summaryrefslogtreecommitdiffstats
path: root/scripts/lib/resulttool/report.py
blob: 90086209e359a729b9a7267176a48eddb3d35d06 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
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
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
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
# test result tool - report text based test results
#
# Copyright (c) 2019, Intel Corporation.
# Copyright (c) 2019, Linux Foundation
#
# This program is free software; you can redistribute it and/or modify it
# under the terms and conditions of the GNU General Public License,
# version 2, as published by the Free Software Foundation.
#
# This program is distributed in the hope it will be useful, but WITHOUT
# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
# FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
# more details.
#
import os
import glob
import json
import resulttool.resultutils as resultutils
from oeqa.utils.git import GitRepo
import oeqa.utils.gitarchive as gitarchive


class ResultsTextReport(object):
    def __init__(self):
        self.ptests = {}
        self.result_types = {'passed': ['PASSED', 'passed'],
                             'failed': ['FAILED', 'failed', 'ERROR', 'error', 'UNKNOWN'],
                             'skipped': ['SKIPPED', 'skipped']}


    def handle_ptest_result(self, k, status, result):
        if k == 'ptestresult.sections':
            # Ensure tests without any test results still show up on the report
            for suite in result['ptestresult.sections']:
                if suite not in self.ptests:
                    self.ptests[suite] = {'passed': 0, 'failed': 0, 'skipped': 0, 'duration' : '-', 'failed_testcases': []}
                if 'duration' in result['ptestresult.sections'][suite]:
                    self.ptests[suite]['duration'] = result['ptestresult.sections'][suite]['duration']
                if 'timeout' in result['ptestresult.sections'][suite]:
                    self.ptests[suite]['duration'] += " T"
            return
        try:
            _, suite, test = k.split(".", 2)
        except ValueError:
            return
        # Handle 'glib-2.0'
        if 'ptestresult.sections' in result and suite not in result['ptestresult.sections']:
            try:
                _, suite, suite1, test = k.split(".", 3)
                if suite + "." + suite1 in result['ptestresult.sections']:
                    suite = suite + "." + suite1
            except ValueError:
                pass
        if suite not in self.ptests:
            self.ptests[suite] = {'passed': 0, 'failed': 0, 'skipped': 0, 'duration' : '-', 'failed_testcases': []}
        for tk in self.result_types:
            if status in self.result_types[tk]:
                self.ptests[suite][tk] += 1

    def get_aggregated_test_result(self, logger, testresult):
        test_count_report = {'passed': 0, 'failed': 0, 'skipped': 0, 'failed_testcases': []}
        result = testresult.get('result', [])
        for k in result:
            test_status = result[k].get('status', [])
            for tk in self.result_types:
                if test_status in self.result_types[tk]:
                    test_count_report[tk] += 1
            if test_status in self.result_types['failed']:
                test_count_report['failed_testcases'].append(k)
            if k.startswith("ptestresult."):
                self.handle_ptest_result(k, test_status, result)
        return test_count_report

    def print_test_report(self, template_file_name, test_count_reports):
        from jinja2 import Environment, FileSystemLoader
        script_path = os.path.dirname(os.path.realpath(__file__))
        file_loader = FileSystemLoader(script_path + '/template')
        env = Environment(loader=file_loader, trim_blocks=True)
        template = env.get_template(template_file_name)
        havefailed = False
        haveptest = bool(self.ptests)
        reportvalues = []
        cols = ['passed', 'failed', 'skipped']
        maxlen = {'passed' : 0, 'failed' : 0, 'skipped' : 0, 'result_id': 0, 'testseries' : 0, 'ptest' : 0 }
        for line in test_count_reports:
            total_tested = line['passed'] + line['failed'] + line['skipped']
            vals = {}
            vals['result_id'] = line['result_id']
            vals['testseries'] = line['testseries']
            vals['sort'] = line['testseries'] + "_" + line['result_id']
            vals['failed_testcases'] = line['failed_testcases']
            for k in cols:
                vals[k] = "%d (%s%%)" % (line[k], format(line[k] / total_tested * 100, '.0f'))
            for k in maxlen:
                if k in vals and len(vals[k]) > maxlen[k]:
                    maxlen[k] = len(vals[k])
            reportvalues.append(vals)
            if line['failed_testcases']:
                havefailed = True
        for ptest in self.ptests:
            if len(ptest) > maxlen['ptest']:
                maxlen['ptest'] = len(ptest)
        output = template.render(reportvalues=reportvalues,
                                 havefailed=havefailed,
                                 haveptest=haveptest,
                                 ptests=self.ptests,
                                 maxlen=maxlen)
        print(output)

    def view_test_report(self, logger, source_dir, branch, commit, tag):
        test_count_reports = []
        if commit:
            if tag:
                logger.warning("Ignoring --tag as --commit was specified")
            tag_name = "{branch}/{commit_number}-g{commit}/{tag_number}"
            repo = GitRepo(source_dir)
            revs = gitarchive.get_test_revs(logger, repo, tag_name, branch=branch)
            rev_index = gitarchive.rev_find(revs, 'commit', commit)
            testresults = resultutils.git_get_result(repo, revs[rev_index][2])
        elif tag:
            repo = GitRepo(source_dir)
            testresults = resultutils.git_get_result(repo, [tag])
        else:
            testresults = resultutils.load_resultsdata(source_dir)
        for testsuite in testresults:
            for resultid in testresults[testsuite]:
                result = testresults[testsuite][resultid]
                test_count_report = self.get_aggregated_test_result(logger, result)
                test_count_report['testseries'] = result['configuration']['TESTSERIES']
                test_count_report['result_id'] = resultid
                test_count_reports.append(test_count_report)
        self.print_test_report('test_report_full_text.txt', test_count_reports)

def report(args, logger):
    report = ResultsTextReport()
    report.view_test_report(logger, args.source_dir, args.branch, args.commit, args.tag)
    return 0

def register_commands(subparsers):
    """Register subcommands from this plugin"""
    parser_build = subparsers.add_parser('report', help='summarise test results',
                                         description='print a text-based summary of the test results',
                                         group='analysis')
    parser_build.set_defaults(func=report)
    parser_build.add_argument('source_dir',
                              help='source file/directory that contain the test result files to summarise')
    parser_build.add_argument('--branch', '-B', default='master', help="Branch to find commit in")
    parser_build.add_argument('--commit', help="Revision to report")
    parser_build.add_argument('-t', '--tag', default='',
                              help='source_dir is a git repository, report on the tag specified from that repository')