bugbug/scripts/test_scheduling_history_retriever.py

# -*- coding: utf-8 -*-

import argparse
import json
import os
import subprocess
from datetime import datetime
from logging import INFO, basicConfig, getLogger

import dateutil.parser
from dateutil.relativedelta import relativedelta
from tqdm import tqdm

from bugbug import commit_features, db, repository, test_scheduling
from bugbug.utils import (
    download_check_etag,
    open_tar_zst,
    zstd_compress,
    zstd_decompress,
)

basicConfig(level=INFO)
logger = getLogger(__name__)

JOBS_TO_CONSIDER = ("test-", "build-")
JOBS_TO_IGNORE = ("build-docker-image-",)

ADR_CACHE_DB = "data/adr_cache.tar"
db.register(
    ADR_CACHE_DB,
    "https://community-tc.services.mozilla.com/api/index/v1/task/project.relman.bugbug.data_test_scheduling_history_push_data.latest/artifacts/public/adr_cache.tar.zst",
    2,
    support_files=[],
)
PUSH_DATA_URL = "https://community-tc.services.mozilla.com/api/index/v1/task/project.relman.bugbug.data_test_scheduling_history_push_data.latest/artifacts/public/push_data.json.zst"

TRAINING_MONTHS = 6


def filter_tasks(tasks, all_tasks):
    return tuple(
        task
        for task in tasks
        if task in all_tasks
        and any(task.startswith(j) for j in JOBS_TO_CONSIDER)
        and not any(task.startswith(j) for j in JOBS_TO_IGNORE)
    )


class Retriever(object):
    def __init__(self):
        os.makedirs("data", exist_ok=True)

    def retrieve_push_data(self):
        # Download previous cache.
        cache_path = os.path.splitext(ADR_CACHE_DB)[0]
        db.download(ADR_CACHE_DB)

        # Setup adr cache configuration.
        os.makedirs(os.path.expanduser("~/.config/adr"), exist_ok=True)
        with open(os.path.expanduser("~/.config/adr/config.toml"), "w") as f:
            f.write(
                f"""[adr.cache.stores]
file = {{ driver = "file", path = "{os.path.abspath(cache_path)}" }}
"""
            )

        # We'll use the past TRAINING_MONTHS months only for training the model,
        # but we use 3 months more than that to calculate the failure statistics.
        subprocess.run(
            [
                "run-adr",
                "ahal/ci-recipes",
                "recipe",
                "-o",
                os.path.abspath("push_data.json"),
                "-f",
                "json",
                "push_data",
                "--",
                "--from",
                f"today-{TRAINING_MONTHS + 3}month",
                "--to",
                "today-3day",
                "--branch",
                "autoland",
            ],
            check=True,
            stdout=subprocess.DEVNULL,  # Redirect to /dev/null, as the logs are too big otherwise.
        )

        with open_tar_zst(f"{ADR_CACHE_DB}.zst") as tar:
            tar.add(cache_path)

        zstd_compress("push_data.json")

    def generate_test_scheduling_history(self):
        if not os.path.exists("push_data.json"):
            download_check_etag(PUSH_DATA_URL, "push_data.json.zst")
            zstd_decompress("push_data.json")
            assert os.path.exists(
                "push_data.json"
            ), "Decompressed push data file exists"

        # Get the commits DB.
        assert db.download(repository.COMMITS_DB)

        HISTORY_DATE_START = datetime.now() - relativedelta(months=TRAINING_MONTHS)

        db.download(test_scheduling.TEST_SCHEDULING_DB, support_files_too=True)

        last_node = None
        for test_data in test_scheduling.get_test_scheduling_history():
            last_node = test_data["revs"][0]

        def generate_all_data():
            past_failures = test_scheduling.get_past_failures()

            push_num = past_failures["push_num"] if "push_num" in past_failures else 0

            # We can start once we get to the last revision we added in the previous run.
            can_start = True if last_node is None else False

            commit_map = {}
            for commit_data in tqdm(repository.get_commits()):
                if not can_start:
                    if last_node == commit_data["node"]:
                        can_start = True

                    continue

                commit_map[commit_data["node"]] = commit_data

            with open("push_data.json", "r") as f:
                push_data = json.load(f)[1:]

            logger.info(f"push data nodes: {len(push_data)}")

            # In the last 28 pushes, we definitely run all possible tasks.
            all_tasks_set = set(
                sum((push_tasks for _, push_tasks, _, _ in push_data[-28:]), [])
            )
            # Filter tasks we don't need.
            all_tasks = filter_tasks(list(all_tasks_set), all_tasks_set)
            all_tasks_set = set(all_tasks)
            logger.info(f"{len(all_tasks_set)} tasks run in the last 28 pushes")

            # Store all tasks in the past_failures DB so it can be used in the evaluation phase.
            past_failures["all_tasks"] = all_tasks
            # XXX: Should we recreate the DB from scratch if the previous all_tasks are not the
            # same as the current ones?

            saved_nodes = set()
            skipped_no_commits = 0
            skipped_too_big_commits = 0
            skipped_no_tasks = 0

            # We can start once we get to the last revision we added in the previous run.
            can_start = True if last_node is None else False

            for i in tqdm(range(len(push_data))):
                (
                    revisions,
                    push_tasks,
                    possible_regressions,
                    likely_regressions,
                ) = push_data.pop(0)

                if not can_start:
                    if last_node == revisions[0]:
                        can_start = True

                    continue

                push_num += 1

                # XXX: Some commits are skipped in the repository mining, e.g. merges and backouts. Maybe we should not skip them.
                commits = tuple(
                    commit_map.pop(revision)
                    for revision in revisions
                    if revision in commit_map
                )
                if len(commits) == 0:
                    skipped_no_commits += 1
                    continue

                merged_commits = commit_features.merge_commits(commits)

                # XXX: For now, skip commits which are too large.
                # In the future we can either:
                #  - Improve shelve perf and go back to consider all files;
                #  - Consider only files which appear with a given frequency, like the "files" feature in commit_features;
                #  - Keep a limit of number of files.
                if len(merged_commits["files"]) > 20:
                    skipped_too_big_commits += 1
                    continue

                # If we considered all_tasks, we'd generate a huge amount of data.
                # So we consider only the tasks which run in this push, and the possible and likely regressions
                # from this push.
                tasks_to_consider = list(
                    set(push_tasks + possible_regressions + likely_regressions)
                )
                tasks_to_consider = filter_tasks(tasks_to_consider, all_tasks_set)

                if len(tasks_to_consider) == 0:
                    skipped_no_tasks += 1
                    continue

                # Sync DB every 250 pushes, so we cleanup the shelve cache (we'd run OOM otherwise!).
                if i % 250 == 0:
                    past_failures.sync()

                pushdate = dateutil.parser.parse(merged_commits["pushdate"])

                for data in test_scheduling.generate_data(
                    past_failures,
                    merged_commits,
                    push_num,
                    tasks_to_consider,
                    possible_regressions,
                    likely_regressions,
                ):
                    if pushdate > HISTORY_DATE_START:
                        saved_nodes.add(i)
                        data["revisions"] = revisions
                        yield data

            logger.info(f"saved push data nodes: {len(saved_nodes)}")
            logger.info(f"skipped {skipped_no_commits} (no commits in our DB)")
            logger.info(f"skipped {skipped_too_big_commits} (too big commits)")
            logger.info(f"skipped {skipped_no_tasks} (no interesting tasks)")

            past_failures["push_num"] = push_num
            past_failures.close()

        db.append(test_scheduling.TEST_SCHEDULING_DB, generate_all_data())

        zstd_compress(test_scheduling.TEST_SCHEDULING_DB)

        with open_tar_zst("data/past_failures.lmdb.tar.zst") as tar:
            tar.add("data/past_failures.lmdb")


def main():
    description = "Retrieve and extract the test scheduling history from ActiveData"
    parser = argparse.ArgumentParser(description=description)

    parser.add_argument(
        "op", help="Which operation to perform.", choices=["retrieve", "generate"]
    )

    args = parser.parse_args()

    retriever = Retriever()
    if args.op == "retrieve":
        retriever.retrieve_push_data()
    elif args.op == "generate":
        retriever.generate_test_scheduling_history()


if __name__ == "__main__":
    main()