path: root/taskcluster/gecko_taskgraph/transforms/source_test.py

# 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/.
"""
Source-test jobs can run on multiple platforms.  These transforms allow jobs
with either `platform` or a list of `platforms`, and set the appropriate
treeherder configuration and attributes for that platform.
"""


import copy
import os

import taskgraph
from taskgraph.transforms.base import TransformSequence
from taskgraph.util.attributes import keymatch
from taskgraph.util.schema import Schema, optionally_keyed_by, resolve_keyed_by
from taskgraph.util.treeherder import join_symbol, split_symbol
from voluptuous import Any, Extra, Optional, Required

from gecko_taskgraph.transforms.job import job_description_schema
from gecko_taskgraph.util.hg import get_json_automationrelevance

source_test_description_schema = Schema(
    {
        # most fields are passed directly through as job fields, and are not
        # repeated here
        Extra: object,
        # The platform on which this task runs.  This will be used to set up attributes
        # (for try selection) and treeherder metadata (for display).  If given as a list,
        # the job will be "split" into multiple tasks, one with each platform.
        Required("platform"): Any(str, [str]),
        # Build labels required for the task. If this key is provided it must
        # contain a build label for the task platform.
        # The task will then depend on a build task, and the installer url will be
        # saved to the GECKO_INSTALLER_URL environment variable.
        Optional("require-build"): optionally_keyed_by("project", {str: str}),
        # These fields can be keyed by "platform", and are otherwise identical to
        # job descriptions.
        Required("worker-type"): optionally_keyed_by(
            "platform", job_description_schema["worker-type"]
        ),
        Required("worker"): optionally_keyed_by(
            "platform", job_description_schema["worker"]
        ),
        Optional("python-version"): [int],
        Optional("dependencies"): {
            k: optionally_keyed_by("platform", v)
            for k, v in job_description_schema["dependencies"].items()
        },
        # A list of artifacts to install from 'fetch' tasks.
        Optional("fetches"): {
            str: optionally_keyed_by(
                "platform", job_description_schema["fetches"][str]
            ),
        },
    }
)

transforms = TransformSequence()

transforms.add_validate(source_test_description_schema)


@transforms.add
def set_job_name(config, jobs):
    for job in jobs:
        if "job-from" in job and job["job-from"] != "kind.yml":
            from_name = os.path.splitext(job["job-from"])[0]
            job["name"] = "{}-{}".format(from_name, job["name"])
        yield job


@transforms.add
def expand_platforms(config, jobs):
    for job in jobs:
        if isinstance(job["platform"], str):
            yield job
            continue

        for platform in job["platform"]:
            pjob = copy.deepcopy(job)
            pjob["platform"] = platform

            if "name" in pjob:
                pjob["name"] = "{}-{}".format(pjob["name"], platform)
            else:
                pjob["label"] = "{}-{}".format(pjob["label"], platform)
            yield pjob


@transforms.add
def split_python(config, jobs):
    for job in jobs:
        key = "python-version"
        versions = job.pop(key, [])
        if not versions:
            yield job
            continue
        for version in versions:
            group = f"py{version}"
            pyjob = copy.deepcopy(job)
            if "name" in pyjob:
                pyjob["name"] += f"-{group}"
            else:
                pyjob["label"] += f"-{group}"
            symbol = split_symbol(pyjob["treeherder"]["symbol"])[1]
            pyjob["treeherder"]["symbol"] = join_symbol(group, symbol)
            pyjob["run"][key] = version
            yield pyjob


@transforms.add
def split_jsshell(config, jobs):
    all_shells = {"sm": "Spidermonkey", "v8": "Google V8"}

    for job in jobs:
        if not job["name"].startswith("jsshell"):
            yield job
            continue

        test = job.pop("test")
        for shell in job.get("shell", all_shells.keys()):
            assert shell in all_shells

            new_job = copy.deepcopy(job)
            new_job["name"] = "{}-{}".format(new_job["name"], shell)
            new_job["description"] = "{} on {}".format(
                new_job["description"], all_shells[shell]
            )
            new_job["shell"] = shell

            group = f"js-bench-{shell}"
            symbol = split_symbol(new_job["treeherder"]["symbol"])[1]
            new_job["treeherder"]["symbol"] = join_symbol(group, symbol)

            run = new_job["run"]
            run["mach"] = run["mach"].format(
                shell=shell, SHELL=shell.upper(), test=test
            )
            yield new_job


def add_build_dependency(config, job):
    """
    Add build dependency to the job and installer_url to env.
    """
    key = job["platform"]
    build_labels = job.pop("require-build", {})
    matches = keymatch(build_labels, key)
    if not matches:
        raise Exception(
            "No build platform found. "
            "Define 'require-build' for {} in the task config.".format(key)
        )

    if len(matches) > 1:
        raise Exception(f"More than one build platform found for '{key}'.")

    label = matches[0]
    deps = job.setdefault("dependencies", {})
    deps.update({"build": label})


@transforms.add
def handle_platform(config, jobs):
    """
    Handle the 'platform' property, setting up treeherder context as well as
    try-related attributes.
    """
    fields = [
        "always-target",
        "fetches.toolchain",
        "require-build",
        "worker-type",
        "worker",
    ]

    for job in jobs:
        platform = job["platform"]

        for field in fields:
            resolve_keyed_by(
                job, field, item_name=job["name"], project=config.params["project"]
            )
        for field in job.get("dependencies", {}):
            resolve_keyed_by(
                job,
                f"dependencies.{field}",
                item_name=job["name"],
                project=config.params["project"],
            )

        if "treeherder" in job:
            job["treeherder"].setdefault("platform", platform)

        if "require-build" in job:
            add_build_dependency(config, job)

        del job["platform"]
        yield job


@transforms.add
def handle_shell(config, jobs):
    """
    Handle the 'shell' property.
    """
    fields = [
        "run-on-projects",
        "worker.env",
    ]

    for job in jobs:
        if not job.get("shell"):
            yield job
            continue

        for field in fields:
            resolve_keyed_by(job, field, item_name=job["name"])

        del job["shell"]
        yield job


@transforms.add
def set_code_review_env(config, jobs):
    """
    Add a CODE_REVIEW environment variable when running in code-review bot mode
    """
    is_code_review = config.params["target_tasks_method"] == "codereview"

    for job in jobs:
        attrs = job.get("attributes", {})
        if is_code_review and attrs.get("code-review") is True:
            env = job["worker"].setdefault("env", {})
            env["CODE_REVIEW"] = "1"

        yield job


@transforms.add
def set_base_revision_in_tgdiff(config, jobs):
    # Don't attempt to download 'json-automation' locally as the revision may
    # not exist in the repository.
    if not os.environ.get("MOZ_AUTOMATION") or taskgraph.fast:
        yield from jobs
        return

    data = get_json_automationrelevance(
        config.params["head_repository"], config.params["head_rev"]
    )
    for job in jobs:
        if job["name"] != "taskgraph-diff":
            yield job
            continue

        job["run"]["command-context"] = {
            "base_rev": data["changesets"][0]["parents"][0]
        }
        yield job


@transforms.add
def set_worker_exit_code(config, jobs):
    for job in jobs:
        worker = job["worker"]
        worker.setdefault("retry-exit-status", [])
        if 137 not in worker["retry-exit-status"]:
            worker["retry-exit-status"].append(137)
        yield job