test_parse.py 31.1 KB
Newer Older
1
2
#!/usr/bin/python3

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

Author: Eckhart Arnold <arnold@badw.de>

Copyright 2017 Bavarian Academy of Sciences and Humanities

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

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

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

import sys
23
from functools import partial
24

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

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


eckhart's avatar
eckhart committed
39
40
41
42
43
44
45
46
47
48
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
49
            l.append(p.pname + p.ptype)
eckhart's avatar
eckhart committed
50
51
52
53
54
55
56
57
58
59
60
61
        gr.root__.apply(visitor)
        s1 = ", ".join(l)
        l = []
        gr.root__.apply(visitor)
        s2 = ", ".join(l)
        l = []
        gr.root__.apply(visitor)
        s3 = ", ".join(l)
        # print(s1);  print(s2);  print(s3)
        assert s1 == s2 == s3


62
class TestInfiLoopsAndRecursion:
63
    def test_direct_left_recursion1(self):
eckhart's avatar
eckhart committed
64
        minilang ="""
65
66
67
68
            expr = expr ("+"|"-") term | term
            term = term ("*"|"/") factor | factor
            factor = /[0-9]+/~
            """
69
        snippet = "9 + 8 + 7 + 6 + 5 + 3 * 4"
70
        parser = grammar_provider(minilang)()
71
        assert parser
Eckhart Arnold's avatar
Eckhart Arnold committed
72
        syntax_tree = parser(snippet)
73
74
        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
75
        if is_logging():
76
            log_ST(syntax_tree, "test_LeftRecursion_direct.cst")
77
            log_parsing_history(parser, "test_LeftRecursion_direct")
78

79
    def test_direct_left_recursion2(self):
80
81
82
83
84
85
        minilang = """
            expr = ex
            ex   = expr ("+"|"-") term | term
            term = term ("*"|"/") factor | factor
            factor = /[0-9]+/~
            """
86
        snippet = "9 + 8 + 7 + 6 + 5 + 3 * 4"
87
        parser = grammar_provider(minilang)()
88
89
        assert parser
        syntax_tree = parser(snippet)
90
91
        assert not is_error(syntax_tree.error_flag), syntax_tree.errors_sorted
        assert snippet == syntax_tree.content
92

93
    def test_indirect_left_recursion1(self):
94
95
        minilang = """
            Expr    = //~ (Product | Sum | Value)
96
97
            Product = Expr { ('*' | '/') Expr }+
            Sum     = Expr { ('+' | '-') Expr }+
98
99
            Value   = /[0-9.]+/~ | '(' Expr ')'
            """
100
        parser = grammar_provider(minilang)()
101
        assert parser
102
        snippet = "8 * 4"
103
        syntax_tree = parser(snippet)
104
        assert not is_error(syntax_tree.error_flag), syntax_tree.errors_sorted
105
106
        snippet = "7 + 8 * 4"
        syntax_tree = parser(snippet)
107
        assert not is_error(syntax_tree.error_flag), syntax_tree.errors_sorted
108
109
        snippet = "9 + 8 * (4 + 3)"
        syntax_tree = parser(snippet)
110
111
        assert not is_error(syntax_tree.error_flag), syntax_tree.errors_sorted
        assert snippet == syntax_tree.content
112
        if is_logging():
113
            log_ST(syntax_tree, "test_LeftRecursion_indirect.cst")
114
            log_parsing_history(parser, "test_LeftRecursion_indirect")
115

eckhart's avatar
eckhart committed
116
117
118
    def test_infinite_loops(self):
        minilang = """not_forever = { // } \n"""
        snippet = " "
119
        parser = grammar_provider(minilang)()
Eckhart Arnold's avatar
Eckhart Arnold committed
120
        syntax_tree = parser(snippet)
121
        assert syntax_tree.error_flag
eckhart's avatar
eckhart committed
122
        # print(syntax_tree.as_sxpr())
123
        # print(syntax_tree.errors())
124

125

Eckhart Arnold's avatar
Eckhart Arnold committed
126
class TestFlowControl:
127
128
    def setup(self):
        self.t1 = """
eckhart's avatar
eckhart committed
129
        All work and no play
130
131
132
133
134
        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
135
136
137
    def test_lookbehind(self):
        ws = RegExp('\s*')
        end = RegExp("END")
138
        doc_end = Lookbehind(RegExp('\\s*?\\n')) + end
Eckhart Arnold's avatar
Eckhart Arnold committed
139
140
141
142
143
        word = RegExp('\w+')
        sequence = OneOrMore(NegativeLookahead(end) + word + ws)
        document = ws + sequence + doc_end + ws

        parser = Grammar(document)
144
145
146
147
148
149
150
        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
151
            parser_initialization__ = ["upon instantiation"]
152
153
            ws = RegExp('\\s*')
            end = RegExp('END')
154
            SUCC_LB = RegExp('\\s*?\\n')
155
156
157
158
159
160
161
162
            doc_end = Series(Lookbehind(SUCC_LB), end)
            word = RegExp('\w+')
            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
163
        assert not cst.error_flag, cst.as_sxpr()
164
        cst = parser(self.t2)
Eckhart Arnold's avatar
Eckhart Arnold committed
165
166
167
        assert cst.error_flag, cst.as_sxpr()


168
169
170
171
172
173
174
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
175
176
        result, messages, syntax_tree = compile_source(mlregex, None, get_ebnf_grammar(),
                        get_ebnf_transformer(), get_ebnf_compiler('MultilineRegexTest'))
177
        assert result
178
        assert not messages, str(messages)
179
        parser = compile_python_object(DHPARSER_IMPORTS + result, '\w+Grammar$')()
eckhart's avatar
eckhart committed
180
181
        node = parser('abc+def', parser.regex)
        assert not node.error_flag
182
        assert node.tag_name == "regex"
183
184
        assert str(node) == 'abc+def'

185
186
187
188
189
190
191
192
193
194
195
196
197
    def test_multilineRegex_wo_Comments(self):
        mlregex = r"""
        regex =  /\w+ 
                  [+]  
                  \w* /
        """
        result, messages, syntax_tree = compile_source(mlregex, None, get_ebnf_grammar(),
                        get_ebnf_transformer(), get_ebnf_compiler('MultilineRegexTest'))
        assert result
        assert not messages, str(messages)
        parser = compile_python_object(DHPARSER_IMPORTS + result, '\w+Grammar$')()
        node = parser('abc+def', parser.regex)
        assert not node.error_flag
198
        assert node.tag_name == "regex"
199
200
        assert str(node) == 'abc+def'

201
202
203
204
205
206
207
208
209
210
211
212
213
214
    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
        parser = compile_python_object(DHPARSER_IMPORTS + result, '\w+Grammar$')()
        node, rest = parser.regex('Alpha')
        assert node
        assert not node.error_flag
        assert rest == ''
215
        assert node.tag_name == "regex"
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
        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
        parser = compile_python_object(DHPARSER_IMPORTS + result, '\w+Grammar$')()
        node, rest = parser.regex('Alpha')
        assert node.error_flag


231
232
233
234
235
236
237
238
239
240
241
242
    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
243
244
        result, messages, syntax_tree = compile_source(tokenlang, None, get_ebnf_grammar(),
                                    get_ebnf_transformer(), get_ebnf_compiler("TokenTest"))
245
        assert result
eckhart's avatar
eckhart committed
246
        assert not messages, str(messages)
247
        parser = compile_python_object(DHPARSER_IMPORTS + result, '\w+Grammar$')()
Eckhart Arnold's avatar
Eckhart Arnold committed
248
        result = parser(testdoc)
249
        # log_parsing_history(parser, "test.log")
250
251
        assert not result.error_flag

252

253
class TestGrammar:
254
    def setup(self):
255
256
257
258
259
260
        grammar = r"""@whitespace = horizontal
        haupt        = textzeile LEERZEILE
        textzeile    = { WORT }+
        WORT         = /[^ \t]+/~
        LEERZEILE    = /\n[ \t]*(?=\n)/~
        """
261
262
263
        self.pyparser, messages, syntax_tree = compile_source(grammar, None, get_ebnf_grammar(),
                                                              get_ebnf_transformer(), get_ebnf_compiler("PosTest"))
        assert self.pyparser
264
        assert not messages
265
266
267
268

    def test_pos_values_initialized(self):
        # checks whether pos values in the parsing result and in the
        # history record have been initialized
269
        with logging("LOGS"):
270
271
272
            grammar = compile_python_object(DHPARSER_IMPORTS + self.pyparser, '\w+Grammar$')()
            grammar("no_file_name*")
        for record in grammar.history__:
273
274
            assert not record.node or record.node.pos >= 0

275
    def test_select_parsing(self):
276
277
278
279
280
        grammar = compile_python_object(DHPARSER_IMPORTS + self.pyparser, '\w+Grammar$')()
        grammar("wort", "WORT")
        grammar("eine Zeile", "textzeile")
        grammar("kein Haupt", "haupt")
        grammar("so ist es richtig", "haupt")
281

di68kap's avatar
di68kap committed
282
283
284
285
286
287
288
289
290
    def test_grammar_subclassing(self):
        class Arithmetic(Grammar):
            '''
            expression =  term  { ("+" | "-") term }
            term       =  factor  { ("*" | "/") factor }
            factor     =  INTEGER | "("  expression  ")"
            INTEGER    =  /\d+/~
            '''
            expression = Forward()
291
292
293
294
            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
295
296
297
298
299
300
            root__ = expression

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

301

302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
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)()
325
326
327
        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
328
        # print(st.errors())
329
        assert st.errors_sorted[0].code == Error.MANDATORY_CONTINUATION
330
        # transitivity of mandatory-operator
331
        st = parser("ABC_");  assert st.error_flag
332
        assert st.errors_sorted[0].code == Error.MANDATORY_CONTINUATION
333
334

    def test_series_composition(self):
335
        TA, TB, TC, TD, TE = (TKN(b) for b in "ABCDE")
336
337
338
339
340
        s1 = Series(TA, TB, TC, mandatory=2)
        s2 = Series(TD, TE)

        combined = Alternative(s1 + s2, RegExp('.*'))
        parser = Grammar(combined)
341
342
343
        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
344
        assert st.errors_sorted[0].code == Error.MANDATORY_CONTINUATION
345
        st = parser("ABC_E");  assert st.error_flag
346
        assert st.errors_sorted[0].code == Error.MANDATORY_CONTINUATION
347
348
349

        combined = Alternative(s2 + s1, RegExp('.*'))
        parser = Grammar(combined)
350
351
352
353
354
355
        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
356
        assert st.errors_sorted[0].code == Error.MANDATORY_CONTINUATION
357

eckhart's avatar
eckhart committed
358
359
360
361
362
363
364
365
366
367
    # 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
368

369

370
371
372
class TestAllOfSomeOf:
    def test_allOf_order(self):
        """Test that parsers of an AllOf-List can match in arbitrary order."""
373
        prefixes = AllOf(TKN("A"), TKN("B"))
374
375
        assert Grammar(prefixes)('A B').content == 'A B'
        assert Grammar(prefixes)('B A').content == 'B A'
376
        # aternative Form
377
        prefixes = AllOf(Series(TKN("B"), TKN("A")))
378
        assert Grammar(prefixes)('A B').content == 'A B'
379
380
381
382

    def test_allOf_completeness(self):
        """Test that an error is raised if not  all parsers of an AllOf-List
        match."""
383
        prefixes = AllOf(TKN("A"), TKN("B"))
384
385
386
387
388
        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."""
389
        prefixes = AllOf(TKN("A"), TKN("B"), TKN("A"))
390
391
392
        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'
393
394
395
396
        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."""
397
        prefixes = SomeOf(TKN("A"), TKN("B"))
398
399
        assert Grammar(prefixes)('A B').content == 'A B'
        assert Grammar(prefixes)('B A').content == 'B A'
400
        # aternative Form
401
        prefixes = SomeOf(Alternative(TKN("B"), TKN("A")))
402
403
        assert Grammar(prefixes)('A B').content == 'A B'
        assert Grammar(prefixes)('B').content == 'B'
404
405
406
407

    def test_someOf_redundance(self):
        """Test that one and the same parser may be listed several times
        and must be matched several times accordingly."""
408
        prefixes = SomeOf(TKN("A"), TKN("B"), TKN("A"))
409
410
411
        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'
412
413
414
        assert Grammar(prefixes)('A B B').error_flag


415
416
417
418
class TestPopRetrieve:
    mini_language = """
        document       = { text | codeblock }
        codeblock      = delimiter { text | (!:delimiter delimiter_sign) } ::delimiter
419
        delimiter      = delimiter_sign  # never use delimiter between capture and pop except for retrival!
420
        delimiter_sign = /`+/
eckhart's avatar
eckhart committed
421
        text           = /[^`]+/
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
        """
    mini_lang2 = """
        @braces_filter=counterpart
        document       = { text | codeblock }
        codeblock      = braces { text | opening_braces | (!:braces closing_braces) } ::braces
        braces         = opening_braces
        opening_braces = /\{+/
        closing_braces = /\}+/
        text           = /[^{}]+/
        """
    mini_lang3 = """
        document       = { text | env }
        env            = (specialtag | opentag) text [closespecial | closetag]
        opentag        = "<" name ">"
        specialtag     = "<" /ABC/ !name ">"
eckhart's avatar
eckhart committed
437
        closetag       = close_slash | close_star
438
439
440
441
442
443
444
445
        close_slash    = "<" ::name "/>"
        close_star     = "<" ::name "*>"
        closespecial   = "<" /ABC/~ ">"
        name           = /\w+/~
        text           = /[^<>]+/
        """

    def setup(self):
446
447
448
        self.minilang_parser = grammar_provider(self.mini_language)()
        self.minilang_parser2 = grammar_provider(self.mini_lang2)()
        self.minilang_parser3 = grammar_provider(self.mini_lang3)()
449
450
451

    @staticmethod
    def opening_delimiter(node, name):
452
        return node.tag_name == name # and not isinstance(node.parser, Retrieve)
453
454
455

    @staticmethod
    def closing_delimiter(node):
456
457
        return node.tag_name in {':Pop', ':Retrieve'}
        # return isinstance(node.parser, Retrieve)
458
459
460
461
462
463
464
465
466

    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)
467
        assert not syntax_tree.error_flag, str(syntax_tree.errors_sorted)
468
469
470

        ambigous_opening = "<ABCnormal> normal tag <ABCnormal/>"
        syntax_tree = self.minilang_parser3(ambigous_opening)
471
        assert not syntax_tree.error_flag, str(syntax_tree.errors_sorted)
472
473
474

        forgot_closing_tag = "<em> where is the closing tag?"
        syntax_tree = self.minilang_parser3(forgot_closing_tag)
475
        assert syntax_tree.error_flag, str(syntax_tree.errors_sorted)
476
477
478

        proper = "<em> has closing tag <em/>"
        syntax_tree = self.minilang_parser3(proper)
479
        assert not syntax_tree.error_flag, str(syntax_tree.errors_sorted)
480
481
482

        proper = "<em> has closing tag <em*>"
        syntax_tree = self.minilang_parser3(proper)
483
        assert not syntax_tree.error_flag, str(syntax_tree.errors_sorted)
484

485
    def test_cache_neutrality(self):
486
487
        """Test that packrat-caching does not interfere with the variable-
        changing parsers: Capture and Retrieve."""
488
489
490
491
492
493
494
495
496
497
        lang = """
            text = opening closing
            opening = (unmarked_package | marked_package)
            closing = ::variable
            unmarked_package = package "."
            marked_package = package "*" "."
            package = "(" variable ")"
            variable = /\w+/~
            """
        case = "(secret)*. secret"
498
        gr = grammar_provider(lang)()
499
        st = gr(case)
500
        assert not st.error_flag, str(st.errors_sorted)
501
502
503
504

    def test_single_line(self):
        teststr = "Anfang ```code block `` <- keine Ende-Zeichen ! ``` Ende"
        syntax_tree = self.minilang_parser(teststr)
505
        assert not syntax_tree.errors_sorted
506
507
        delim = str(next(syntax_tree.select(partial(self.opening_delimiter, name="delimiter"))))
        pop = str(next(syntax_tree.select(self.closing_delimiter)))
508
509
        assert delim == pop
        if is_logging():
510
            log_ST(syntax_tree, "test_PopRetrieve_single_line.cst")
511
512
513
514
515
516
517
518
519
520
521

    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)
522
        assert not syntax_tree.errors_sorted
523
524
        delim = str(next(syntax_tree.select(partial(self.opening_delimiter, name="delimiter"))))
        pop = str(next(syntax_tree.select(self.closing_delimiter)))
525
526
        assert delim == pop
        if is_logging():
527
            log_ST(syntax_tree, "test_PopRetrieve_multi_line.cst")
528
529
530
531

    def test_single_line_complement(self):
        teststr = "Anfang {{{code block }} <- keine Ende-Zeichen ! }}} Ende"
        syntax_tree = self.minilang_parser2(teststr)
532
        assert not syntax_tree.errors_sorted
533
534
        delim = str(next(syntax_tree.select(partial(self.opening_delimiter, name="braces"))))
        pop = str(next(syntax_tree.select(self.closing_delimiter)))
535
536
        assert len(delim) == len(pop) and delim != pop
        if is_logging():
537
            log_ST(syntax_tree, "test_PopRetrieve_single_line.cst")
538
539
540
541
542
543
544
545
546
547
548

    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)
549
        assert not syntax_tree.errors_sorted
550
551
        delim = str(next(syntax_tree.select(partial(self.opening_delimiter, name="braces"))))
        pop = str(next(syntax_tree.select(self.closing_delimiter)))
552
553
        assert len(delim) == len(pop) and delim != pop
        if is_logging():
554
            log_ST(syntax_tree, "test_PopRetrieve_multi_line.cst")
555
556


557
class TestWhitespaceHandling:
Eckhart Arnold's avatar
Eckhart Arnold committed
558
    minilang = """
559
560
561
562
563
564
565
566
567
        doc = A B
        A = "A"
        B = "B"
        Rdoc = ar br
        ar = /A/
        br = /B/
        """

    def setup(self):
568
        self.gr = grammar_provider(self.minilang)()
569
570
571
572
573
574
575
576
577
578
579
580

    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
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600


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

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

    def test_error_propagation(self):
        testcode1 = "halloB"
        testcode2 = "XYZ"
        testcode3 = "hallo "
        cst = self.parser(testcode1)
601
        assert not cst.error_flag, str(cst.errors_sorted)
di68kap's avatar
di68kap committed
602
603
604
605
        cst = self.parser(testcode2)
        assert not cst.error_flag
        cst = self.parser(testcode3)
        assert cst.error_flag
606
607


di68kap's avatar
di68kap committed
608
609
610
611
612
613
614
class TestBorderlineCases:
    def test_not_matching(self):
        minilang = """parser = /X/"""
        gr = grammar_provider(minilang)()
        cst = gr('X', 'parser')
        assert not cst.error_flag
        cst = gr(' ', 'parser')
615
        assert cst.error_flag and cst.errors_sorted[0].code == Error.PARSER_DID_NOT_MATCH
di68kap's avatar
di68kap committed
616
        cst = gr('', 'parser')
617
        assert cst.error_flag and cst.errors_sorted[0].code == Error.PARSER_DID_NOT_MATCH
di68kap's avatar
di68kap committed
618
619
620
621
622
623
624

    def test_matching(self):
        minilang = """parser = /.?/"""
        gr = grammar_provider(minilang)()
        cst = gr(' ', 'parser')
        assert not cst.error_flag
        cst = gr('  ', 'parser')
625
        assert cst.error_flag and cst.errors_sorted[0].code == Error.PARSER_STOPPED_BEFORE_END
di68kap's avatar
di68kap committed
626
627
628
629
        cst = gr('', 'parser')
        assert not cst.error_flag


Eckhart Arnold's avatar
Eckhart Arnold committed
630
class TestReentryAfterError:
631
    def setup(self):
Eckhart Arnold's avatar
Eckhart Arnold committed
632
633
634
635
636
        lang = """
        document = alpha [beta] gamma "."
          alpha = "ALPHA" abc
            abc = §"a" "b" "c"
          beta = "BETA" (bac | bca)
637
638
            bac = "b" "a" §"c"
            bca = "b" "c" §"a"
Eckhart Arnold's avatar
Eckhart Arnold committed
639
640
641
642
          gamma = "GAMMA" §(cab | cba)
            cab = "c" "a" §"b"
            cba = "c" "b" §"a"
        """
643
        self.gr = grammar_provider(lang)()
644

645
646
    def test_no_resume_rules(self):
        gr = self.gr;  gr.resume_rules = dict()
647
648
        content = 'ALPHA acb BETA bac GAMMA cab .'
        cst = gr(content)
649
650
651
652
653
        # print(cst.as_sxpr())
        assert cst.error_flag
        assert cst.content == content
        assert cst.pick('alpha').content.startswith('ALPHA')

eckhart's avatar
eckhart committed
654
655
656
657
658
659
660
661
662
    def test_no_resume_rules_partial_parsing(self):
        gr = self.gr;  gr.resume_rules = dict()
        content = 'ALPHA acb'
        cst = gr(content, 'alpha')
        # print(cst.as_sxpr())
        assert cst.error_flag
        assert cst.content == content
        assert cst.pick('alpha').content.startswith('ALPHA')

663
664
    def test_simple_resume_rule(self):
        gr = self.gr;  gr.resume_rules = dict()
665
        gr.resume_rules__['alpha'] = ['BETA']
666
        content = 'ALPHA acb BETA bac GAMMA cab .'
667
668
669
670
671
672
        cst = gr(content)
        # print(cst.as_sxpr())
        assert cst.error_flag
        assert cst.content == content
        assert cst.pick('alpha').content.startswith('ALPHA')
        # because of resuming, there should be only on error message
673
        assert len(cst.errors_sorted) == 1
674

675
676
    def test_failing_resume_rule(self):
        gr = self.gr;  gr.resume_rules = dict()
677
        gr.resume_rules__['alpha'] = ['XXX']
678
        content = 'ALPHA acb BETA bac GAMMA cab .'
679
680
681
682
683
684
685
        cst = gr(content)
        # print(cst.as_sxpr())
        assert cst.error_flag
        assert cst.content == content
        # assert cst.pick('alpha').content.startswith('ALPHA')
        # because of resuming, there should be only on error message

686
687
    def test_severl_reentry_points(self):
        gr = self.gr;  gr.resume_rules = dict()
688
        gr.resume_rules__['alpha'] = ['BETA', 'GAMMA']
689
        content = 'ALPHA acb BETA bac GAMMA cab .'
690
691
692
693
694
695
        cst = gr(content)
        # print(cst.as_sxpr())
        assert cst.error_flag
        assert cst.content == content
        assert cst.pick('alpha').content.startswith('ALPHA')
        # because of resuming, there should be only on error message
696
        assert len(cst.errors_sorted) == 1
697

698
699
    def test_several_reentry_points_second_point_matching(self):
        gr = self.gr;  gr.resume_rules = dict()
700
701
702
        gr.resume_rules__['alpha'] = ['BETA', 'GAMMA']
        content = 'ALPHA acb GAMMA cab .'
        cst = gr(content)
703
        # print(cst.as_sxpr())
Eckhart Arnold's avatar
Eckhart Arnold committed
704
        assert cst.error_flag
705
        assert cst.content == content
706
707
        assert cst.pick('alpha').content.startswith('ALPHA')
        # because of resuming, there should be only on error message
708
        assert len(cst.errors_sorted) == 1
709

710
711
712
713
714
715
716
717
718
719
720
721
    def test_several_resume_rules_innermost_rule_matching(self):
        gr = self.gr;  gr.resume_rules = dict()
        gr.resume_rules__['alpha'] = ['BETA', 'GAMMA']
        gr.resume_rules__['beta'] = ['GAMMA']
        gr.resume_rules__['bac'] = ['GAMMA']
        content = 'ALPHA abc BETA bad GAMMA cab .'
        cst = gr(content)
        # print(cst.as_sxpr())
        assert cst.error_flag
        assert cst.content == content
        assert cst.pick('alpha').content.startswith('ALPHA')
        # because of resuming, there should be only on error message
722
        assert len(cst.errors_sorted) == 1
723
724
725
726
727
728
729
730
        # multiple failures
        content = 'ALPHA acb BETA bad GAMMA cab .'
        cst = gr(content)
        # print(cst.as_sxpr())
        assert cst.error_flag
        assert cst.content == content
        assert cst.pick('alpha').content.startswith('ALPHA')
        # because of resuming, there should be only on error message
731
        assert len(cst.errors_sorted) == 2
732

733

734
735
736
737
738
739
740
741
742
743
class TestConfiguredErrorMessages:
    def test_(self):
        lang = """
            document = series | /.*/
            @series_error = "a badly configured error message {5}"
            series = "X" | head §"C" "D"
            head = "A" "B"
            """
        parser = grammar_provider(lang)()
        st = parser("AB_D");  assert st.error_flag
744
745
        assert st.errors_sorted[0].code == Error.MALFORMED_ERROR_STRING
        assert st.errors_sorted[1].code == Error.MANDATORY_CONTINUATION
746
        # print(st.errors())
747
748


749
750
751
752
753
754
755
756
757
758
class TestUnknownParserError:
    def test_unknown_parser_error(self):
        gr = Grammar()
        try:
            gr("", "NonExistantParser")
            assert False, "UnknownParserError expected!"
        except UnknownParserError:
            pass


759
760
761
762
763
764
class TestEarlyTokenWhitespaceDrop:
    def setup(self):
        self.lang = """
            @ drop = token, whitespace
            expression = term  { ("+" | "-") term}
            term       = factor  { ("*"|"/") factor}
765
766
            factor     = number | variable | "("  expression  ")" 
                       | constant | fixed
767
            variable   = /[a-z]/~
768
769
770
            number     = /\d+/~
            constant   = "A" | "B"
            fixed      = "X"   
771
772
773
            """
        self.gr = grammar_provider(self.lang)()

774
    def test_drop(self):
775
776
777
778
        cst = self.gr('4 + 3 * 5')
        # print(cst.as_sxpr())
        assert not cst.pick(':Token')
        assert not cst.pick(':Whitespace')
779
780
781
782
783
784
785
786
        cst = self.gr('A + B')
        try:
            _ = next(cst.select(lambda node: node.content == 'A'))
            assert False, "Tokens in compound expressions should be dropped!"
        except StopIteration:
            pass
        cst = self.gr('X * y')
        assert next(cst.select(lambda node: node.content == 'X'))
787

788

Eckhart Arnold's avatar
Eckhart Arnold committed
789
790
class TestMetaParser:
    def test_meta_parser(self):
791
792
        save = get_config_value('flatten_tree_while_parsing')
        set_config_value('flatten_tree_while_parsing', True)
Eckhart Arnold's avatar
Eckhart Arnold committed
793
        mp = MetaParser()
794
        mp.grammar = Grammar()  # override placeholder warning
Eckhart Arnold's avatar
Eckhart Arnold committed
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
        mp.pname = "named"
        mp.tag_name = mp.pname
        nd = mp._return_value(Node('tagged', 'non-empty'))
        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"
        nd = mp._return_value(Node('tagged', ''))
        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
        nd = mp._return_value(Node(':anonymous', 'content'))
        assert nd.tag_name == 'named', nd.as_sxpr()
        assert not nd.children
        assert nd.result == 'content'
        nd = mp._return_value(Node(':anonymous', ''))
        assert nd.tag_name == 'named', nd.as_sxpr()
        assert not nd.children
        assert not nd.content
        mp.pname = ''
        mp.tag_name = ':unnamed'
        nd = mp._return_value(Node('tagged', 'non-empty'))
        assert nd.tag_name == 'tagged', nd.as_sxpr()
        assert len(nd.children) == 0
        assert nd.content == 'non-empty'
        nd = mp._return_value(Node('tagged', ''))
        assert nd.tag_name == 'tagged', nd.as_sxpr()
        assert len(nd.children) == 0
        assert not nd.content
        nd = mp._return_value(Node(':anonymous', 'content'))
        assert nd.tag_name == ':anonymous', nd.as_sxpr()
        assert not nd.children
        assert nd.result == 'content'
        nd = mp._return_value(Node('', ''))
        assert nd.tag_name == '', nd.as_sxpr()
        assert not nd.children
        assert not nd.content
        assert mp._return_value(None) == EMPTY_NODE
834
        set_config_value('flatten_tree_while_parsing', save)
Eckhart Arnold's avatar
Eckhart Arnold committed
835
836
837



838
if __name__ == "__main__":
839
    from DHParser.testing import runner
840
841
    with logging(False):
        runner("", globals())