test_parse.py 49.9 KB
Newer Older
1
#!/usr/bin/env python3
2

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.
"""

eckhart's avatar
eckhart committed
22
import copy
23
import os
24
import sys
25
from functools import partial
26
from typing import List
27

28
29
scriptpath = os.path.dirname(__file__) or '.'
sys.path.append(os.path.abspath(os.path.join(scriptpath, '..')))
30

eckhart's avatar
eckhart committed
31
from DHParser.configuration import get_config_value, set_config_value
32
from DHParser.toolkit import compile_python_object, re
33
from DHParser.log import is_logging, log_ST, log_parsing_history
eckhart's avatar
eckhart committed
34
from DHParser.error import Error, is_error, adjust_error_locations
35
from DHParser.parse import ParserError, Parser, Grammar, Forward, TKN, ZeroOrMore, RE, \
eckhart's avatar
eckhart committed
36
    RegExp, Lookbehind, NegativeLookahead, OneOrMore, Series, Alternative, \
37
38
    Interleave, UnknownParserError, MetaParser, Token, EMPTY_NODE, Capture, Drop, Whitespace, \
    GrammarError
39
from DHParser import compile_source
40
from DHParser.ebnf import get_ebnf_grammar, get_ebnf_transformer, get_ebnf_compiler, \
eckhart's avatar
eckhart committed
41
    parse_ebnf, DHPARSER_IMPORTS
42
from DHParser.dsl import grammar_provider
43
from DHParser.syntaxtree import Node, parse_sxpr
44
from DHParser.stringview import StringView
45
from DHParser.trace import set_tracer, trace_history, resume_notices_on
46
47


48
49
50
51
52
53

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

Eckhart Arnold's avatar
Eckhart Arnold committed
54
    def test_non_empty_derivation(self):
55
56
        pass

57

58
59
class TestParserError:
    def test_parser_error_str(self):
60
        pe = ParserError(Node('TAG', 'test').with_pos(0), StringView('Beispiel'), None, True)
61
62
        assert str(pe).find('Beispiel') >= 0 and str(pe).find('TAG') >= 0

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)()
71
72
        set_tracer(gr, trace_history)
        st = gr('hard-time')
73
        assert not st.errors
74
        st = gr('hard-')
75
76
        assert st.errors and not any(e.code == 1045 for e in st.errors)

eckhart's avatar
eckhart committed
77

eckhart's avatar
eckhart committed
78
79
80
81
82
83
84
85
86
class TestParserClass:
    def test_apply(self):
        minilang ="""
            expr = expr ("+"|"-") term | term
            term = term ("*"|"/") factor | factor
            factor = /[0-9]+/~
            """
        gr = grammar_provider(minilang)()
        l = []
87
88
        def visitor(context: List[Parser]):
            p = context[-1]
di68kap's avatar
di68kap committed
89
            l.append(p.pname + p.ptype)
eckhart's avatar
eckhart committed
90
91
92
93
94
95
96
97
98
99
        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

100
101
102
103
104
105
106
107
    def test_symbol(self):
        class MyGrammar(Grammar):
            wrong = Token('wrong')
            word = OneOrMore(wrong) + Whitespace(r'\s*') + OneOrMore(RegExp(r'\w+'))
            root__ = word
        gr = MyGrammar()
        regex = gr['word'].parsers[-1].parser
        result = gr.associated_symbol(regex).symbol
108
        assert result == 'word', result
109

eckhart's avatar
eckhart committed
110

111
class TestInfiLoopsAndRecursion:
112
    def test_direct_left_recursion1(self):
eckhart's avatar
eckhart committed
113
        minilang = """
114
115
116
117
            expr = 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
Eckhart Arnold's avatar
Eckhart Arnold committed
121
        syntax_tree = parser(snippet)
122
123
        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
124
        if is_logging():
125
            log_ST(syntax_tree, "test_LeftRecursion_direct.cst")
126
            log_parsing_history(parser, "test_LeftRecursion_direct")
127

128
    def test_direct_left_recursion2(self):
129
130
131
132
133
134
        minilang = """
            expr = ex
            ex   = expr ("+"|"-") term | term
            term = term ("*"|"/") factor | factor
            factor = /[0-9]+/~
            """
135
        snippet = "9 + 8 + 7 + 6 + 5 + 3 * 4"
136
        parser = grammar_provider(minilang)()
137
        assert parser
138
        syntax_tree = parser(snippet)
139
140
        assert not is_error(syntax_tree.error_flag), syntax_tree.errors_sorted
        assert snippet == syntax_tree.content
141

142
    def test_indirect_left_recursion1(self):
143
144
        minilang = """
            Expr    = //~ (Product | Sum | Value)
145
146
            Product = Expr { ('*' | '/') Expr }+
            Sum     = Expr { ('+' | '-') Expr }+
147
            Value   = /[0-9.]+/~ | '(' §Expr ')'
148
            """
149
        parser = grammar_provider(minilang)()
150
        assert parser
151
        snippet = "8 * 4"
152
        syntax_tree = parser(snippet)
153
        assert not is_error(syntax_tree.error_flag), syntax_tree.errors_sorted
154
155
        snippet = "7 + 8 * 4"
        syntax_tree = parser(snippet)
156
        assert not is_error(syntax_tree.error_flag), syntax_tree.errors_sorted
157
158
        snippet = "9 + 8 * (4 + 3)"
        syntax_tree = parser(snippet)
159
160
        assert not is_error(syntax_tree.error_flag), syntax_tree.errors_sorted
        assert snippet == syntax_tree.content
161
        if is_logging():
162
            log_ST(syntax_tree, "test_LeftRecursion_indirect.cst")
163
            log_parsing_history(parser, "test_LeftRecursion_indirect")
164

165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
    # # 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
185

186
187
188
    def test_break_inifnite_loop_ZeroOrMore(self):
        forever = ZeroOrMore(RegExp(''))
        result = Grammar(forever)('')  # infinite loops will automatically be broken
189
        assert repr(result) == "Node(':EMPTY', '')", repr(result)
190
191
192
193

    def test_break_inifnite_loop_OneOrMore(self):
        forever = OneOrMore(RegExp(''))
        result = Grammar(forever)('')  # infinite loops will automatically be broken
194
        assert repr(result) == "Node(':EMPTY', '')", repr(result)
195

196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
    # 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)
221

eckhart's avatar
eckhart committed
222

Eckhart Arnold's avatar
Eckhart Arnold committed
223
class TestFlowControl:
224
225
    def setup(self):
        self.t1 = """
eckhart's avatar
eckhart committed
226
        All work and no play
227
228
229
230
231
        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
232
    def test_lookbehind(self):
233
        ws = RegExp(r'\s*')
Eckhart Arnold's avatar
Eckhart Arnold committed
234
        end = RegExp("END")
235
        doc_end = Lookbehind(RegExp('\\s*?\\n')) + end
236
        word = RegExp(r'\w+')
Eckhart Arnold's avatar
Eckhart Arnold committed
237
238
239
240
        sequence = OneOrMore(NegativeLookahead(end) + word + ws)
        document = ws + sequence + doc_end + ws

        parser = Grammar(document)
241
242
243
244
245
246
247
        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
248
            parser_initialization__ = ["upon instantiation"]
249
            ws = RegExp(r'\s*')
250
            end = RegExp('END')
251
            SUCC_LB = RegExp('\\s*?\\n')
252
            doc_end = Series(Lookbehind(SUCC_LB), end)
253
            word = RegExp(r'\w+')
254
255
256
257
258
259
            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
260
        assert not cst.error_flag, cst.as_sxpr()
261
        cst = parser(self.t2)
Eckhart Arnold's avatar
Eckhart Arnold committed
262
263
264
        assert cst.error_flag, cst.as_sxpr()


265
266
267
268
269
270
271
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
272
        result, messages, _ = compile_source(mlregex, None, get_ebnf_grammar(),
Eckhart Arnold's avatar
Eckhart Arnold committed
273
                        get_ebnf_transformer(), get_ebnf_compiler('MultilineRegexTest'))
274
        assert result
275
        assert not messages, str(messages)
276
        parser = compile_python_object(DHPARSER_IMPORTS + result, r'\w+Grammar$')()
eckhart's avatar
eckhart committed
277
278
        node = parser('abc+def', parser.regex)
        assert not node.error_flag
279
        assert node.tag_name == "regex"
280
281
        assert str(node) == 'abc+def'

282
283
    def test_multilineRegex_wo_Comments(self):
        mlregex = r"""
eckhart's avatar
eckhart committed
284
285
        regex =  /\w+
                  [+]
286
287
                  \w* /
        """
Eckhart Arnold's avatar
Eckhart Arnold committed
288
        result, messages, _ = compile_source(mlregex, None, get_ebnf_grammar(),
289
290
291
                        get_ebnf_transformer(), get_ebnf_compiler('MultilineRegexTest'))
        assert result
        assert not messages, str(messages)
292
        parser = compile_python_object(DHPARSER_IMPORTS + result, r'\w+Grammar$')()
293
294
        node = parser('abc+def', parser.regex)
        assert not node.error_flag
295
        assert node.tag_name == "regex"
296
297
        assert str(node) == 'abc+def'

298
    def test_ignore_case(self):
299
300
301
302
        mlregex = r"""
        @ ignorecase = True
        regex = /alpha/
        """
Eckhart Arnold's avatar
Eckhart Arnold committed
303
        result, messages, _ = compile_source(mlregex, None, get_ebnf_grammar(),
304
305
306
                        get_ebnf_transformer(), get_ebnf_compiler('MultilineRegexTest'))
        assert result
        assert not messages
307
        parser = compile_python_object(DHPARSER_IMPORTS + result, r'\w+Grammar$')()
308
        node, rest = parser.regex(StringView('Alpha'))
309
310
        assert node
        assert rest == ''
311
        assert node.tag_name == "regex"
312
313
314
315
316
317
        assert str(node) == 'Alpha'

        mlregex = r"""
        @ ignorecase = False
        regex = /alpha/
        """
Eckhart Arnold's avatar
Eckhart Arnold committed
318
        result, messages, _ = compile_source(mlregex, None, get_ebnf_grammar(),
319
320
321
                        get_ebnf_transformer(), get_ebnf_compiler('MultilineRegexTest'))
        assert result
        assert not messages
322
        parser = compile_python_object(DHPARSER_IMPORTS + result, r'\w+Grammar$')()
323
324
        node, rest = parser.regex(StringView('Alpha'))
        assert node is None
325

326
327
328
329
330
331
332
333
334
335
336
337
    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
338
        result, messages, _ = compile_source(
eckhart's avatar
eckhart committed
339
340
            tokenlang, None, get_ebnf_grammar(), get_ebnf_transformer(),
            get_ebnf_compiler("TokenTest"))
341
        assert result
eckhart's avatar
eckhart committed
342
        assert not messages, str(messages)
343
        parser = compile_python_object(DHPARSER_IMPORTS + result, r'\w+Grammar$')()
Eckhart Arnold's avatar
Eckhart Arnold committed
344
        result = parser(testdoc)
345
        # log_parsing_history(parser, "test.log")
346
        assert not result.error_flag, str(result.errors_sorted)
347

348

349
class TestGrammar:
350
    def setup(self):
351
352
353
354
355
356
        grammar = r"""@whitespace = horizontal
        haupt        = textzeile LEERZEILE
        textzeile    = { WORT }+
        WORT         = /[^ \t]+/~
        LEERZEILE    = /\n[ \t]*(?=\n)/~
        """
Eckhart Arnold's avatar
Eckhart Arnold committed
357
358
        self.pyparser, messages, _ = compile_source(grammar, None, get_ebnf_grammar(),
                                    get_ebnf_transformer(), get_ebnf_compiler("PosTest"))
359
        assert self.pyparser
360
        assert not messages
361
362
363
364

    def test_pos_values_initialized(self):
        # checks whether pos values in the parsing result and in the
        # history record have been initialized
365
366
        grammar = compile_python_object(DHPARSER_IMPORTS + self.pyparser, r'\w+Grammar$')()
        grammar("no_file_name*")
367
        for record in grammar.history__:
368
369
            assert not record.node or record.node.pos >= 0

370
    def test_select_parsing(self):
371
        grammar = compile_python_object(DHPARSER_IMPORTS + self.pyparser, r'\w+Grammar$')()
372
373
374
375
        grammar("wort", "WORT")
        grammar("eine Zeile", "textzeile")
        grammar("kein Haupt", "haupt")
        grammar("so ist es richtig", "haupt")
376

di68kap's avatar
di68kap committed
377
378
    def test_grammar_subclassing(self):
        class Arithmetic(Grammar):
379
            r'''
di68kap's avatar
di68kap committed
380
381
382
383
384
385
            expression =  term  { ("+" | "-") term }
            term       =  factor  { ("*" | "/") factor }
            factor     =  INTEGER | "("  expression  ")"
            INTEGER    =  /\d+/~
            '''
            expression = Forward()
386
387
388
389
            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
390
391
392
393
394
395
            root__ = expression

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

396
397
398
399
400
401
402
403
404
405
406
    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

407
    def test_synonym(self):
di68kap's avatar
di68kap committed
408
        lang = r"""
409
410
411
412
413
414
415
416
417
418
419
420
            doc  = { word | number }
            word = /\w+/ S
            number = [VZ] /\d+/ S 
            S    = ~        # let S by a synonym for anonymous whitespace
            VZ   = "-"
        """
        gr = grammar_provider(lang)()
        st = gr('eins 1 zwei2drei 3')
        # set_config_value('compiled_EBNF_log', 'grammar.log')
        gr = grammar_provider("@drop = whitespace, token" + lang)()
        st = gr('eins 1 zwei2drei 3')
        st = gr('-3')
421
        assert str(gr['S']) == "S = ~", str(gr['S'])
422

423

424
425
426
427
428
429
430
class TestSeries:
    def test_non_mandatory(self):
        lang = """
        document = series | /.*/
        series = "A" "B" "C" "D"
        """
        parser = grammar_provider(lang)()
Eckhart Arnold's avatar
Eckhart Arnold committed
431
        st = parser("ABCD")
432
        assert not st.error_flag
Eckhart Arnold's avatar
Eckhart Arnold committed
433
        st = parser("A_CD")
434
        assert not st.error_flag
Eckhart Arnold's avatar
Eckhart Arnold committed
435
        st = parser("AB_D")
436
437
438
439
440
441
442
443
444
445
446
        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)()
447
448
449
        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
450
        assert st.errors_sorted[0].code == Error.MANDATORY_CONTINUATION
451
        # transitivity of mandatory-operator
452
        st = parser("ABC_");  assert st.error_flag
453
        assert st.errors_sorted[0].code == Error.MANDATORY_CONTINUATION
454
455

    def test_series_composition(self):
456
        TA, TB, TC, TD, TE = (TKN(b) for b in "ABCDE")
457
458
459
460
461
        s1 = Series(TA, TB, TC, mandatory=2)
        s2 = Series(TD, TE)

        combined = Alternative(s1 + s2, RegExp('.*'))
        parser = Grammar(combined)
462
463
464
        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
465
        assert st.errors_sorted[0].code == Error.MANDATORY_CONTINUATION
466
        st = parser("ABC_E");  assert st.error_flag
467
        assert st.errors_sorted[0].code == Error.MANDATORY_CONTINUATION
468
469
470

        combined = Alternative(s2 + s1, RegExp('.*'))
        parser = Grammar(combined)
471
472
473
474
475
476
        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
477
        assert st.errors_sorted[0].code == Error.MANDATORY_CONTINUATION
478

eckhart's avatar
eckhart committed
479
480
481
482
483
484
485
486
487
488
    # 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
489

490
491
    def test_ebnf_serialization(self):
        ebnf_grammar = get_ebnf_grammar()
eckhart's avatar
eckhart committed
492
        # TODO: Add test here
493
        ebnf = ebnf_grammar.as_ebnf()
494
        # print(ebnf)
495
496


497
498
499
class TestAllOfSomeOf:
    def test_allOf_order(self):
        """Test that parsers of an AllOf-List can match in arbitrary order."""
eckhart's avatar
eckhart committed
500
        prefixes = Interleave(TKN("A"), TKN("B"))
501
502
        assert Grammar(prefixes)('A B').content == 'A B'
        assert Grammar(prefixes)('B A').content == 'B A'
503
504
505
506

    def test_allOf_completeness(self):
        """Test that an error is raised if not  all parsers of an AllOf-List
        match."""
eckhart's avatar
eckhart committed
507
        prefixes = Interleave(TKN("A"), TKN("B"))
508
509
510
511
512
        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."""
eckhart's avatar
eckhart committed
513
        prefixes = Interleave(TKN("A"), TKN("B"), TKN("A"))
514
515
516
        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'
517
518
519
520
        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."""
eckhart's avatar
eckhart committed
521
        prefixes = Interleave(TKN("A"), TKN("B"))
522
523
        assert Grammar(prefixes)('A B').content == 'A B'
        assert Grammar(prefixes)('B A').content == 'B A'
eckhart's avatar
eckhart committed
524
525
526
        st = Grammar(prefixes)('B')
        assert st.error_flag
        prefixes = Interleave(TKN("B"), TKN("A"), repetitions=((0, 1), (0, 1)))
527
        assert Grammar(prefixes)('A B').content == 'A B'
eckhart's avatar
eckhart committed
528
529
530
        st = Grammar(prefixes)('B')
        assert not st.error_flag
        assert st.content == 'B'
531
532
533
534

    def test_someOf_redundance(self):
        """Test that one and the same parser may be listed several times
        and must be matched several times accordingly."""
eckhart's avatar
eckhart committed
535
        prefixes = Interleave(TKN("A"), TKN("B"), TKN("A"))
536
537
538
        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'
539
540
541
        assert Grammar(prefixes)('A B B').error_flag


542
543
544
545
546
class TestInterleave:
    def test_interleave_most_simple(self):
        letterset = Interleave(Token("A"), Token("B"), Token("C"))
        gr = Grammar(letterset)
        st = gr('ABC')
547
        assert not st.errors, str(st.errors)
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
        assert st.content == "ABC"
        st = gr('BCA')
        assert not st.errors
        assert st.content == "BCA"
        st = gr('BCBA')
        assert st.errors
        st = gr('AB')
        assert st.errors

    def test_interleave(self):
        letterset = Interleave(Token("A"), Token("B"), Token("C"),
                               repetitions=[(1, 1000), (0, 1), (1, 1)])
        gr = Grammar(letterset)
        st = gr('AABC')
        assert not st.errors
563
564
565
566
567
568
569
570
571
572
        st = gr('BACAAA')
        assert not st.errors
        st = gr('ABCC')
        assert st.errors
        st = gr('AAACAAA')
        assert not st.errors
        st = gr('AAABAAA')
        assert st.errors


573
574
575
576
class TestErrorRecovery:
    def test_series_skip(self):
        lang = """
        document = series | /.*/
577
        @series_skip = /(?=[A-Z])/
578
579
580
581
582
        series = "A" "B" §"C" "D"
        """
        parser = grammar_provider(lang)()
        st = parser('AB_D')
        assert len(st.errors) == 1  # no additional "stopped before end"-error!
583
        resume_notices_on(parser)
584
585
        st = parser('AB_D')
        assert len(st.errors) == 2 and any(err.code == Error.RESUME_NOTICE for err in st.errors)
eckhart's avatar
eckhart committed
586
        assert 'Skipping' in str(st.errors_sorted[1])
587

588
    def test_Interleave_skip(self):
589
590
591
        lang = """
        document = allof | /.*/
        @allof_skip = /[A-Z]/
592
        allof = "A" ° §"B" ° "C" ° "D"
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
        """
        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"
609
610
        st = parser('A_CB')
        assert st['allof'].content == "A_CB"
611
        st = parser('BC_A')
612
        assert 'allof' not in st
613
614


615
class TestPopRetrieve:
616
    mini_language = r"""
617
618
        document       = { text | codeblock }
        codeblock      = delimiter { text | (!:delimiter delimiter_sign) } ::delimiter
619
        delimiter      = delimiter_sign  # never use delimiter between capture and pop except for retrival!
620
        delimiter_sign = /`+/
eckhart's avatar
eckhart committed
621
        text           = /[^`]+/
622
        """
623
    mini_lang2 = r"""
624
        @braces_filter=matching_bracket
625
626
627
628
629
630
631
        document       = { text | codeblock }
        codeblock      = braces { text | opening_braces | (!:braces closing_braces) } ::braces
        braces         = opening_braces
        opening_braces = /\{+/
        closing_braces = /\}+/
        text           = /[^{}]+/
        """
632
    mini_lang3 = r"""
633
634
635
636
        document       = { text | env }
        env            = (specialtag | opentag) text [closespecial | closetag]
        opentag        = "<" name ">"
        specialtag     = "<" /ABC/ !name ">"
eckhart's avatar
eckhart committed
637
        closetag       = close_slash | close_star
638
639
640
641
642
643
        close_slash    = "<" ::name "/>"
        close_star     = "<" ::name "*>"
        closespecial   = "<" /ABC/~ ">"
        name           = /\w+/~
        text           = /[^<>]+/
        """
644
645
646
647
648
649
650
651
    mini_lang4 = r"""
        document       = { text | env }
        env            = opentag document closetag
        opentag        = "<" name ">"
        closetag       = "</" :?name ">"
        name           = /\w+/~
        text           = /[^<>]+/        
    """
652
653

    def setup(self):
654
655
656
        self.minilang_parser = grammar_provider(self.mini_language)()
        self.minilang_parser2 = grammar_provider(self.mini_lang2)()
        self.minilang_parser3 = grammar_provider(self.mini_lang3)()
657
        self.minilang_parser4 = grammar_provider(self.mini_lang4)()
658
659

    @staticmethod
660
    def has_tag_name(node, name):
661
        return node.tag_name == name # and not isinstance(node.parser, Retrieve)
662

663
664
    def test_capture_assertions(self):
        try:
665
666
667
            _ = Grammar(Capture(Drop(Whitespace(r'\s*'))))
            assert False, "GrammarError expected!"
        except GrammarError as ge:
668
669
            pass
        try:
670
            _ = Grammar(Capture(Series(Token(' '), Drop(Whitespace(r'\s*')))))
671
            assert False, "ValueError expected!"
672
        except GrammarError:
673
            pass
674
675
676
        cp = Capture(RegExp(r'\w+'))
        cp.pname = "capture"
        _ = Grammar(cp)
677

678
679
680
681
    def test_compile_mini_language(self):
        assert self.minilang_parser
        assert self.minilang_parser2
        assert self.minilang_parser3
682
        assert self.minilang_parser4
683
684
685
686

    def test_stackhandling(self):
        ambigous_opening = "<ABCnormal> normal tag <ABCnormal*>"
        syntax_tree = self.minilang_parser3(ambigous_opening)
687
        assert not syntax_tree.error_flag, str(syntax_tree.errors_sorted)
688
689
690

        ambigous_opening = "<ABCnormal> normal tag <ABCnormal/>"
        syntax_tree = self.minilang_parser3(ambigous_opening)
691
        assert not syntax_tree.error_flag, str(syntax_tree.errors_sorted)
692
693
694

        forgot_closing_tag = "<em> where is the closing tag?"
        syntax_tree = self.minilang_parser3(forgot_closing_tag)
695
        assert syntax_tree.error_flag, str(syntax_tree.errors_sorted)
696
697
698

        proper = "<em> has closing tag <em/>"
        syntax_tree = self.minilang_parser3(proper)
699
        assert not syntax_tree.error_flag, str(syntax_tree.errors_sorted)
700
701
702

        proper = "<em> has closing tag <em*>"
        syntax_tree = self.minilang_parser3(proper)
703
        assert not syntax_tree.error_flag, str(syntax_tree.errors_sorted)
704

705
706
707
    def test_optional_match(self):
        test1 = '<info>Hey, you</info>'
        st = self.minilang_parser4(test1)
708
        assert not st.error_flag, str(st.errors_sorted)
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
        test12 = '<info>Hey, <emph>you</emph></info>'
        st = self.minilang_parser4(test1)
        assert not st.error_flag
        test2 = '<info>Hey, you</>'
        st = self.minilang_parser4(test2)
        assert not st.error_flag
        test3 = '<info>Hey, <emph>you</></>'
        st = self.minilang_parser4(test3)
        assert not st.error_flag
        test4 = '<info>Hey, <emph>you</></info>'
        st = self.minilang_parser4(test4)
        assert not st.error_flag

    def test_rollback_behaviour_of_optional_match(self):
        test1 = '<info>Hey, you</info*>'
        st = self.minilang_parser4(test1)
        assert not self.minilang_parser4.variables__['name']
        assert st.error_flag
        test2 = '<info>Hey, you</*>'
        st = self.minilang_parser4(test2)
        assert not self.minilang_parser4.variables__['name']
        assert st.error_flag

732
    def test_cache_neutrality(self):
733
734
        """Test that packrat-caching does not interfere with the variable-
        changing parsers: Capture and Retrieve."""
735
        lang = r"""
736
737
738
739
740
741
742
743
744
            text = opening closing
            opening = (unmarked_package | marked_package)
            closing = ::variable
            unmarked_package = package "."
            marked_package = package "*" "."
            package = "(" variable ")"
            variable = /\w+/~
            """
        case = "(secret)*. secret"
745
        gr = grammar_provider(lang)()
746
        st = gr(case)
747
        assert not st.error_flag, str(st.errors_sorted)
748
749
750
751

    def test_single_line(self):
        teststr = "Anfang ```code block `` <- keine Ende-Zeichen ! ``` Ende"
        syntax_tree = self.minilang_parser(teststr)
752
753
        assert not syntax_tree.errors_sorted, \
            ''.join(str(error) for error in syntax_tree.errors_sorted)
754
755
756
        matchf = partial(self.has_tag_name, name="delimiter")
        delim = str(next(syntax_tree.select_if(matchf)))
        pop = str(next(syntax_tree.select_if(matchf, reverse=True)))
757
758
        assert delim == pop
        if is_logging():
759
            log_ST(syntax_tree, "test_PopRetrieve_single_line.cst")
760
761
762
763
764
765
766
767
768
769
770

    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)
771
        assert not syntax_tree.errors_sorted
772
773
774
        matchf = partial(self.has_tag_name, name="delimiter")
        delim = str(next(syntax_tree.select_if(matchf)))
        pop = str(next(syntax_tree.select_if(matchf, reverse=True)))
775
776
        assert delim == pop
        if is_logging():
777
            log_ST(syntax_tree, "test_PopRetrieve_multi_line.cst")
778
779
780
781

    def test_single_line_complement(self):
        teststr = "Anfang {{{code block }} <- keine Ende-Zeichen ! }}} Ende"
        syntax_tree = self.minilang_parser2(teststr)
782
        assert not syntax_tree.errors_sorted
783
784
785
786
787
        matchf = partial(self.has_tag_name, name="braces")
        delim = str(next(syntax_tree.select_if(matchf)))
        pop = str(next(syntax_tree.select_if(matchf, reverse=True)))
        assert len(delim) == len(pop)
        assert delim != pop
788
        if is_logging():
789
            log_ST(syntax_tree, "test_PopRetrieve_single_line.cst")
790
791
792
793
794
795
796
797
798
799
800

    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)
801
        assert not syntax_tree.errors_sorted
802
803
804
        matchf = partial(self.has_tag_name, name="braces")
        delim = str(next(syntax_tree.select_if(matchf)))
        pop = str(next(syntax_tree.select_if(matchf, reverse=True)))
805
806
        assert len(delim) == len(pop) and delim != pop
        if is_logging():
807
            log_ST(syntax_tree, "test_PopRetrieve_multi_line.cst")
808

809
    def test_autoretrieve(self):
810
811
        lang = r"""
            document   = { definition } § EOF
812
            definition = symbol :defsign value
813
            symbol     = /\w+/~                      
814
815
            defsign    = "=" | ":="
            value      = /\d+/~
816
            EOF        = !/./ [:?defsign]   # eat up captured defsigns
817
        """
818
819
        # code, _, _ = compile_ebnf(lang)
        # print(code)
820
        parser = grammar_provider(lang)()
821
822
823
        st = parser("X := 1")
        assert not st.error_flag
        st1 = st
824
825
        st = parser("")
        assert not st.error_flag
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849

        lines = [line for line in lang.split('\n') if line.strip()]
        eof_line = lines.pop()
        lines.insert(1, eof_line)
        lang = '\n'.join(lines)
        parser = grammar_provider(lang)()
        st = parser("X := 1")
        assert not st.errors
        assert st.equals(st1)

        del lines[1]
        lines.insert(2, eof_line)
        lang = '\n'.join(lines)
        parser = grammar_provider(lang)()
        st = parser("X := 1")
        assert not st.errors
        assert st.equals(st1)

        # and, finally...
        lang_variant = r"""
            document   = { definition } § EOF
            symbol     = /\w+/~                      
            defsign    = "=" | ":="
            value      = /\d+/~
850
            EOF        = !/./ :?defsign   # eat up captured defsign, only if it has been retrieved
851
852
853
854
855
856
            definition = symbol :defsign value
        """
        parser = grammar_provider(lang_variant)()
        st = parser("X := 1")
        assert not st.errors
        assert st.equals(st1)
857
858
        st = parser('')
        assert "EOF expected" in str(st.errors)
di68kap's avatar
di68kap committed
859

860

861
class TestWhitespaceHandling:
Eckhart Arnold's avatar
Eckhart Arnold committed
862
    minilang = """
863
864
865
866
867
868
869
870
871
        doc = A B
        A = "A"
        B = "B"
        Rdoc = ar br
        ar = /A/
        br = /B/
        """

    def setup(self):
872
        self.gr = grammar_provider(self.minilang)()
873
874
875
876
877
878
879
880
881
882
883
884

    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
885
886
887
888
889


class TestErrorReporting:
    grammar = """
        root      = series alpha | anything
eckhart's avatar
eckhart committed
890
        series    = subseries &alpha
di68kap's avatar
di68kap committed
891
892
893
894
895
896
897
898
899
900
901
902
903
904
        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)
905
        assert not cst.error_flag, str(cst.errors_sorted)
di68kap's avatar
di68kap committed
906
907
908
909
        cst = self.parser(testcode2)
        assert not cst.error_flag
        cst = self.parser(testcode3)
        assert cst.error_flag
910
911


di68kap's avatar
di68kap committed
912
913
class TestBorderlineCases:
    def test_not_matching(self):
914
        minilang = """parser = /X/\n"""
di68kap's avatar
di68kap committed
915
916
917
918
        gr = grammar_provider(minilang)()
        cst = gr('X', 'parser')
        assert not cst.error_flag
        cst = gr(' ', 'parser')
919
        assert cst.error_flag and cst.errors_sorted[0].code == Error.PARSER_DID_NOT_MATCH
di68kap's avatar
di68kap committed
920
        cst = gr('', 'parser')
921
        assert cst.error_flag and cst.errors_sorted[0].code == Error.PARSER_DID_NOT_MATCH
di68kap's avatar
di68kap committed
922
923
924
925
926
927
928

    def test_matching(self):
        minilang = """parser = /.?/"""
        gr = grammar_provider(minilang)()
        cst = gr(' ', 'parser')
        assert not cst.error_flag
        cst = gr('  ', 'parser')
929
        assert cst.error_flag and cst.errors_sorted[0].code == Error.PARSER_STOPPED_BEFORE_END
di68kap's avatar
di68kap committed
930
931
932
933
        cst = gr('', 'parser')
        assert not cst.error_flag


eckhart's avatar
eckhart committed
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
EBNF_with_Errors = r"""# Test code with errors. All places marked by a "$" should yield and error

@ comment    = /#.*(?:\n|$)/
@ whitespace = /\s*/
@ literalws  = right
@ anonymous  = pure_elem, EOF
@ drop       = whitespace, EOF


# re-entry-rules for resuming after parsing-error
@ definition_resume = /\n\s*(?=@|\w+\w*\s*=)/
@ directive_resume  = /\n\s*(?=@|\w+\w*\s*=)/

# specialized error messages for certain cases

@ definition_error  = /,/, 'Delimiter "," not expected in definition!\nEither this was meant to '
                           'be a directive and the directive symbol @ is missing\nor the error is '
                           'due to inconsistent use of the comma as a delimiter\nfor the elements '
                           'of a sequence.'

#: top-level

syntax     = [~//] { definition | directive } §EOF
definition = symbol §:DEF~ expression :ENDL~
directive  = "@" §symbol "="
             (regexp | literals | symbol)
             { "," (regexp | literals | symbol) }

#: components

expression = sequence { :OR~ sequence }
sequence   = ["§"] ( interleave | lookaround )
             { :AND~ ["§"] ( interleave | lookaround ) }
interleave = difference { "°" ["§"] difference }
lookaround = flowmarker § (oneormore | pure_elem)
difference = term ["-" § (oneormore $ pure_elem)]               # <- ERROR
term       = oneormore | repetition | option | pure_elem        # resuming expected her

#: elements

pure_elem  = element § !/[?*+]/
element    = [retrieveop] symbol !DEF
           | literal
           | plaintext
           | regexp
           | whitespace
           | group$                                             # <- ERROR

#: flow-operators

flowmarker = "!"  | "&"                                         # resuming expected her
           | "<-!" | "<-&"
retr$ieveop = "::" | ":?" | ":"

#: groups

group      = "(" §expression ")"
oneormore  = "{" expression "}+" | element "+"
repetition = "{" §expressi$on "}" | element "*"                 # <- ERROR
option     = "[" §expression "]" | element "?"                  # resuming expected here

#: leaf-elements

symbol     = /(?!\d)\w+/~
$literals   = { literal }+                                      # <- ERROR
literal    = /"(?:(?<!\\)\\"|[^"])*?"/~                         # resuming expected her
           | /'(?:(?<!\\)\\'|[^'])*?'/~