testing.py 14.8 KB
Newer Older
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
"""testing.py - test support for DHParser based grammars and compilers

Copyright 2016  by Eckhart Arnold (arnold@badw.de)
                Bavarian Academy of Sciences an Humanities (badw.de)

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.
"""
18
import collections
19
import configparser
20
import copy
di68kap's avatar
di68kap committed
21
import inspect
22
23
import json
import os
24
25
26
27
try:
    import regex as re
except ImportError:
    import re
28
29

from DHParser import Node, error_messages
30
from DHParser.toolkit import compact_sexpr, is_logging
31
32
from DHParser.syntaxtree import MockParser
from DHParser.ebnf import grammar_changed
33
from DHParser.dsl import compile_on_disk
34
35
36


def mock_syntax_tree(sexpr):
37
38
    """
    Generates a tree of nodes from an S-expression.
39
40

    Example: 
41
    >>> mock_syntax_tree("(a (b c))").as_sxpr()
42
    '(a\\n    (b\\n        "c"\\n    )\\n)'
43
44
45
46
47
48
    """
    def next_block(s):
        s = s.strip()
        while s[0] != ')':
            if s[0] != '(': raise ValueError('"(" expected, not ' + s[:10])
            # assert s[0] == '(', s
49
            level = 1
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
            i = 1
            while level > 0:
                if s[i] == '(':
                    level += 1
                elif s[i] == ')':
                    level -= 1
                i += 1
            yield s[:i]
            s = s[i:].strip()

    sexpr = sexpr.strip()
    if sexpr[0] != '(': raise ValueError('"(" expected, not ' + sexpr[:10])
    # assert sexpr[0] == '(', sexpr
    sexpr = sexpr[1:].strip()
    m = re.match('[\w:]+', sexpr)
    name, class_name = (sexpr[:m.end()].split(':') + [''])[:2]
    sexpr = sexpr[m.end():].strip()
    if sexpr[0] == '(':
        result = tuple(mock_syntax_tree(block) for block in next_block(sexpr))
    else:
        lines = []
        while sexpr and sexpr[0] != ')':
            for qm in ['"""', "'''", '"', "'"]:
                m = re.match(qm + r'.*?' + qm, sexpr)
                if m:
                    i = len(qm)
                    lines.append(sexpr[i:m.end() - i])
                    sexpr = sexpr[m.end():].strip()
                    break
            else:
                m = re.match(r'(?:(?!\)).)*', sexpr)
                lines.append(sexpr[:m.end()])
                sexpr = sexpr[m.end():]
        result = "\n".join(lines)
    return Node(MockParser(name, ':' + class_name), result)


87
def recompile_grammar(ebnf_filename, force=False) -> bool:
88
89
    """
    Recompiles an ebnf-grammar if necessary, that is if either no
90
    corresponding 'XXXXCompiler.py'-file exists or if that file is
91
92
93
94
95
96
97
    outdated.
    
    Parameters:
        ebnf_filename(str):  The filename of the ebnf-source of the 
            grammar. In case this is a directory and not a file all
            files within this directory ending with .ebnf will be 
            compiled.
98
99
        force(bool):  If False (default), the grammar will only be
            recompiled if it has been changed.
100
101
    """
    if os.path.isdir(ebnf_filename):
102
        success = True
103
104
        for entry in os.listdir(ebnf_filename):
            if entry.lower().endswith('.ebnf') and os.path.isfile(entry):
105
106
                success = success and recompile_grammar(entry, force)
        return success
107
108

    base, ext = os.path.splitext(ebnf_filename)
109
    compiler_name = base + 'Compiler.py'
110
    error_file_name = base + '_ebnf_ERRORS.txt'
111
    errors = []
112
    if (not os.path.exists(compiler_name) or force or
113
114
115
116
117
        grammar_changed(compiler_name, ebnf_filename)):
        # print("recompiling parser for: " + ebnf_filename)
        errors = compile_on_disk(ebnf_filename)
        if errors:
            # print("Errors while compiling: " + ebnf_filename + '!')
118
            with open(error_file_name, 'w') as f:
119
120
121
                for e in errors:
                    f.write(e)
                    f.write('\n')
122
            return False
123

124
125
    if not errors and os.path.exists(error_file_name):
        os.remove(error_file_name)
126
    return True
127

128

129
UNIT_STAGES = {'match', 'fail', 'ast', 'cst', '__ast__', '__cst__'}
130
131


132
def unit_from_configfile(config_filename):
133
134
    """
    Reads a grammar unit test from a config file.
135
    """
136
    cfg = configparser.ConfigParser(interpolation=None)
137
    cfg.read(config_filename)
138
139
    OD = collections.OrderedDict
    unit = OD()
140
141
    for section in cfg.sections():
        symbol, stage = section.split(':')
142
143
        if stage not in UNIT_STAGES:
            if symbol in UNIT_STAGES:
144
145
                symbol, stage = stage, symbol
            else:
146
                raise ValueError('Test stage %s not in: ' % (stage, str(UNIT_STAGES)))
147
        for testkey, testcode in cfg[section].items():
148
149
150
151
            if testcode[:3] + testcode[-3:] in {"''''''", '""""""'}:
                testcode = testcode[3:-3]
            elif testcode[:1] + testcode[-1:] in {"''", '""'}:
                testcode = testcode[1:-1]
152
            unit.setdefault(symbol, OD()).setdefault(stage, OD())[testkey] = testcode
153
154
    # print(json.dumps(unit, sort_keys=True, indent=4))
    return unit
155

156

157
def unit_from_json(json_filename):
158
159
    """
    Reads a grammar unit test from a json file.
160
    """
161
    with open(json_filename, 'r') as f:
162
163
164
165
166
167
168
        unit = json.load(f)
    for symbol in unit:
        for stage in unit[symbol]:
            if stage not in UNIT_STAGES:
                raise ValueError('Test stage %s not in: ' % (stage, str(UNIT_STAGES)))
    return unit

169
# TODO: add support for yaml, cson, toml
170
171


172
def unit_from_file(filename):
173
174
    """
    Reads a grammar unit test from a file. The format of the file is
175
176
    determined by the ending of its name.
    """
177
178
179
180
    if filename.endswith(".json"):
        return unit_from_json(filename)
    elif filename.endswith(".ini"):
        return unit_from_configfile(filename)
181
    else:
182
        raise ValueError("Unknown unit test file type: " + filename[filename.rfind('.'):])
183
184


185
def get_report(test_unit):
186
187
    """
    Returns a text-report of the results of a grammar unit test.
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
    """
    report = []
    for parser_name, tests in test_unit.items():
        heading = 'Test of parser: "%s"' % parser_name
        report.append('\n\n%s\n%s\n' % (heading, '=' * len(heading)))
        for test_name, test_code in tests.get('match', dict()).items():
            heading = 'Match-test "%s"' % test_name
            report.append('\n%s\n%s\n' % (heading, '-' * len(heading)))
            report.append('### Test-code:')
            report.append(test_code)
            error = tests.get('__err__', {}).get(test_name, "")
            if error:
                report.append('\n### Error:')
                report.append(error)
            ast = tests.get('__ast__', {}).get(test_name, None)
203
204
205
            cst = tests.get('__cst__', {}).get(test_name, None)
            if cst and (not ast or cst == ast):
                report.append('\n### CST')
206
                report.append(cst.as_sxpr())
207
            elif ast:
208
                report.append('\n### AST')
209
                report.append(ast.as_sxpr())
210
211
212
    return '\n'.join(report)


213
def grammar_unit(test_unit, parser_factory, transformer_factory, report=True, verbose=False):
214
215
    """
    Unit tests for a grammar-parser and ast transformations.
216
    """
217
    if isinstance(test_unit, str):
218
        unit_dir, unit_name = os.path.split(os.path.splitext(test_unit)[0])
219
        test_unit = unit_from_file(test_unit)
220
    else:
221
        unit_dir = ""
222
        unit_name = str(id(test_unit))
223
224
    if verbose:
        print("\nUnit: " + unit_name)
225
226
227
    errata = []
    parser = parser_factory()
    transform = transformer_factory()
228
229
    for parser_name, tests in test_unit.items():
        assert set(tests.keys()).issubset(UNIT_STAGES)
230
231
        if verbose:
            print('  Match-Tests for parser "' + parser_name + '"')
232
        for test_name, test_code in tests.get('match', dict()).items():
233
234
235
            if verbose:
                infostr = '    match-test "' + test_name + '" ... '
                errflag = len(errata)
236
            cst = parser(test_code, parser_name)
237
238
            if is_logging():
                cst.log("match_%s_%s.cst" % (parser_name, test_name))
239
                parser.log_parsing_history__("match_%s_%s.log" % (parser_name, test_name))
240
            tests.setdefault('__cst__', {})[test_name] = cst
241
            if "ast" in tests or report:
242
243
244
                ast = copy.deepcopy(cst)
                transform(ast)
                tests.setdefault('__ast__', {})[test_name] = ast
245
246
                if is_logging():
                    ast.log("match_%s_%s.ast" % (parser_name, test_name))
247
            if cst.error_flag:
248
                errata.append('Match test "%s" for parser "%s" failed:\n\tExpr.:  %s\n\n\t%s' %
249
                              (test_name, parser_name, '\n\t'.join(test_code.split('\n')),
250
251
                               '\n\t'.join(m.replace('\n', '\n\t\t') for m in
                                           error_messages(test_code, cst.collect_errors()))))
252
                tests.setdefault('__err__', {})[test_name] = errata[-1]
253
254
            elif "cst" in tests and mock_syntax_tree(tests["cst"][test_name]) != cst:
                    errata.append('Concrete syntax tree test "%s" for parser "%s" failed:\n%s' %
255
                                  (test_name, parser_name, cst.as_sxpr()))
256
257
258
259
260
261
            elif "ast" in tests:
                compare = mock_syntax_tree(tests["ast"][test_name])
                if compare != ast:
                    errata.append('Abstract syntax tree test "%s" for parser "%s" failed:'
                                  '\n\tExpr.:     %s\n\tExpected:  %s\n\tReceived:  %s'
                                  % (test_name, parser_name, '\n\t'.join(test_code.split('\n')),
262
263
                                     compact_sexpr(compare.as_sxpr()),
                                     compact_sexpr(ast.as_sxpr())))
264
                    tests.setdefault('__err__', {})[test_name] = errata[-1]
265
            if verbose:
266
                print(infostr + ("OK" if len(errata) == errflag else "FAIL"))
267

268
        if verbose and 'fail' in tests:
269
            print('  Fail-Tests for parser "' + parser_name + '"')
270
        for test_name, test_code in tests.get('fail', dict()).items():
271
272
273
            if verbose:
                infostr = '    fail-test  "' + test_name + '" ... '
                errflag = len(errata)
274
            cst = parser(test_code, parser_name)
275
276
            if is_logging():
                cst.log("fail_%s_%s.cst" % (parser_name, test_name))
277
                parser.log_parsing_history__("fail_%s_%s.log" % (parser_name, test_name))
278
279
280
            if not cst.error_flag:
                errata.append('Fail test "%s" for parser "%s" yields match instead of '
                              'expected failure!' % (test_name, parser_name))
281
                tests.setdefault('__err__', {})[test_name] = errata[-1]
282
283
            if verbose:
                print(infostr + "OK" if len(errata) == errflag else "FAIL")
284

285
286
    # write test-report
    if report:
287
288
289
290
        report_dir = os.path.join(unit_dir, "REPORT")
        if not os.path.exists(report_dir):
            os.mkdir(report_dir)
        with open(os.path.join(report_dir, unit_name + '.report'), 'w') as f:
291
            f.write(get_report(test_unit))
292

293
294
295
    return errata


296
def grammar_suite(directory, parser_factory, transformer_factory, ignore_unknown_filetypes=False,
297
                  report=True, verbose=False):
298
299
    """
    Runs all grammar unit tests in a directory. A file is considered a test
300
301
302
    unit, if it has the word "test" in its name.
    """
    all_errors = collections.OrderedDict()
303
304
    if verbose:
        print("\nScanning test-directory: " + directory)
305
306
307
    for filename in sorted(os.listdir(directory)):
        if filename.lower().find("test") >= 0:
            try:
308
309
                if verbose:
                    print("\nRunning grammar tests from: " + filename)
310
                errata = grammar_unit(os.path.join(directory, filename),
311
                                      parser_factory, transformer_factory, report, verbose)
312
313
314
                if errata:
                    all_errors[filename] = errata
            except ValueError as e:
315
                if not ignore_unknown_filetypes or str(e).find("Unknown") < 0:
316
317
318
319
320
321
322
323
                    raise e
    error_report = []
    if all_errors:
        for filename in all_errors:
            error_report.append('Errors found by unit test "%s":' % filename)
            for error in all_errors[filename]:
                error_report.append('\t' + '\n\t'.join(error.split('\n')))
    if error_report:
324
        return ('Test suite "%s" revealed some errors:\n\n' % directory) + '\n'.join(error_report)
325
326
327
    return ''


328
def runner(tests, namespace):
329
330
    """
    Runs all or some selected Python unit tests found in the
331
332
    namespace. To run all tests in a module, call 
    ``runner("", globals())`` from within that module.
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351

    Args:
        tests: Either a string or a list of strings that contains the
            names of test or test classes. Each test and, in the case
            of a test class, all tests within the test class will be
            run.
        namespace: The namespace for running the test, usually 
            ``globals()`` should be used.
            
    Example:
        class TestSomething()
            def setup(self):
                pass
            def teardown(self):
                pass
            def test_something(self):
                pass
                
        if __name__ == "__main__":
di68kap's avatar
di68kap committed
352
            from DHParser.testing import runner
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
            runner("", globals())        
    """
    def instantiate(cls_name):
        exec("obj = " + cls_name + "()", namespace)
        obj = namespace["obj"]
        if "setup" in dir(obj):
            obj.setup()
        return obj

    if tests:
        if isinstance(tests, str):
            tests = tests.split(" ")
    else:
        # collect all test classes, in case no methods or classes have been passed explicitly
        tests = []
        for name in namespace.keys():
            if name.lower().startswith('test') and inspect.isclass(namespace[name]):
                tests.append(name)

    obj = None
    for test in tests:
        try:
            if test.find('.') >= 0:
                cls_name, method_name = test.split('.')
                obj = instantiate(cls_name)
                print("Running " + cls_name + "." + method_name)
                exec('obj.' + method_name + '()')
            else:
                obj = instantiate(test)
                for name in dir(obj):
                    if name.lower().startswith("test"):
                        print("Running " + test + "." + name)
                        exec('obj.' + name + '()')
        finally:
            if "teardown" in dir(obj):
                obj.teardown()