delimiter(node):
582
583
        return node.tag_name in {':Pop', ':Retrieve'}
        # return isinstance(node.parser, Retrieve)
584
585
586
587
588
589
590
591
592

    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)
593
        assert not syntax_tree.error_flag, str(syntax_tree.errors_sorted)
594
595
596

        ambigous_opening = "<ABCnormal> normal tag <ABCnormal/>"
        syntax_tree = self.minilang_parser3(ambigous_opening)
597
        assert not syntax_tree.error_flag, str(syntax_tree.errors_sorted)
598
599
600

        forgot_closing_tag = "<em> where is the closing tag?"
        syntax_tree = self.minilang_parser3(forgot_closing_tag)
601
        assert syntax_tree.error_flag, str(syntax_tree.errors_sorted)
602
603
604

        proper = "<em> has closing tag <em/>"
        syntax_tree = self.minilang_parser3(proper)
605
        assert not syntax_tree.error_flag, str(syntax_tree.errors_sorted)
606
607
608

        proper = "<em> has closing tag <em*>"
        syntax_tree = self.minilang_parser3(proper)
609
        assert not syntax_tree.error_flag, str(syntax_tree.errors_sorted)
610

611
    def test_cache_neutrality(self):
612
613
        """Test that packrat-caching does not interfere with the variable-
        changing parsers: Capture and Retrieve."""
614
        lang = r"""
615
616
617
618
619
620
621
622
623
            text = opening closing
            opening = (unmarked_package | marked_package)
            closing = ::variable
            unmarked_package = package "."
            marked_package = package "*" "."
            package = "(" variable ")"
            variable = /\w+/~
            """
        case = "(secret)*. secret"
624
        gr = grammar_provider(lang)()
625
        st = gr(case)
626
        assert not st.error_flag, str(st.errors_sorted)
627
628
629
630

    def test_single_line(self):
        teststr = "Anfang ```code block `` <- keine Ende-Zeichen ! ``` Ende"
        syntax_tree = self.minilang_parser(teststr)
631
        assert not syntax_tree.errors_sorted
632
633
        delim = str(next(syntax_tree.select_if(partial(self.opening_delimiter, name="delimiter"))))
        pop = str(next(syntax_tree.select_if(self.closing_delimiter)))
634
635
        assert delim == pop
        if is_logging():
636
            log_ST(syntax_tree, "test_PopRetrieve_single_line.cst")
637
638
639
640
641
642
643
644
645
646
647

    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)
648
        assert not syntax_tree.errors_sorted
649
650
        delim = str(next(syntax_tree.select_if(partial(self.opening_delimiter, name="delimiter"))))
        pop = str(next(syntax_tree.select_if(self.closing_delimiter)))
651
652
        assert delim == pop
        if is_logging():
653
            log_ST(syntax_tree, "test_PopRetrieve_multi_line.cst")
654
655
656
657

    def test_single_line_complement(self):
        teststr = "Anfang {{{code block }} <- keine Ende-Zeichen ! }}} Ende"
        syntax_tree = self.minilang_parser2(teststr)
658
        assert not syntax_tree.errors_sorted
659
660
        delim = str(next(syntax_tree.select_if(partial(self.opening_delimiter, name="braces"))))
        pop = str(next(syntax_tree.select_if(self.closing_delimiter)))
661
662
        assert len(delim) == len(pop) and delim != pop
        if is_logging():
663
            log_ST(syntax_tree, "test_PopRetrieve_single_line.cst")
664
665
666
667
668
669
670
671
672
673
674

    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)
675
        assert not syntax_tree.errors_sorted
676
677
        delim = str(next(syntax_tree.select_if(partial(self.opening_delimiter, name="braces"))))
        pop = str(next(syntax_tree.select_if(self.closing_delimiter)))
678
679
        assert len(delim) == len(pop) and delim != pop
        if is_logging():
680
            log_ST(syntax_tree, "test_PopRetrieve_multi_line.cst")
681
682


683
class TestWhitespaceHandling:
Eckhart Arnold's avatar
Eckhart Arnold committed
684
    minilang = """
685
686
687
688
689
690
691
692
693
        doc = A B
        A = "A"
        B = "B"
        Rdoc = ar br
        ar = /A/
        br = /B/
        """

    def setup(self):
694
        self.gr = grammar_provider(self.minilang)()
695
696
697
698
699
700
701
702
703
704
705
706

    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
707
708
709
710
711


class TestErrorReporting:
    grammar = """
        root      = series alpha | anything
eckhart's avatar
eckhart committed
712
        series    = subseries &alpha
di68kap's avatar
di68kap committed
713
714
715
716
717
718
719
720
721
722
723
724
725
726
        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)
727
        assert not cst.error_flag, str(cst.errors_sorted)
di68kap's avatar
di68kap committed
728
729
730
731
        cst = self.parser(testcode2)
        assert not cst.error_flag
        cst = self.parser(testcode3)
        assert cst.error_flag
732
733


di68kap's avatar
di68kap committed
734
735
class TestBorderlineCases:
    def test_not_matching(self):
736
        minilang = """parser = /X/\n"""
di68kap's avatar
di68kap committed
737
738
739
740
        gr = grammar_provider(minilang)()
        cst = gr('X', 'parser')
        assert not cst.error_flag
        cst = gr(' ', 'parser')
741
        assert cst.error_flag and cst.errors_sorted[0].code == Error.PARSER_DID_NOT_MATCH
di68kap's avatar
di68kap committed
742
        cst = gr('', 'parser')
743
        assert cst.error_flag and cst.errors_sorted[0].code == Error.PARSER_DID_NOT_MATCH
di68kap's avatar
di68kap committed
744
745
746
747
748
749
750

    def test_matching(self):
        minilang = """parser = /.?/"""
        gr = grammar_provider(minilang)()
        cst = gr(' ', 'parser')
        assert not cst.error_flag
        cst = gr('  ', 'parser')
751
        assert cst.error_flag and cst.errors_sorted[0].code == Error.PARSER_STOPPED_BEFORE_END
di68kap's avatar
di68kap committed
752
753
754
755
        cst = gr('', 'parser')
        assert not cst.error_flag


Eckhart Arnold's avatar
Eckhart Arnold committed
756
class TestReentryAfterError:
757
    def setup(self):
Eckhart Arnold's avatar
Eckhart Arnold committed
758
759
760
761
762
        lang = """
        document = alpha [beta] gamma "."
          alpha = "ALPHA" abc
            abc = §"a" "b" "c"
          beta = "BETA" (bac | bca)
763
764
            bac = "b" "a" §"c"
            bca = "b" "c" §"a"
Eckhart Arnold's avatar
Eckhart Arnold committed
765
766
767
768
          gamma = "GAMMA" §(cab | cba)
            cab = "c" "a" §"b"
            cba = "c" "b" §"a"
        """
769
        self.gr = grammar_provider(lang)()
770

771
772
    def test_no_resume_rules(self):
        gr = self.gr;  gr.resume_rules = dict()
773
774
        content = 'ALPHA acb BETA bac GAMMA cab .'
        cst = gr(content)
775
776
777
778
        assert cst.error_flag
        assert cst.content == content
        assert cst.pick('alpha').content.startswith('ALPHA')

eckhart's avatar
eckhart committed
779
780
781
782
783
784
785
786
    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')

787
788
    def test_simple_resume_rule(self):
        gr = self.gr;  gr.resume_rules = dict()
789
        gr.resume_rules__['alpha'] = [re.compile(r'(?=BETA)')]
790
        content = 'ALPHA acb BETA bac GAMMA cab .'
791
792
793
794
        cst = gr(content)
        assert cst.error_flag
        assert cst.content == content
        assert cst.pick('alpha').content.startswith('ALPHA')
795
        # because of resuming, there should be only one error message
796
        assert len(cst.errors_sorted) == 1
797

798
799
    def test_failing_resume_rule(self):
        gr = self.gr;  gr.resume_rules = dict()
800
        gr.resume_rules__['alpha'] = [re.compile(r'(?=XXX)')]
801
        content = 'ALPHA acb BETA bac GAMMA cab .'
802
803
804
805
806
        cst = gr(content)
        assert cst.error_flag
        assert cst.content == content
        # assert cst.pick('alpha').content.startswith('ALPHA')

807
808
    def test_severl_reentry_points(self):
        gr = self.gr;  gr.resume_rules = dict()
809
        gr.resume_rules__['alpha'] = [re.compile(r'(?=BETA)'), re.compile(r'(?=GAMMA)')]
810
        content = 'ALPHA acb BETA bac GAMMA cab .'
811
812
813
814
        cst = gr(content)
        assert cst.error_flag
        assert cst.content == content
        assert cst.pick('alpha').content.startswith('ALPHA')
815
        # because of resuming, there should be only one error message
816
        assert len(cst.errors_sorted) == 1
817

818
819
    def test_several_reentry_points_second_point_matching(self):
        gr = self.gr;  gr.resume_rules = dict()
820
        gr.resume_rules__['alpha'] = [re.compile(r'(?=BETA)'), re.compile(r'(?=GAMMA)')]
821
822
        content = 'ALPHA acb GAMMA cab .'
        cst = gr(content)
Eckhart Arnold's avatar
Eckhart Arnold committed
823
        assert cst.error_flag
824
        assert cst.content == content
825
        assert cst.pick('alpha').content.startswith('ALPHA')
826
        # because of resuming, there should be only one error message
827
        assert len(cst.errors_sorted) == 1
828

829
830
    def test_several_resume_rules_innermost_rule_matching(self):
        gr = self.gr;  gr.resume_rules = dict()
831
832
833
        gr.resume_rules__['alpha'] = [re.compile(r'(?=BETA)'), re.compile(r'(?=GAMMA)')]
        gr.resume_rules__['beta'] = [re.compile(r'(?=GAMMA)')]
        gr.resume_rules__['bac'] = [re.compile(r'(?=GAMMA)')]
834
835
836
837
838
        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')
839
        # because of resuming, there should be only one error message
840
        assert len(cst.errors_sorted) == 1
841
842
843
844
845
846
        # 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')
847
        # there should be only two error messages
848
        assert len(cst.errors_sorted) == 2
849

850
    def test_skip_comment_on_resume(self):
851
852
        lang = r"""
            @ comment =  /(?:\/\/.*)|(?:\/\*(?:.|\n)*?\*\/)/  # Kommentare im C++-Stil
853
            document = block_A block_B
854
            @ block_A_resume = /(?=x)/
855
856
            block_A = "a" §"b" "c"
            block_B = "x" "y" "z"
857
        """
858
859
860
861
862
863
864
865
866
867
868
869
        def mini_suite(grammar):
            tree = grammar('abc/*x*/xyz')
            assert not tree.errors
            tree = grammar('abDxyz')
            mandatory_cont = (Error.MANDATORY_CONTINUATION, Error.MANDATORY_CONTINUATION_AT_EOF)
            assert len(tree.errors) == 1 and tree.errors[0].code in mandatory_cont
            tree = grammar('abD/*x*/xyz')
            assert len(tree.errors) == 1 and tree.errors[0].code in mandatory_cont
            tree = grammar('aD /*x*/ c /* a */ /*x*/xyz')
            assert len(tree.errors) == 1 and tree.errors[0].code in mandatory_cont

        # test regex-defined resume rule
870
        grammar = grammar_provider(lang)()
871
        mini_suite(grammar)
872

873

874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
    def test_unambiguous_error_location(self):
        lang = r"""
            @ drop        = whitespace, token  # drop tokens and whitespace early
           
            @object_resume = /(?<=\})/
           
            json       = ~ value EOF
            value      = object | string 
            object     = "{" [member { "," §member }] "}"
            member     = string §":" value
            string     = `"` CHARACTERS `"` ~

            CHARACTERS = { /[^"\\]+/ }                  
            EOF      =  !/./        # no more characters ahead, end of file reached
            """
        test_case = """{
                "missing member": "abcdef",
            }"""
        gr = grammar_provider(lang)()
        cst = gr(test_case)
        assert any(err.code == Error.MANDATORY_CONTINUATION for err in cst.errors)


897
class TestConfiguredErrorMessages:
898
    def test_configured_error_message(self):
899
900
901
        lang = """
            document = series | /.*/
            @series_error = "a badly configured error message {5}"
902
            series = /X/ | head §"C" "D"
903
904
905
906
            head = "A" "B"
            """
        parser = grammar_provider(lang)()
        st = parser("AB_D");  assert st.error_flag
907
908
        assert st.errors_sorted[0].code == Error.MALFORMED_ERROR_STRING
        assert st.errors_sorted[1].code == Error.MANDATORY_CONTINUATION
909
910


911
912
913
914
915
916
917
918
919
920
class TestUnknownParserError:
    def test_unknown_parser_error(self):
        gr = Grammar()
        try:
            gr("", "NonExistantParser")
            assert False, "UnknownParserError expected!"
        except UnknownParserError:
            pass


921
922
class TestEarlyTokenWhitespaceDrop:
    def setup(self):
923
        self.lang = r"""
924
925
926
            @ drop = token, whitespace
            expression = term  { ("+" | "-") term}
            term       = factor  { ("*"|"/") factor}
eckhart's avatar
eckhart committed
927
            factor     = number | variable | "("  expression  ")"
928
                       | constant | fixed
929
            variable   = /[a-z]/~
930
931
            number     = /\d+/~
            constant   = "A" | "B"
eckhart's avatar
eckhart committed
932
            fixed      = "X"
933
934
935
            """
        self.gr = grammar_provider(self.lang)()

936
    def test_drop(self):
937
938
939
        cst = self.gr('4 + 3 * 5')
        assert not cst.pick(':Token')
        assert not cst.pick(':Whitespace')
940
941
        cst = self.gr('A + B')
        try:
942
            _ = next(cst.select_if(lambda node: node.content == 'A'))
943
944
945
946
            assert False, "Tokens in compound expressions should be dropped!"
        except StopIteration:
            pass
        cst = self.gr('X * y')
947
        assert next(cst.select_if(lambda node: node.content == 'X'))
948

949

Eckhart Arnold's avatar
Eckhart Arnold committed
950
class TestMetaParser:
951
952
953
954
    def setup(self):
        self.mp = MetaParser()
        self.mp.grammar = Grammar()  # override placeholder warning
        self.mp.pname = "named"
955
        self.mp.anonymous = False
956
957
958
        self.mp.tag_name = self.mp.pname

    def test_return_value(self):
959
960
        save = get_config_value('flatten_tree_while_parsing')
        set_config_value('flatten_tree_while_parsing', True)
961
        nd = self.mp._return_value(Node('tagged', 'non-empty'))
Eckhart Arnold's avatar
Eckhart Arnold committed
962
963
964
965
        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"
966
        nd = self.mp._return_value(Node('tagged', ''))
Eckhart Arnold's avatar
Eckhart Arnold committed
967
968
969
970
        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
971
        nd = self.mp._return_value(Node(':anonymous', 'content'))
Eckhart Arnold's avatar
Eckhart Arnold committed
972
973
974
        assert nd.tag_name == 'named', nd.as_sxpr()
        assert not nd.children
        assert nd.result == 'content'
975
        nd = self.mp._return_value(Node(':anonymous', ''))
Eckhart Arnold's avatar
Eckhart Arnold committed
976
977
978
        assert nd.tag_name == 'named', nd.as_sxpr()
        assert not nd.children
        assert not nd.content
979
980
981
        nd = self.mp._return_value(EMPTY_NODE)
        assert nd.tag_name == 'named' and not nd.children, nd.as_sxpr()
        self.mp.pname = ''
982
        self.mp.anonymous = True
983
984
        self.mp.tag_name = ':unnamed'
        nd = self.mp._return_value(Node('tagged', 'non-empty'))
Eckhart Arnold's avatar
Eckhart Arnold committed
985
986
987
        assert nd.tag_name == 'tagged', nd.as_sxpr()
        assert len(nd.children) == 0
        assert nd.content == 'non-empty'
988
        nd = self.mp._return_value(Node('tagged', ''))
Eckhart Arnold's avatar
Eckhart Arnold committed
989
990
991
        assert nd.tag_name == 'tagged', nd.as_sxpr()
        assert len(nd.children) == 0
        assert not nd.content
992
        nd = self.mp._return_value(Node(':anonymous', 'content'))
Eckhart Arnold's avatar
Eckhart Arnold committed
993
994
995
        assert nd.tag_name == ':anonymous', nd.as_sxpr()
        assert not nd.children
        assert nd.result == 'content'
996
        nd = self.mp._return_value(Node('', ''))
Eckhart Arnold's avatar
Eckhart Arnold committed
997
998
999
        assert nd.tag_name == '', nd.as_sxpr()
        assert not nd.children
        assert not nd.content
1000
        assert self.mp._return_value(None) == EMPTY_NODE