testing.py 12 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

Eckhart Arnold's avatar
Eckhart Arnold committed
29
from DHParser.toolkit import is_logging, clear_logs, error_messages
Eckhart Arnold's avatar
Eckhart Arnold committed
30
from DHParser.syntaxtree import mock_syntax_tree, flatten_sxpr
31

32
__all__ = ('unit_from_configfile',
33
34
35
36
37
38
39
           'unit_from_json',
           'unit_from_file',
           'get_report',
           'grammar_unit',
           'grammar_suite',
           'runner')

40
UNIT_STAGES = {'match', 'fail', 'ast', 'cst', '__ast__', '__cst__'}
41
42


43
def unit_from_configfile(config_filename):
44
45
    """
    Reads a grammar unit test from a config file.
46
    """
47
    cfg = configparser.ConfigParser(interpolation=None)
di68kap's avatar
di68kap committed
48
    cfg.read(config_filename, encoding="utf8")
49
50
    OD = collections.OrderedDict
    unit = OD()
51
52
    for section in cfg.sections():
        symbol, stage = section.split(':')
53
54
        if stage not in UNIT_STAGES:
            if symbol in UNIT_STAGES:
55
56
                symbol, stage = stage, symbol
            else:
57
                raise ValueError('Test stage %s not in: ' % (stage, str(UNIT_STAGES)))
58
        for testkey, testcode in cfg[section].items():
59
60
            if testcode[:3] + testcode[-3:] in {"''''''", '""""""'}:
                testcode = testcode[3:-3]
di68kap's avatar
di68kap committed
61
62
                # testcode = testcode.replace('\\#', '#')
                testcode = re.sub(r'(?<!\\)\\#', '#', testcode).replace('\\\\', '\\')
63
64
            elif testcode[:1] + testcode[-1:] in {"''", '""'}:
                testcode = testcode[1:-1]
65
            unit.setdefault(symbol, OD()).setdefault(stage, OD())[testkey] = testcode
66
67
    # print(json.dumps(unit, sort_keys=True, indent=4))
    return unit
68

69

70
def unit_from_json(json_filename):
71
72
    """
    Reads a grammar unit test from a json file.
73
    """
di68kap's avatar
di68kap committed
74
    with open(json_filename, 'r', encoding='utf8') as f:
75
76
77
78
79
80
81
        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

82
# TODO: add support for yaml, cson, toml
83
84


85
def unit_from_file(filename):
86
87
    """
    Reads a grammar unit test from a file. The format of the file is
88
89
    determined by the ending of its name.
    """
90
91
92
93
    if filename.endswith(".json"):
        return unit_from_json(filename)
    elif filename.endswith(".ini"):
        return unit_from_configfile(filename)
94
    else:
95
        raise ValueError("Unknown unit test file type: " + filename[filename.rfind('.'):])
96
97


98
def get_report(test_unit):
99
100
    """
    Returns a text-report of the results of a grammar unit test.
101
    """
102
103
104
105
    def indent(txt):
        lines = txt.split('\n')
        lines[0] = '    ' + lines[0]
        return "\n    ".join(lines)
106
107
108
109
110
111
112
113
    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:')
114
            report.append(indent(test_code))
115
116
117
118
119
            error = tests.get('__err__', {}).get(test_name, "")
            if error:
                report.append('\n### Error:')
                report.append(error)
            ast = tests.get('__ast__', {}).get(test_name, None)
120
121
122
            cst = tests.get('__cst__', {}).get(test_name, None)
            if cst and (not ast or cst == ast):
                report.append('\n### CST')
123
                report.append(indent(cst.as_sxpr()))
124
            elif ast:
125
                report.append('\n### AST')
126
                report.append(indent(ast.as_sxpr()))
127
128
129
    return '\n'.join(report)


130
def grammar_unit(test_unit, parser_factory, transformer_factory, report=True, verbose=False):
131
132
    """
    Unit tests for a grammar-parser and ast transformations.
133
    """
134
    if isinstance(test_unit, str):
135
        unit_dir, unit_name = os.path.split(os.path.splitext(test_unit)[0])
136
        test_unit = unit_from_file(test_unit)
137
    else:
138
        unit_dir = ""
139
        unit_name = str(id(test_unit))
140
141
    if verbose:
        print("\nUnit: " + unit_name)
142
143
144
    errata = []
    parser = parser_factory()
    transform = transformer_factory()
145
146
    for parser_name, tests in test_unit.items():
        assert set(tests.keys()).issubset(UNIT_STAGES)
147
148
        if verbose:
            print('  Match-Tests for parser "' + parser_name + '"')
149
        for test_name, test_code in tests.get('match', dict()).items():
150
151
152
            if verbose:
                infostr = '    match-test "' + test_name + '" ... '
                errflag = len(errata)
153
            cst = parser(test_code, parser_name)
Eckhart Arnold's avatar
Eckhart Arnold committed
154
            cst.log("match_%s_%s.cst" % (parser_name, test_name))
155
            tests.setdefault('__cst__', {})[test_name] = cst
156
            if "ast" in tests or report:
157
158
159
                ast = copy.deepcopy(cst)
                transform(ast)
                tests.setdefault('__ast__', {})[test_name] = ast
Eckhart Arnold's avatar
Eckhart Arnold committed
160
                ast.log("match_%s_%s.ast" % (parser_name, test_name))
161
            if cst.error_flag:
Eckhart Arnold's avatar
Eckhart Arnold committed
162
                errata.append('Match test "%s" for parser "%s" failed:\n\tExpr.:  %s\n\n\t%s\n\n' %
163
                              (test_name, parser_name, '\n\t'.join(test_code.split('\n')),
164
165
                               '\n\t'.join(m.replace('\n', '\n\t\t') for m in
                                           error_messages(test_code, cst.collect_errors()))))
166
                tests.setdefault('__err__', {})[test_name] = errata[-1]
167
168
                # write parsing-history log only in case of failure!
                parser.log_parsing_history__("match_%s_%s.log" % (parser_name, test_name))
169
170
            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' %
171
                                  (test_name, parser_name, cst.as_sxpr()))
172
173
174
175
176
177
            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
178
179
                                     flatten_sxpr(compare.as_sxpr()),
                                     flatten_sxpr(ast.as_sxpr())))
180
                    tests.setdefault('__err__', {})[test_name] = errata[-1]
181
            if verbose:
182
                print(infostr + ("OK" if len(errata) == errflag else "FAIL"))
183

184
        if verbose and 'fail' in tests:
185
            print('  Fail-Tests for parser "' + parser_name + '"')
186
        for test_name, test_code in tests.get('fail', dict()).items():
187
188
189
            if verbose:
                infostr = '    fail-test  "' + test_name + '" ... '
                errflag = len(errata)
190
191
192
193
            cst = parser(test_code, parser_name)
            if not cst.error_flag:
                errata.append('Fail test "%s" for parser "%s" yields match instead of '
                              'expected failure!' % (test_name, parser_name))
194
                tests.setdefault('__err__', {})[test_name] = errata[-1]
195
196
                # write parsing-history log only in case of test-failure
                parser.log_parsing_history__("fail_%s_%s.log" % (parser_name, test_name))
197
198
            if verbose:
                print(infostr + "OK" if len(errata) == errflag else "FAIL")
199

200
201
    # write test-report
    if report:
202
        report_dir = "REPORT"
203
204
        if not os.path.exists(report_dir):
            os.mkdir(report_dir)
di68kap's avatar
di68kap committed
205
        with open(os.path.join(report_dir, unit_name + '.md'), 'w', encoding='utf8') as f:
206
            f.write(get_report(test_unit))
207

208
209
210
    return errata


211
def grammar_suite(directory, parser_factory, transformer_factory, ignore_unknown_filetypes=False,
212
                  report=True, verbose=False):
213
214
    """
    Runs all grammar unit tests in a directory. A file is considered a test
215
216
217
    unit, if it has the word "test" in its name.
    """
    all_errors = collections.OrderedDict()
218
219
    if verbose:
        print("\nScanning test-directory: " + directory)
220
221
    save_cwd = os.getcwd()
    os.chdir(directory)
Eckhart Arnold's avatar
Eckhart Arnold committed
222
    if is_logging():    clear_logs()
223
    for filename in sorted(os.listdir()):
224
225
        if filename.lower().find("test") >= 0:
            try:
226
227
                if verbose:
                    print("\nRunning grammar tests from: " + filename)
228
229
                errata = grammar_unit(filename, parser_factory,
                                      transformer_factory, report, verbose)
230
231
232
                if errata:
                    all_errors[filename] = errata
            except ValueError as e:
233
                if not ignore_unknown_filetypes or str(e).find("Unknown") < 0:
234
                    raise e
235
    os.chdir(save_cwd)
236
237
238
239
240
241
242
    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:
243
        return ('Test suite "%s" revealed some errors:\n\n' % directory) + '\n'.join(error_report)
244
245
246
    return ''


247
def runner(tests, namespace):
248
249
    """
    Runs all or some selected Python unit tests found in the
250
251
    namespace. To run all tests in a module, call 
    ``runner("", globals())`` from within that module.
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270

    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
271
            from DHParser.testing import runner
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
301
302
303
304
305
306
307
            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()