gecko-dev/testing/runtimes/writeruntimes

#!/bin/sh
# -*- Mode: python; indent-tabs-mode: nil; tab-width: 40 -*-
# vim: set filetype=python:

# This Source Code Form is subject to the terms of the Mozilla Public
# License, v. 2.0. If a copy of the MPL was not distributed with this
# file, You can obtain one at http://mozilla.org/MPL/2.0/.

# The beginning of this script is both valid shell and valid python,
# such that the script starts with the shell and is reexecuted python
''':'
which mach > /dev/null 2>&1 && exec mach python "$0" "$@" ||
echo "mach not found, either add it to your \$PATH or run this script via ./mach python testing/runtimes/writeruntimes"; exit  # noqa
'''

from __future__ import absolute_import, print_function

import datetime
import json
import os
import sys
import time
from argparse import ArgumentParser
from collections import defaultdict

import requests

from moztest.resolve import (
    TestManifestLoader,
    TestResolver,
    TEST_SUITES,
)

here = os.path.abspath(os.path.dirname(__file__))
ACTIVE_DATA_URL = "https://activedata.allizom.org/query"
EXCEED_LIMIT = [
    # Suites that exceed 10,000 ActiveData result limit will be defined here.
    'web-platform-tests',
    'web-platform-tests-reftest',
]
MAX_RETRIES = 10
RETRY_INTERVAL = 10


def construct_query(suite, platform):
    if platform in ('windows', 'android'):
        platform_clause = '{"find":{"run.machine.platform": "%s"}}' % platform
    else:
        # Bundle macosx and linux results together - they are not too different.
        platform_clause = '''
            {
                "not": {
                    "or": [
                        {"find":{"run.machine.platform": "windows"}},
                        {"find":{"run.machine.platform": "android"}}
                    ]
                }
            }
        '''

    # Only use this if the suite being queried exceeeds 10,000 results.
    output_clause = '"destination": "url",\n"format": "list",' if suite in EXCEED_LIMIT else ''

    query = """
{
    "from":"unittest",
    "limit":200000,
    "groupby":["result.test"],
    "select":{"value":"result.duration","aggregate":"median"},
    %s
    "where":{"and":[
        {"eq":{"repo.branch.name": "mozilla-central"}},
        {"in":{"result.status": ["OK", "PASS", "FAIL"]}},
        {"gt":{"run.timestamp": {"date": "today-week"}}},
        {"eq":{"run.suite.fullname":"%s"}},
        %s
    ]}
}
""" % (output_clause, suite, platform_clause)

    return query


def query_activedata(suite, platform):
    query = construct_query(suite, platform)
    print("Querying ActiveData for '{}' tests on '{}' platforms.. "
            .format(suite, platform), end='')
    sys.stdout.flush()
    response = requests.post(ACTIVE_DATA_URL,
                             data=query,
                             stream=True)
    response.raise_for_status()

    # Presence of destination clause in the query requires additional processing
    # to produce the dataset that can be used.
    if suite in EXCEED_LIMIT:
        # The output_url is where result of the query will be stored.
        output_url = response.json()["url"]

        tried = 0
        while tried < MAX_RETRIES:
            # Use the requests.Session object to manage requests, since the output_url
            # can often return 403 Forbidden.
            session = requests.Session()
            response = session.get(output_url)
            if response.status_code == 200:
                break
            # A non-200 status code means we should retry after some wait.
            time.sleep(RETRY_INTERVAL)
            tried += 1

        # Data returned from destination is in format of:
        # {data: [result: {test: test_name, duration: duration}]}
        # Normalize it to the format expected by compute_manifest_runtimes.
        raw_data = response.json()["data"]
        data = dict([[item['result']['test'], item['result']['duration']] for item in raw_data])
    else:
        data = dict(response.json()["data"])

    print("{} found".format(len(data)))
    return data


def write_runtimes(manifest_runtimes, platform, suite, outdir=here):
    if not os.path.exists(outdir):
        os.makedirs(outdir)

    outfilename = os.path.join(outdir, "manifest-runtimes-{}.json".format(platform))
    # If file is not present, initialize a file with empty JSON object.
    if not os.path.exists(outfilename):
        with open(outfilename, 'w+') as f:
            json.dump({}, f)

    # Load the entire file.
    with open(outfilename, 'r') as f:
        data = json.load(f)

    # Update the specific suite with the new runtime information and write to file.
    data[suite] = manifest_runtimes
    with open(outfilename, 'w') as f:
        json.dump(data, f, indent=2, sort_keys=True)


def compute_manifest_runtimes(suite, platform):
    resolver = TestResolver.from_environment(cwd=here, loader_cls=TestManifestLoader)

    crashtest_prefixes = {
        'http': '/tests/',
        'chrome': '/reftest/content/',
        'file': '/reftest/tests/',
    }
    manifest_runtimes = defaultdict(float)
    data = query_activedata(suite, platform)

    if "web-platform-tests" in suite:
        wpt_groups = {t["name"]: t["manifest"]
                        for t in resolver.resolve_tests(flavor="web-platform-tests")}

    for path, duration in data.items():
        # Returned data did not contain a test path, so go to next result.
        if not path:
            continue

        if suite in ('reftest', 'crashtest') and ' ' in path:
            path = path.split()[0]

        if suite == 'crashtest' and '://' in path:
            # Crashtest paths are URLs with various schemes and prefixes.
            # Normalize it to become relative to mozilla-central.
            scheme = path[:path.index('://')]
            if ':' in scheme:
                scheme = scheme.split(':')[-1]
            prefix = crashtest_prefixes[scheme]
            path = path.split(prefix, 1)[-1]
        elif suite == 'xpcshell' and ':' in path:
            path = path.split(':', 1)[-1]

        if "web-platform-tests" in suite:
            if path in wpt_groups:
                manifest_runtimes[wpt_groups[path]] += duration
            continue

        if path not in resolver.tests_by_path:
            continue

        for test in resolver.tests_by_path[path]:
            manifest = test.get('ancestor_manifest') or test['manifest_relpath']
            manifest_runtimes[manifest] += duration

    manifest_runtimes = {k: round(v, 2) for k, v in manifest_runtimes.items()}
    return manifest_runtimes


def cli(args=sys.argv[1:]):
    default_suites = [suite for suite, obj in TEST_SUITES.items() if 'build_flavor' in obj]
    default_platforms = ['android', 'windows', 'unix']

    parser = ArgumentParser()
    parser.add_argument('-o', '--output-directory', dest='outdir', default=here,
                        help="Directory to save runtime data.")
    parser.add_argument('-s', '--suite', dest='suites', action='append',
                        default=None, choices=default_suites,
                        help="Suite(s) to include in the data set (default: all)")
    parser.add_argument('-p', '--platform', dest='platforms', action='append',
                        default=None, choices=default_platforms,
                        help="Platform(s) to gather runtime information on "
                             "(default: all).")
    args = parser.parse_args(args)

    # If a suite was specified, use that. Otherwise, use the full default set.
    suites = args.suites or default_suites
    # Same as above, but for the platform clause.
    platforms = args.platforms or default_platforms

    for platform in platforms:
        for suite in suites:
            runtimes = compute_manifest_runtimes(suite, platform)
            if not runtimes:
                print("Not writing runtime data for '{}' for '{}' as no data was found".format(suite, platform))
                continue

            write_runtimes(runtimes, platform, suite, outdir=args.outdir)


if __name__ == "__main__":
    sys.exit(cli())