test_parse.py 62.8 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, Tuple
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, start_logging
Eckhart Arnold's avatar
Eckhart Arnold committed
34
from DHParser.error import Error, is_error, add_source_locations, MANDATORY_CONTINUATION, \
eckhart's avatar
eckhart committed
35
    MALFORMED_ERROR_STRING, MANDATORY_CONTINUATION_AT_EOF, RESUME_NOTICE, PARSER_STOPPED_BEFORE_END, \
36
    PARSER_NEVER_TOUCHES_DOCUMENT, CAPTURE_DROPPED_CONTENT_WARNING
37
from DHParser.parse import ParserError, Parser, Grammar, Forward, TKN, ZeroOrMore, RE, \
eckhart's avatar
eckhart committed
38
    RegExp, Lookbehind, NegativeLookahead, OneOrMore, Series, Alternative, \
39
    Interleave, CombinedParser, Text, EMPTY_NODE, Capture, Drop, Whitespace, \
40
    GrammarError, Counted, Always, INFINITE, longest_match
41
from DHParser import compile_source
42
from DHParser.ebnf import get_ebnf_grammar, get_ebnf_transformer, get_ebnf_compiler, \
43
    parse_ebnf, DHPARSER_IMPORTS, compile_ebnf
Eckhart Arnold's avatar
Eckhart Arnold committed
44
45
from DHParser.dsl import grammar_provider, create_parser
from DHParser.preprocess import gen_neutral_srcmap_func
46
from DHParser.syntaxtree import Node, parse_sxpr
47
from DHParser.stringview import StringView
48
from DHParser.trace import set_tracer, trace_history, resume_notices_on
49
50


51
52
53
54
55
56

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

Eckhart Arnold's avatar
Eckhart Arnold committed
57
    def test_non_empty_derivation(self):
58
59
        pass

60

61
62
class TestParserError:
    def test_parser_error_str(self):
63
64
        pe = ParserError(Parser(), Node('TAG', 'test').with_pos(0),
                         StringView('Beispiel'), None, first_throw=True)
65
66
        assert str(pe).find('Beispiel') >= 0 and str(pe).find('TAG') >= 0

67
    def test_false_lookahead_only_message(self):
68
        """PARSER_LOOKAHEAD_*_ONLY errors must not be reported if there
69
70
71
72
73
74
        no lookahead parser in the history!"""
        lang = """
        word = letters { letters | `-` letters }
        letters = /[A-Za-z]+/
        """
        gr = grammar_provider(lang)()
75
76
        set_tracer(gr, trace_history)
        st = gr('hard-time')
77
        assert not st.errors
78
        st = gr('hard-')
79
80
        assert st.errors and not any(e.code == 1045 for e in st.errors)

eckhart's avatar
eckhart committed
81

eckhart's avatar
eckhart committed
82
83
84
85
86
87
88
89
90
class TestParserClass:
    def test_apply(self):
        minilang ="""
            expr = expr ("+"|"-") term | term
            term = term ("*"|"/") factor | factor
            factor = /[0-9]+/~
            """
        gr = grammar_provider(minilang)()
        l = []
91
92
        def visitor(context: List[Parser]):
            p = context[-1]
di68kap's avatar
di68kap committed
93
            l.append(p.pname + p.ptype)
eckhart's avatar
eckhart committed
94
95
96
97
98
99
100
101
102
103
        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

104
105
    def test_symbol(self):
        class MyGrammar(Grammar):
106
            wrong = Text('wrong')
107
108
109
110
            word = OneOrMore(wrong) + Whitespace(r'\s*') + OneOrMore(RegExp(r'\w+'))
            root__ = word
        gr = MyGrammar()
        regex = gr['word'].parsers[-1].parser
111
        result = gr.associated_symbol__(regex).symbol
112
        assert result == 'word', result
113

eckhart's avatar
eckhart committed
114

115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
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
178
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
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
class TestInfiLoopsAndRecursion:
    def setup(self):
        pass
        # set_config_value('history_tracking', True)
        # set_config_value('resume_notices', True)
        # start_logging('LOGS')

    def test_very_simple(self):
        minilang = """
            term = term (`*`|`/`) factor | factor
            factor = /[0-9]+/
            """
        grammar_factory = grammar_provider(minilang)
        parser = grammar_factory()
        snippet = "5*4*3*2"
        # set_tracer(parser, trace_history)
        st = parser(snippet)
        if is_logging():
            log_ST(st, 'test_LeftRecursion_very_simple.cst')
            log_parsing_history(parser, 'test_LeftRecursion_very_simple')
        assert not is_error(st.error_flag), str(st.errors)
        st = parser("1*2*3*4*5*6*7*8*9")
        # if is_logging():
        #     log_ST(st, 'test_LeftRecursion_very_simple_2.cst')
        #     log_parsing_history(parser, 'test_LeftRecursion_very_simple_2')
        assert not is_error(st.error_flag)

    def test_direct_left_recursion1(self):
        minilang = """@literalws = right
            expr = expr ("+"|"-") term | term
            term = term ("*"|"/") factor | factor
            factor = /[0-9]+/~
            """
        snippet = "9 + 8 + 7 + 6 + 5 + 3 * 4"
        parser = grammar_provider(minilang)()
        # print(raw_compileEBNF(minilang).result)
        assert parser
        syntax_tree = parser(snippet)
        if is_logging():
            log_ST(syntax_tree, "test_LeftRecursion_direct1.cst")
            log_parsing_history(parser, "test_LeftRecursion_direct1")
        assert not is_error(syntax_tree.error_flag), str(syntax_tree.errors_sorted)
        assert snippet == syntax_tree.content, str(syntax_tree)

    def test_direct_left_recursion2(self):
        minilang = """@literalws = right
            expr = ex
            ex   = expr ("+"|"-") term | term
            term = tr
            tr   = term ("*"|"/") factor | factor
            factor = /[0-9]+/~
            """
        snippet = "9 + 8 + 7 + 6 + 5 + 3 * 4"
        parser = grammar_provider(minilang)()
        assert parser
        syntax_tree = parser(snippet)
        assert not is_error(syntax_tree.error_flag), syntax_tree.errors_sorted
        assert snippet == syntax_tree.content
        if is_logging():
            log_ST(syntax_tree, "test_LeftRecursion_direct2.cst")
            log_parsing_history(parser, "test_LeftRecursion_direct2")

    def test_indirect_left_recursion1(self):
        minilang = """@literalws = right
            Expr    = //~ (Product | Sum | Value)
            Product = Expr { ('*' | '/') Expr }+
            Sum     = Expr { ('+' | '-') Expr }+
            Value   = /[0-9.]+/~ | '(' §Expr ')'
            """
        # print(raw_compileEBNF(minilang).result)
        parser = grammar_provider(minilang)()
        snippet = "8 * 4"
        syntax_tree = parser(snippet)
        assert not is_error(syntax_tree.error_flag), syntax_tree.errors_sorted
        snippet = "7 + 8 * 4"
        syntax_tree = parser(snippet)
        if is_logging():
            log_ST(syntax_tree, "test_LeftRecursion_indirect1.cst")
            log_parsing_history(parser, "test_LeftRecursion_indirect1")
        assert not is_error(syntax_tree.error_flag), syntax_tree.errors_sorted
        snippet = "9 + 8 * (4 + 3)"
        syntax_tree = parser(snippet)
        assert not is_error(syntax_tree.error_flag), syntax_tree.errors_sorted
        assert snippet == syntax_tree.content
        snippet = "9 + 8 * (4 - 3 / (5 - 1))"
        syntax_tree = parser(snippet)
        assert not is_error(syntax_tree.error_flag), syntax_tree.errors_sorted
        assert snippet == syntax_tree.content
        if is_logging():
            log_ST(syntax_tree, "test_LeftRecursion_indirect1.cst")
            log_parsing_history(parser, "test_LeftRecursion_indirect1")

    # BEWARE: EXPERIMENTAL TEST can be long running
    def test_indirect_left_recursion2(self):
        arithmetic_syntax = r"""@literalws = right
            expression     = addition | subtraction  # | term
            addition       = (expression | term) "+" (expression | term)
            subtraction    = (expression | term) "-" (expression | term)
            term           = multiplication | division  # | factor
            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)()
        assert arithmetic
        syntax_tree = arithmetic("(a + b) * (a - b)")
        assert syntax_tree.errors
        if is_logging():
            log_ST(syntax_tree, "test_LeftRecursion_indirect2.cst")
            log_parsing_history(arithmetic, "test_LeftRecursion_indirect2")

    def test_indirect_left_recursion3(self):
        arithmetic_syntax = r"""@literalws = right
            expression     = addition | subtraction | term
            addition       = (expression | term) "+" (expression | term)
            subtraction    = (expression | term) "-" (expression | term)
            term           = multiplication | division | factor
            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)()
        assert arithmetic
        syntax_tree = arithmetic("(a + b) * (a - b)")
        assert not syntax_tree.errors
        if is_logging():
            log_ST(syntax_tree, "test_LeftRecursion_indirect3.cst")
            log_parsing_history(arithmetic, "test_LeftRecursion_indirect3")


    def test_break_inifnite_loop_ZeroOrMore(self):
        forever = ZeroOrMore(RegExp(''))
        result = Grammar(forever)('')  # infinite loops will automatically be broken
256
        assert repr(result) == "Node('root', '')", repr(result)
257
258
259
260

    def test_break_inifnite_loop_OneOrMore(self):
        forever = OneOrMore(RegExp(''))
        result = Grammar(forever)('')  # infinite loops will automatically be broken
261
        assert repr(result) == "Node('root', '')", repr(result)
262
263
264
265

    def test_break_infinite_loop_Counted(self):
        forever = Counted(Always(), (0, INFINITE))
        result = Grammar(forever)('')  # if this takes very long, something is wrong
266
        assert repr(result) == "Node('root', '')", repr(result)
267
268
        forever = Counted(Always(), (5, INFINITE))
        result = Grammar(forever)('')  # if this takes very long, something is wrong
269
        assert repr(result) == "Node('root', '')", repr(result)
270
271
        forever = Counted(Always(), (INFINITE, INFINITE))
        result = Grammar(forever)('')  # if this takes very long, something is wrong
272
        assert repr(result) == "Node('root', '')", repr(result)
273
274
        forever = Counted(Always(), (1000, INFINITE - 1))
        result = Grammar(forever)('')  # if this takes very long, something is wrong
275
        assert repr(result) == "Node('root', '')", repr(result)
276
277

    def test_break_infinite_loop_Interleave(self):
278
        forever = Interleave(Always(), repetitions=[(0, INFINITE)])
279
        result = Grammar(forever)('')  # if this takes very long, something is wrong
280
        assert repr(result) == "Node('root', '')", repr(result)
281
        forever = Interleave(Always(), Always(),
282
                             repetitions=[(5, INFINITE), (INFINITE, INFINITE)])
283
        result = Grammar(forever)('')  # if this takes very long, something is wrong
284
        assert repr(result) == "Node('root', '')", repr(result)
285
        forever = Interleave(Always(), repetitions=[(1000, INFINITE - 1)])
286
        result = Grammar(forever)('')  # if this takes very long, something is wrong
287
        assert repr(result) == "Node('root', '')", repr(result)
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306

    # def test_infinite_loops(self):
    #     minilang = """forever = { // } \n"""
    #     try:
    #         parser_class = grammar_provider(minilang)
    #     except CompilationError as error:
    #         assert all(e.code == 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 == INFINITE_LOOP
    #     set_config_value('static_analysis', 'none')
    #     parser = provider()
    #     snippet = " "
    #     syntax_tree = parser(snippet)
    #     assert any(e.code == INFINITE_LOOP for e in syntax_tree.errors)
307
    #     res = parser.static_analysis__()
308
309
310
    #     assert res and res[0][2].code == INFINITE_LOOP
    #     minilang = """not_forever = { / / } \n"""
    #     parser = grammar_provider(minilang)()
311
    #     res = parser.static_analysis__()
312
313
    #     assert not res
    #     set_config_value('static_analysis', save)
314

eckhart's avatar
eckhart committed
315

316
317
318
319
320
321
# class TestStaticAnalysis:
#     def test_alternative(self):
#         lang = 'doc = "A" | "AB"'
#         parser = create_parser(lang)


Eckhart Arnold's avatar
Eckhart Arnold committed
322
class TestFlowControl:
323
324
325
326
327
328
    t1 = """
         All work and no play
         makes Jack a dull boy
         END
         """
    t2 = "All word and not play makes Jack a dull boy END\n"
329

Eckhart Arnold's avatar
Eckhart Arnold committed
330
    def test_lookbehind(self):
331
332
        ws = RegExp(r'\s*');  ws.pname = "ws"
        end = RegExp("END");  end.pname = "end"
333
        doc_end = Lookbehind(RegExp('\\s*?\\n')) + end
334
        word = RegExp(r'\w+');  word.pname = "word"
Eckhart Arnold's avatar
Eckhart Arnold committed
335
336
337
        sequence = OneOrMore(NegativeLookahead(end) + word + ws)
        document = ws + sequence + doc_end + ws
        parser = Grammar(document)
338
339
340
341
342
        cst = parser(self.t1)
        assert not cst.error_flag, cst.as_sxpr()
        cst = parser(self.t2)
        assert cst.error_flag, cst.as_sxpr()

343
344
345
346
347
348
349
350
        cst = parser(self.t2, parser['ws'], complete_match=False)
        assert cst.did_match() and len(cst) == 0 and not cst.errors
        cst = parser(self.t2, parser['word'], complete_match=False)
        assert cst.did_match() and cst.content == "All" and not cst.errors
        cst = parser(self.t2, parser['end'], complete_match=False)
        assert not cst.did_match()


351
352
    def test_lookbehind_indirect(self):
        class LookbehindTestGrammar(Grammar):
eckhart's avatar
eckhart committed
353
            parser_initialization__ = ["upon instantiation"]
354
            ws = RegExp(r'\s*')
355
            end = RegExp('END')
356
            SUCC_LB = RegExp('\\s*?\\n')
357
            doc_end = Series(Lookbehind(SUCC_LB), end)
358
            word = RegExp(r'\w+')
359
360
361
362
363
364
            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
365
        assert not cst.error_flag, cst.as_sxpr()
366
        cst = parser(self.t2)
Eckhart Arnold's avatar
Eckhart Arnold committed
367
368
369
        assert cst.error_flag, cst.as_sxpr()


370
371
372
373
374
375
376
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
377
        result, messages, _ = compile_source(mlregex, None, get_ebnf_grammar(),
Eckhart Arnold's avatar
Eckhart Arnold committed
378
                        get_ebnf_transformer(), get_ebnf_compiler('MultilineRegexTest'))
379
        assert result
380
        assert not messages, str(messages)
381
        parser = compile_python_object(DHPARSER_IMPORTS + result, r'\w+Grammar$')()
eckhart's avatar
eckhart committed
382
383
        node = parser('abc+def', parser.regex)
        assert not node.error_flag
384
        assert node.tag_name == "regex"
385
386
        assert str(node) == 'abc+def'

387
388
    def test_multilineRegex_wo_Comments(self):
        mlregex = r"""
eckhart's avatar
eckhart committed
389
390
        regex =  /\w+
                  [+]
391
392
                  \w* /
        """
Eckhart Arnold's avatar
Eckhart Arnold committed
393
        result, messages, _ = compile_source(mlregex, None, get_ebnf_grammar(),
394
395
396
                        get_ebnf_transformer(), get_ebnf_compiler('MultilineRegexTest'))
        assert result
        assert not messages, str(messages)
397
        parser = compile_python_object(DHPARSER_IMPORTS + result, r'\w+Grammar$')()
398
399
        node = parser('abc+def', parser.regex)
        assert not node.error_flag
400
        assert node.tag_name == "regex"
401
402
        assert str(node) == 'abc+def'

403
    def test_ignore_case(self):
404
405
406
407
        mlregex = r"""
        @ ignorecase = True
        regex = /alpha/
        """
Eckhart Arnold's avatar
Eckhart Arnold committed
408
        result, messages, _ = compile_source(mlregex, None, get_ebnf_grammar(),
409
410
411
                        get_ebnf_transformer(), get_ebnf_compiler('MultilineRegexTest'))
        assert result
        assert not messages
412
        parser = compile_python_object(DHPARSER_IMPORTS + result, r'\w+Grammar$')()
413
        node, rest = parser.regex(StringView('Alpha'))
414
415
        assert node
        assert rest == ''
416
        assert node.tag_name == "regex"
417
418
419
420
421
422
        assert str(node) == 'Alpha'

        mlregex = r"""
        @ ignorecase = False
        regex = /alpha/
        """
Eckhart Arnold's avatar
Eckhart Arnold committed
423
        result, messages, _ = compile_source(mlregex, None, get_ebnf_grammar(),
424
425
426
                        get_ebnf_transformer(), get_ebnf_compiler('MultilineRegexTest'))
        assert result
        assert not messages
427
        parser = compile_python_object(DHPARSER_IMPORTS + result, r'\w+Grammar$')()
428
429
        node, rest = parser.regex(StringView('Alpha'))
        assert node is None
430

431
    def test_token(self):
432
        tokenlang = r"""@literalws = right
433
434
435
436
437
438
439
440
441
442
            @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
443
        result, messages, _ = compile_source(
eckhart's avatar
eckhart committed
444
445
            tokenlang, None, get_ebnf_grammar(), get_ebnf_transformer(),
            get_ebnf_compiler("TokenTest"))
446
        assert result
eckhart's avatar
eckhart committed
447
        assert not messages, str(messages)
448
        parser = compile_python_object(DHPARSER_IMPORTS + result, r'\w+Grammar$')()
Eckhart Arnold's avatar
Eckhart Arnold committed
449
        result = parser(testdoc)
450
        # log_parsing_history(parser, "test.log")
451
        assert not result.error_flag, str(result.errors_sorted)
452

453

454
class TestGrammar:
455
456
457
458
459
460
461
462
    grammar = r"""@whitespace = horizontal
    haupt        = textzeile LEERZEILE
    textzeile    = { WORT }+
    WORT         = /[^ \t]+/~
    LEERZEILE    = /\n[ \t]*(?=\n)/~
    """
    pyparser, messages, _ = compile_source(grammar, None, get_ebnf_grammar(),
                                           get_ebnf_transformer(), get_ebnf_compiler("PosTest"))
eckhart's avatar
eckhart committed
463
    assert pyparser, str(messages)
eckhart's avatar
eckhart committed
464
    assert not messages, str(messages)
465
466
467
468

    def test_pos_values_initialized(self):
        # checks whether pos values in the parsing result and in the
        # history record have been initialized
469
470
        grammar = compile_python_object(DHPARSER_IMPORTS + self.pyparser, r'\w+Grammar$')()
        grammar("no_file_name*")
471
        for record in grammar.history__:
472
473
            assert not record.node or record.node.pos >= 0

474
    def test_select_parsing(self):
475
        grammar = compile_python_object(DHPARSER_IMPORTS + self.pyparser, r'\w+Grammar$')()
476
477
478
479
        grammar("wort", "WORT")
        grammar("eine Zeile", "textzeile")
        grammar("kein Haupt", "haupt")
        grammar("so ist es richtig", "haupt")
480

di68kap's avatar
di68kap committed
481
482
    def test_grammar_subclassing(self):
        class Arithmetic(Grammar):
483
            r'''
di68kap's avatar
di68kap committed
484
485
486
487
488
489
            expression =  term  { ("+" | "-") term }
            term       =  factor  { ("*" | "/") factor }
            factor     =  INTEGER | "("  expression  ")"
            INTEGER    =  /\d+/~
            '''
            expression = Forward()
490
491
492
493
            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
494
495
496
497
498
499
            root__ = expression

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

500
501
502
503
504
505
506
    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')
507
        assert st.errors[0].code == PARSER_STOPPED_BEFORE_END
508
509
510
        st = gr('eins zwei', complete_match=False)
        assert not st.errors

511
    def test_synonym(self):
di68kap's avatar
di68kap committed
512
        lang = r"""
513
514
515
516
517
518
519
520
521
            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')
522
        gr = grammar_provider("@drop = whitespace, strings" + lang)()
523
524
        st = gr('eins 1 zwei2drei 3')
        st = gr('-3')
525
        assert str(gr['S']) == "S = ~", str(gr['S'])
526

527

528
529
530
531
532
533
534
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
535
        st = parser("ABCD")
536
        assert not st.error_flag
Eckhart Arnold's avatar
Eckhart Arnold committed
537
        st = parser("A_CD")
538
        assert not st.error_flag
Eckhart Arnold's avatar
Eckhart Arnold committed
539
        st = parser("AB_D")
540
541
542
543
544
545
546
547
548
549
550
        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)()
551
552
553
        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
554
        assert st.errors_sorted[0].code == MANDATORY_CONTINUATION
Eckhart Arnold's avatar
bugfix    
Eckhart Arnold committed
555
        assert str(st.errors_sorted[0]).find("series") >= 0
556
        # transitivity of mandatory-operator
557
        st = parser("ABC_");  assert st.error_flag
558
        assert st.errors_sorted[0].code == MANDATORY_CONTINUATION
559
560

    def test_series_composition(self):
561
        TA, TB, TC, TD, TE = (TKN(b) for b in "ABCDE")
562
563
564
565
566
        s1 = Series(TA, TB, TC, mandatory=2)
        s2 = Series(TD, TE)

        combined = Alternative(s1 + s2, RegExp('.*'))
        parser = Grammar(combined)
567
568
569
        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
570
        assert st.errors_sorted[0].code == MANDATORY_CONTINUATION
571
        st = parser("ABC_E");  assert st.error_flag
572
        assert st.errors_sorted[0].code == MANDATORY_CONTINUATION
573
574
575

        combined = Alternative(s2 + s1, RegExp('.*'))
        parser = Grammar(combined)
576
577
578
579
580
581
        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
582
        assert st.errors_sorted[0].code == MANDATORY_CONTINUATION
583

eckhart's avatar
eckhart committed
584
585
586
587
588
589
590
591
592
593
    # 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
594

595
596
    def test_ebnf_serialization(self):
        ebnf_grammar = get_ebnf_grammar()
eckhart's avatar
eckhart committed
597
        # TODO: Add test here
598
        ebnf = ebnf_grammar.as_ebnf__()
599
        # print(ebnf)
600
601


602
603
604
class TestAllOfSomeOf:
    def test_allOf_order(self):
        """Test that parsers of an AllOf-List can match in arbitrary order."""
eckhart's avatar
eckhart committed
605
        prefixes = Interleave(TKN("A"), TKN("B"))
606
607
        assert Grammar(prefixes)('A B').content == 'A B'
        assert Grammar(prefixes)('B A').content == 'B A'
608
609
610
611

    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
612
        prefixes = Interleave(TKN("A"), TKN("B"))
613
614
615
616
617
        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
618
        prefixes = Interleave(TKN("A"), TKN("B"), TKN("A"))
619
620
621
        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'
622
623
624
625
        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
626
        prefixes = Interleave(TKN("A"), TKN("B"))
627
628
        assert Grammar(prefixes)('A B').content == 'A B'
        assert Grammar(prefixes)('B A').content == 'B A'
eckhart's avatar
eckhart committed
629
630
631
        st = Grammar(prefixes)('B')
        assert st.error_flag
        prefixes = Interleave(TKN("B"), TKN("A"), repetitions=((0, 1), (0, 1)))
632
        assert Grammar(prefixes)('A B').content == 'A B'
eckhart's avatar
eckhart committed
633
634
635
        st = Grammar(prefixes)('B')
        assert not st.error_flag
        assert st.content == 'B'
636
637
638
639

    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
640
        prefixes = Interleave(TKN("A"), TKN("B"), TKN("A"))
641
642
643
        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'
644
645
646
        assert Grammar(prefixes)('A B B').error_flag


647
648
class TestInterleave:
    def test_interleave_most_simple(self):
649
        letterset = Interleave(Text("A"), Text("B"), Text("C"))
650
651
        gr = Grammar(letterset)
        st = gr('ABC')
652
        assert not st.errors, str(st.errors)
653
654
655
656
657
658
659
660
661
662
        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):
663
        letterset = Interleave(Text("A"), Text("B"), Text("C"),
664
665
666
667
                               repetitions=[(1, 1000), (0, 1), (1, 1)])
        gr = Grammar(letterset)
        st = gr('AABC')
        assert not st.errors
668
669
670
671
672
673
674
675
676
677
        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


678
679
680
681
class TestErrorRecovery:
    def test_series_skip(self):
        lang = """
        document = series | /.*/
682
        @series_skip = /(?=[A-Z])/
683
684
685
686
687
        series = "A" "B" §"C" "D"
        """
        parser = grammar_provider(lang)()
        st = parser('AB_D')
        assert len(st.errors) == 1  # no additional "stopped before end"-error!
688
        resume_notices_on(parser)
689
690
691
692
        st = parser('AB_D')
        assert len(st.errors) == 2 and any(err.code == RESUME_NOTICE for err in st.errors)
        assert 'Skipping' in str(st.errors_sorted[1])

693
    def test_series_skip2(self):
694
        grammar = r"""
695
696
        @whitespace = vertical
        @literalws = right
697
698
699
700
701
702
        @document_skip = /\s+|(?=$)/
        document = { sentence § sentence_continuation }+ EOF
            sentence_continuation = &(sentence | EOF)
        @sentence_skip = /\s+|(?=\.|$)/
        sentence = { word § word_continuation }+ "."
            word_continuation = &(word | `.` | EOF)
703
704
705
706
707
708
709
710
711
712
713
714
715
        word = /[A-Za-z]+/~
        EOF = !/./ 
        """
        data = "Time is out of joint. Oh cursed spite that I was ever born to set it right."
        parser = grammar_provider(grammar)()
        st = parser(data)
        assert not st.errors, str(st.errors)
        data2 = data.replace('cursed', 'cur?ed')
        st = parser(data2)
        assert len(st.errors) == 1
        zombie = st.pick('ZOMBIE__')
        assert zombie.content == '?ed ', zombie.content

716
717
718
719
720
721
722
723
724
725
726
727
    def test_irrelevance_of_error_definition_order(self):
        lang = """
        document = series | /.*/
        series = "A" "B" §"C" "D"
        @series_skip = /(?=[A-Z])/
        @series_error = '', "Unerwartetes Zeichen"
        """
        parser = grammar_provider(lang)()
        st = parser('AB_D')
        assert len(st.errors) == 1  # no additional "stopped before end"-error!
        assert st.errors[0].message == "Unerwartetes Zeichen"
        resume_notices_on(parser)
728
        st = parser('AB_D')
729
        assert len(st.errors) == 2 and any(err.code == RESUME_NOTICE for err in st.errors)
eckhart's avatar
eckhart committed
730
        assert 'Skipping' in str(st.errors_sorted[1])
731

732
    def test_Interleave_skip(self):
733
734
735
        lang = """
        document = allof | /.*/
        @allof_skip = /[A-Z]/
736
        allof = "A" ° §"B" ° "C" ° "D"
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
        """
        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"
753
754
        st = parser('A_CB')
        assert st['allof'].content == "A_CB"
755
        st = parser('BC_A')
756
        assert 'allof' not in st
757
758


759
class TestPopRetrieve:
760
    mini_language = r"""
761
762
        document       = { text | codeblock }
        codeblock      = delimiter { text | (!:delimiter delimiter_sign) } ::delimiter
763
        delimiter      = delimiter_sign  # never use delimiter between capture and pop except for retrieval!
764
        delimiter_sign = /`+/
eckhart's avatar
eckhart committed
765
        text           = /[^`]+/
766
        """
767
    mini_lang2 = r"""
768
        @braces_filter = matching_bracket()
769
770
771
772
773
774
775
        document       = { text | codeblock }
        codeblock      = braces { text | opening_braces | (!:braces closing_braces) } ::braces
        braces         = opening_braces
        opening_braces = /\{+/
        closing_braces = /\}+/
        text           = /[^{}]+/
        """
776
    mini_lang3 = r"""@literalws = right
777
        document       = { text | env }
eckhart's avatar
eckhart committed
778
        env            = (specialtag | opentag) text [ closespecial | closetag ]
779
780
        opentag        = "<" name ">"
        specialtag     = "<" /ABC/ !name ">"
eckhart's avatar
eckhart committed
781
        closetag       = close_slash | close_star
782
783
784
785
786
787
        close_slash    = "<" ::name "/>"
        close_star     = "<" ::name "*>"
        closespecial   = "<" /ABC/~ ">"
        name           = /\w+/~
        text           = /[^<>]+/
        """
788
    mini_lang4 = r"""@literalws = right
789
790
791
792
793
794
795
        document       = { text | env }
        env            = opentag document closetag
        opentag        = "<" name ">"
        closetag       = "</" :?name ">"
        name           = /\w+/~
        text           = /[^<>]+/        
    """
796
797

    def setup(self):
798
799
800
        self.minilang_parser = grammar_provider(self.mini_language)()
        self.minilang_parser2 = grammar_provider(self.mini_lang2)()
        self.minilang_parser3 = grammar_provider(self.mini_lang3)()
801
        self.minilang_parser4 = grammar_provider(self.mini_lang4)()
802
803

    @staticmethod
804
    def has_tag_name(node, name):
805
        return node.tag_name == name # and not isinstance(node.parser, Retrieve)
806

807
808
    def test_capture_assertions(self):
        try:
809
            _ = Grammar(Capture(Drop(Whitespace(r'\s*'))))
810
            assert False, "GrammarError expected"
811
        except GrammarError as ge:
812
813
            assert ge.errors and ge.errors[0][-1].code == CAPTURE_DROPPED_CONTENT_WARNING, \
                "Capture-dropped-content-Warning expected"
814
        try:
815
            _ = Grammar(Capture(Series(Text(' '), Drop(Whitespace(r'\s*')))))
816
817
818
819
            assert False, "GrammarError expected"
        except GrammarError as ge:
            assert ge.errors and ge.errors[0][-1].code == CAPTURE_DROPPED_CONTENT_WARNING, \
                "Capture-dropped-content-Warning expected"
820
821
822
        cp = Capture(RegExp(r'\w+'))
        cp.pname = "capture"
        _ = Grammar(cp)
823

824
825
826
827
    def test_compile_mini_language(self):
        assert self.minilang_parser
        assert self.minilang_parser2
        assert self.minilang_parser3
828
        assert self.minilang_parser4
829
830
831
832

    def test_stackhandling(self):
        ambigous_opening = "<ABCnormal> normal tag <ABCnormal*>"
        syntax_tree = self.minilang_parser3(ambigous_opening)
833
        assert not syntax_tree.error_flag, str(syntax_tree.errors_sorted)
834
835
836

        ambigous_opening = "<ABCnormal> normal tag <ABCnormal/>"
        syntax_tree = self.minilang_parser3(ambigous_opening)
837
        assert not syntax_tree.error_flag, str(syntax_tree.errors_sorted)
838
839
840

        forgot_closing_tag = "<em> where is the closing tag?"
        syntax_tree = self.minilang_parser3(forgot_closing_tag)
841
        assert syntax_tree.error_flag, str(syntax_tree.errors_sorted)
842
843
844

        proper = "<em> has closing tag <em/>"
        syntax_tree = self.minilang_parser3(proper)
845
        assert not syntax_tree.error_flag, str(syntax_tree.errors_sorted)
846
847
848

        proper = "<em> has closing tag <em*>"
        syntax_tree = self.minilang_parser3(proper)
849
        assert not syntax_tree.error_flag, str(syntax_tree.errors_sorted)
850

851
    def test_optional_match(self):
852
853
854
855
        # from DHParser.dsl import compileEBNF
        # src = compileEBNF(self.mini_lang4)
        # print(src)
        # return
856
857
858
859
860
861
        test1 = '<info>Hey, you</info>'
        st = self.minilang_parser4(test1)
        assert not st.error_flag, str(st.errors_sorted)
        test12 = '<info>Hey, <emph>you</emph></info>'
        st = self.minilang_parser4(test1)
        assert not st.error_flag
862
        test2 = '<info>Hey, you</>'
863
864
865
        # set_config_value('history_tracking', True)
        # set_tracer(self.minilang_parser4, trace_history)
        # start_logging('LOGS')
866
        st = self.minilang_parser4(test2)
867
868
869
870
871
872
873
874
875
        # log_parsing_history(self.minilang_parser4, "optional_match")
        # print(st.as_sxpr())
        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
876
877
878
879
880
881
882
883
884
885
886

    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

887
    def test_cache_neutrality_1(self):
888
889
        """Test that packrat-caching does not interfere with the variable-
        changing parsers: Capture and Retrieve."""
890
        lang = r"""@literalws = right
891
892
893
894
895
896
897
898
899
            text = opening closing
            opening = (unmarked_package | marked_package)
            closing = ::variable
            unmarked_package = package "."
            marked_package = package "*" "."
            package = "(" variable ")"
            variable = /\w+/~
            """
        case = "(secret)*. secret"
900
        gr = grammar_provider(lang)()
901
        st = gr(case)
902
        assert not st.error_flag, str(st.errors_sorted)
903

904
905
906
907
908
909
910
911
912
913
914
    def test_cache_neutrality_2(self):
        lang = r'''document = variantA | variantB
            variantA  = delimiter `X` ::delimiter `!` 
            variantB  = `A` delimiter ::delimiter `!` 
            delimiter = `A` | `X`
        '''
        gr = grammar_provider(lang)()
        case = 'AXA!'
        st = gr(case)
        assert not st.errors
        case = 'AXX!'
915
916
917
        # set_config_value('history_tracking', True)
        # start_logging('LOGS')
        # set_tracer(gr, trace_history)
918
        st = gr(case)
919
        # log_parsing_history(gr, 'test_cache_neutrality_2')
920
        # print(st.as_sxpr())
921
922
923
924
925
        assert not st.errors
        assert str(st) == "AXX!"

    def test_cache_neutrality_3(selfself):
        lang = r'''document = variantA | variantB
926
927
            variantA  = delimiter `X` [`---`] check ::delimiter `!` 
            variantB  = `A` delimiter [`---`] check ::delimiter `!`
928
929
930
931
            check = :delimiter 
            delimiter = `A` | `X`
        '''
        gr = grammar_provider(lang)()
932
        case = 'AX---XX!'
933
934
935
936
937
938
        # set_config_value('history_tracking', True)
        # start_logging('LOGS')
        # set_tracer(gr, trace_history)
        st = gr(case)
        # log_parsing_history(gr, 'test_cache_neutrality_3')
        # print(st.as_sxpr())
eckhart's avatar
eckhart committed
939
        assert not st.errors, str(errors)
940
941
942
        case = 'AXXX!'
        st = gr(case)
        assert not st.errors
943

944
945
946
    def test_single_line(self):
        teststr = "Anfang ```code block `` <- keine Ende-Zeichen ! ``` Ende"
        syntax_tree = self.minilang_parser(teststr)
947
948
        assert not syntax_tree.errors_sorted, \
            ''.join(str(error) for error in syntax_tree.errors_sorted)
949
950
951
        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)))
952
953
        assert delim == pop
        if is_logging():
954
            log_ST(syntax_tree, "test_PopRetrieve_single_line.cst")
955
956
957
958
959
960
961
962
963
964
965

    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)
966
        assert not syntax_tree.errors_sorted
967
968
969
        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)))
970
971
        assert delim == pop
        if is_logging():
972
            log_ST(syntax_tree, "test_PopRetrieve_multi_line.cst")
973
974
975
976

    def test_single_line_complement(self):
        teststr = "Anfang {{{code block }} <- keine Ende-Zeichen ! }}} Ende"
        syntax_tree = self.minilang_parser2(teststr)
977
        assert not syntax_tree.errors_sorted
978
979
980
981
982
        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
983
        if is_logging():
984
            log_ST(syntax_tree, "test_PopRetrieve_single_line.cst")
985
986
987
988
989
990
991
992
993
994
995

    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)
996
        assert not syntax_tree.errors_sorted
997
998
999
        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)))
1000
        assert len(delim) == len(pop) and delim != pop