testing.py 11.7 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 error_messages
30
from DHParser.toolkit import is_logging
Eckhart Arnold's avatar
Eckhart Arnold committed
31
from DHParser.syntaxtree import mock_syntax_tree, oneliner_sxpr
32

33
__all__ = ('unit_from_configfile',
34
35
36
37
38
39
40
           'unit_from_json',
           'unit_from_file',
           'get_report',
           'grammar_unit',
           'grammar_suite',
           'runner')

41
UNIT_STAGES = {'match', 'fail', 'ast', 'cst', '__ast__', '__cst__'}
42
43


44
def unit_from_configfile(config_filename):
45
46
    """
    Reads a grammar unit test from a config file.
47
    """
48
    cfg = configparser.ConfigParser(interpolation=None)
49
    cfg.read(config_filename)
50
51
    OD = collections.OrderedDict
    unit = OD()
52
53
    for section in cfg.sections():
        symbol, stage = section.split(':')
54
55
        if stage not in UNIT_STAGES:
            if symbol in UNIT_STAGES:
56
57
                symbol, stage = stage, symbol
            else:
58
                raise ValueError('Test stage %s not in: ' % (stage, str(UNIT_STAGES)))
59
        for testkey, testcode in cfg[section].items():
60
61
62
63
            if testcode[:3] + testcode[-3:] in {"''''''", '""""""'}:
                testcode = testcode[3:-3]
            elif testcode[:1] + testcode[-1:] in {"''", '""'}:
                testcode = testcode[1:-1]
64
            unit.setdefault(symbol, OD()).setdefault(stage, OD())[testkey] = testcode
65
66
    # print(json.dumps(unit, sort_keys=True, indent=4))
    return unit
67

68

69
def unit_from_json(json_filename):
70
71
    """
    Reads a grammar unit test from a json file.
72
    """
73
    with open(json_filename, 'r') as f:
74
75
76
77
78
79
80
        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

81
# TODO: add support for yaml, cson, toml
82
83


84
def unit_from_file(filename):
85
86
    """
    Reads a grammar unit test from a file. The format of the file is
87
88
    determined by the ending of its name.
    """
89
90
91
92
    if filename.endswith(".json"):
        return unit_from_json(filename)
    elif filename.endswith(".ini"):
        return unit_from_configfile(filename)
93
    else:
94
        raise ValueError("Unknown unit test file type: " + filename[filename.rfind('.'):])
95
96


97
def get_report(test_unit):
98
99
    """
    Returns a text-report of the results of a grammar unit test.
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
    """
    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)
115
116
117
            cst = tests.get('__cst__', {}).get(test_name, None)
            if cst and (not ast or cst == ast):
                report.append('\n### CST')
118
                report.append(cst.as_sxpr())
119
            elif ast:
120
                report.append('\n### AST')
121
                report.append(ast.as_sxpr())
122
123
124
    return '\n'.join(report)


125
def grammar_unit(test_unit, parser_factory, transformer_factory, report=True, verbose=False):
126
127
    """
    Unit tests for a grammar-parser and ast transformations.
128
    """
129
    if isinstance(test_unit, str):
130
        unit_dir, unit_name = os.path.split(os.path.splitext(test_unit)[0])
131
        test_unit = unit_from_file(test_unit)
132
    else:
133
        unit_dir = ""
134
        unit_name = str(id(test_unit))
135
136
    if verbose:
        print("\nUnit: " + unit_name)
137
138
139
    errata = []
    parser = parser_factory()
    transform = transformer_factory()
140
141
    for parser_name, tests in test_unit.items():
        assert set(tests.keys()).issubset(UNIT_STAGES)
142
143
        if verbose:
            print('  Match-Tests for parser "' + parser_name + '"')
144
        for test_name, test_code in tests.get('match', dict()).items():
145
146
147
            if verbose:
                infostr = '    match-test "' + test_name + '" ... '
                errflag = len(errata)
148
            cst = parser(test_code, parser_name)
149
150
            if is_logging():
                cst.log("match_%s_%s.cst" % (parser_name, test_name))
151
                parser.log_parsing_history__("match_%s_%s.log" % (parser_name, test_name))
152
            tests.setdefault('__cst__', {})[test_name] = cst
153
            if "ast" in tests or report:
154
155
156
                ast = copy.deepcopy(cst)
                transform(ast)
                tests.setdefault('__ast__', {})[test_name] = ast
157
158
                if is_logging():
                    ast.log("match_%s_%s.ast" % (parser_name, test_name))
159
            if cst.error_flag:
160
                errata.append('Match test "%s" for parser "%s" failed:\n\tExpr.:  %s\n\n\t%s' %
161
                              (test_name, parser_name, '\n\t'.join(test_code.split('\n')),
162
163
                               '\n\t'.join(m.replace('\n', '\n\t\t') for m in
                                           error_messages(test_code, cst.collect_errors()))))
164
                tests.setdefault('__err__', {})[test_name] = errata[-1]
165
166
            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' %
167
                                  (test_name, parser_name, cst.as_sxpr()))
168
169
170
171
172
173
            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')),
Eckhart Arnold's avatar
Eckhart Arnold committed
174
175
                                     oneliner_sxpr(compare.as_sxpr()),
                                     oneliner_sxpr(ast.as_sxpr())))
176
                    tests.setdefault('__err__', {})[test_name] = errata[-1]
177
            if verbose:
178
                print(infostr + ("OK" if len(errata) == errflag else "FAIL"))
179

180
        if verbose and 'fail' in tests:
181
            print('  Fail-Tests for parser "' + parser_name + '"')
182
        for test_name, test_code in tests.get('fail', dict()).items():
183
184
185
            if verbose:
                infostr = '    fail-test  "' + test_name + '" ... '
                errflag = len(errata)
186
            cst = parser(test_code, parser_name)
187
188
            if is_logging():
                cst.log("fail_%s_%s.cst" % (parser_name, test_name))
189
                parser.log_parsing_history__("fail_%s_%s.log" % (parser_name, test_name))
190
191
192
            if not cst.error_flag:
                errata.append('Fail test "%s" for parser "%s" yields match instead of '
                              'expected failure!' % (test_name, parser_name))
193
                tests.setdefault('__err__', {})[test_name] = errata[-1]
194
195
            if verbose:
                print(infostr + "OK" if len(errata) == errflag else "FAIL")
196

197
198
    # write test-report
    if report:
199
200
201
202
        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:
203
            f.write(get_report(test_unit))
204

205
206
207
    return errata


208
def grammar_suite(directory, parser_factory, transformer_factory, ignore_unknown_filetypes=False,
209
                  report=True, verbose=False):
210
211
    """
    Runs all grammar unit tests in a directory. A file is considered a test
212
213
214
    unit, if it has the word "test" in its name.
    """
    all_errors = collections.OrderedDict()
215
216
    if verbose:
        print("\nScanning test-directory: " + directory)
217
218
219
    for filename in sorted(os.listdir(directory)):
        if filename.lower().find("test") >= 0:
            try:
220
221
                if verbose:
                    print("\nRunning grammar tests from: " + filename)
222
                errata = grammar_unit(os.path.join(directory, filename),
223
                                      parser_factory, transformer_factory, report, verbose)
224
225
226
                if errata:
                    all_errors[filename] = errata
            except ValueError as e:
227
                if not ignore_unknown_filetypes or str(e).find("Unknown") < 0:
228
229
230
231
232
233
234
235
                    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:
236
        return ('Test suite "%s" revealed some errors:\n\n' % directory) + '\n'.join(error_report)
237
238
239
    return ''


240
def runner(tests, namespace):
241
242
    """
    Runs all or some selected Python unit tests found in the
243
244
    namespace. To run all tests in a module, call 
    ``runner("", globals())`` from within that module.
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263

    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
264
            from DHParser.testing import runner
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
            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()