python/run-tests.py - master - githubmirror/spark

python/run-tests.py
584 строки22 KB
Tian Gao
[SPARK-56186][PYTHON] Retire pypy
25 мар 2026, 01:48
25 мар 2026, 01:48d51b39b
#!/usr/bin/env python3

#
# Licensed to the Apache Software Foundation (ASF) under one or more
# contributor license agreements.  See the NOTICE file distributed with
# this work for additional information regarding copyright ownership.
# The ASF licenses this file to You 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.
#

import asyncio
import logging
from argparse import ArgumentParser
import os
import io
import platform
import pty
import re
import shutil
import subprocess
import sys
import tempfile
from threading import Thread, Lock
import time
import uuid
import queue as Queue
from multiprocessing import Manager


# Append `SPARK_HOME/dev` to the Python path so that we can import the sparktestsupport module
sys.path.append(os.path.join(os.path.dirname(os.path.realpath(__file__)), "../dev/"))


from sparktestsupport import SPARK_HOME
from sparktestsupport.shellutils import which, subprocess_check_output
from sparktestsupport.modules import all_modules, pyspark_sql  # noqa


python_modules = dict((m.name, m) for m in all_modules if m.python_test_goals if m.name != "root")


def print_red(text):
    print("\033[31m" + text + "\033[0m")


def get_valid_filename(s):
    """Replace whitespaces and special characters in the given string to get a valid file name."""
    s = s.strip().replace(" ", "_").replace(os.sep, "_")
    return re.sub(r"(?u)[^-\w.]", "", s)


SKIPPED_TESTS = None
LOG_FILE = os.path.join(SPARK_HOME, "python/unit-tests.log")
FAILURE_REPORTING_LOCK = Lock()
LOGGER = logging.getLogger()

# Find out where the assembly jars are located.
# TODO: revisit for Scala 2.13
SPARK_DIST_CLASSPATH = ""
if "SPARK_SKIP_CONNECT_COMPAT_TESTS" not in os.environ:
    for scala in ["2.13"]:
        build_dir = os.path.join(SPARK_HOME, "assembly", "target", "scala-" + scala)
        if os.path.isdir(build_dir):
            SPARK_DIST_CLASSPATH = os.path.join(build_dir, "jars", "*")
            break
    else:
        raise RuntimeError("Cannot find assembly build directory, please build Spark first.")


class TestRunner:
    def __init__(self, test_name, cmd, env, test_output, timeout=None):
        self.test_name = test_name
        self.cmd = cmd
        self.env = env
        self.test_output = test_output
        self.timeout = timeout
        self.p = None
        self.pdb_mode = False
        self.master_fd = None
        self.write_task = None
        self.read_task = None
        self.timeout_task = None

    def run(self):
        """
        Run a command in subprocess, with stdin, stdout, stderr hooked.
        In normaly case, all the outputs from subprocess will be redirected to
        the test_output file.
        When `(Pdb)` is detected, the subprocess will be in interactive mode,
        and the output will be redirected to the console.
        """
        self.master_fd, slave_fd = pty.openpty()

        # Start child connected to the PTY
        self.p = subprocess.Popen(
            self.cmd,
            env=self.env,
            stdin=slave_fd,
            stdout=slave_fd,
            stderr=slave_fd,
        )
        os.close(slave_fd)

        try:
            asyncio.run(self.handle_inout())
        except subprocess.TimeoutExpired:
            LOGGER.error(f"Test {self.test_name} timed out after {self.timeout} seconds")
        try:
            return self.p.wait(timeout=30)
        except subprocess.TimeoutExpired:
            # If SIGTERM is intercepted, do a hard kill
            self.p.kill()
            return self.p.wait()

    async def handle_inout(self):
        tasks = []
        self.read_task = asyncio.create_task(self.read_from_child())
        tasks.append(self.read_task)
        if self.timeout is not None:
            self.timeout_task = asyncio.create_task(self.check_timeout())
            tasks.append(self.timeout_task)
        try:
            await asyncio.gather(*tasks)
        except asyncio.CancelledError:
            pass

    def output_line(self, line):
        if self.pdb_mode:
            sys.stdout.write(line.decode("utf-8", "replace"))
            sys.stdout.flush()
        else:
            if isinstance(self.test_output, io.TextIOBase):
                self.test_output.write(line.decode("utf-8", "replace"))
            else:
                self.test_output.write(line)

    def process_buffer(self, buffer, force_flush=False):
        # Process all full lines first
        while (nl := buffer.find(b"\n")) != -1:
            self.output_line(buffer[: nl + 1])
            buffer = buffer[nl + 1 :]
        # Process the remaining buffer
        if b"(Pdb)" in buffer:
            self.pdb_mode = True
            self.output_line(buffer)
            return b""
        elif force_flush:
            self.output_line(buffer)
            return b""
        else:
            return buffer

    # Reader: forward child output to our stdout
    async def read_from_child(self):
        buffer = b""
        while True:
            try:
                data = await asyncio.to_thread(os.read, self.master_fd, 1024)
            except OSError:
                break
            if not data:
                break
            buffer += data
            buffer = self.process_buffer(buffer)
            if self.pdb_mode and self.write_task is None:
                self.write_task = asyncio.create_task(self.write_to_child())
        buffer = self.process_buffer(buffer, force_flush=True)
        self.test_output.flush()

        if self.write_task is not None:
            self.write_task.cancel()
            try:
                await self.write_task
            except asyncio.CancelledError:
                pass
            self.write_task = None
            self.pdb_mode = False

        if self.timeout_task is not None:
            self.timeout_task.cancel()
            self.timeout_task = None

    # Writer: forward our stdin to child tty
    async def write_to_child(self):
        while True:
            data = await self.loop.run_in_executor(None, sys.stdin.buffer.read, 1)
            if not data:
                break
            os.write(self.master_fd, data)

    # Kill the child process if the timeout is reached
    async def check_timeout(self):
        await asyncio.sleep(self.timeout)
        if self.pdb_mode:
            # We don't want to kill the process if it's in pdb mode
            return
        if self.p.poll() is None:
            if sys.platform == "linux":
                self.thread_dump(self.p.pid)
            self.p.terminate()
            raise subprocess.TimeoutExpired(self.cmd, self.timeout)

    def thread_dump(self, pid):
        pyspark_python = self.env["PYSPARK_PYTHON"]
        python_path = f"{os.path.join(SPARK_HOME, 'python')}"
        py4j_path = f"{os.path.join(SPARK_HOME, 'python/lib/py4j-0.10.9.9-src.zip')}"
        p = subprocess.run(
            [pyspark_python, "-m", "pyspark.threaddump", "-p", str(pid)],
            env={
                **self.env,
                "PYTHONPATH": f"{python_path}:{py4j_path}:{os.environ.get('PYTHONPATH', '')}",
            },
            stdout=subprocess.PIPE,
            stderr=subprocess.STDOUT,
        )
        if p.returncode == 0:
            LOGGER.error(f"Thread dump:\n{p.stdout.decode('utf-8')}")
        elif p.returncode == 5:
            # pystack or psutil not installed, that's okay
            pass
        else:
            LOGGER.error(
                f"Failed to get thread dump, exit code {p.returncode}:\n{p.stdout.decode('utf-8')}"
            )


def run_individual_python_test(target_dir, test_name, pyspark_python, keep_test_output):
    """
    Runs an individual test. This function is called by the multi-process runner of all tests.

    Parameters
    ----------
    target_dir
        Destination for the Hive and log directory.
    test_name
        Test name.
    pyspark_python
        Python version used to run the test.
    keep_test_output
        Flag indicating if the test output should be retained after successful execution.

    """
    env = dict(os.environ)
    env.update(
        {
            "SPARK_DIST_CLASSPATH": SPARK_DIST_CLASSPATH,
            "SPARK_TESTING": "1",
            "SPARK_PREPEND_CLASSES": "1",
            "PYSPARK_PYTHON": which(pyspark_python),
            "PYSPARK_DRIVER_PYTHON": which(pyspark_python),
        }
    )

    if "SPARK_CONNECT_TESTING_REMOTE" in os.environ:
        env.update({"SPARK_CONNECT_TESTING_REMOTE": os.environ["SPARK_CONNECT_TESTING_REMOTE"]})
    if "SPARK_SKIP_CONNECT_COMPAT_TESTS" in os.environ:
        env.update({"SPARK_SKIP_JVM_REQUIRED_TESTS": os.environ["SPARK_SKIP_CONNECT_COMPAT_TESTS"]})

    # Create a unique temp directory under 'target/' for each run. The TMPDIR variable is
    # recognized by the tempfile module to override the default system temp directory.
    tmp_dir = os.path.join(target_dir, str(uuid.uuid4()))
    while os.path.isdir(tmp_dir):
        tmp_dir = os.path.join(target_dir, str(uuid.uuid4()))
    os.mkdir(tmp_dir)
    sock_dir = os.getenv("TMPDIR") or os.getenv("TEMP") or os.getenv("TMP") or "/tmp"
    env["TMPDIR"] = tmp_dir
    metastore_dir = os.path.join(tmp_dir, str(uuid.uuid4()))
    while os.path.isdir(metastore_dir):
        metastore_dir = os.path.join(metastore_dir, str(uuid.uuid4()))
    os.mkdir(metastore_dir)

    # Also override the JVM's temp directory and log4j conf by setting driver and executor options.
    log4j2_path = os.path.join(SPARK_HOME, "python/test_support/log4j2.properties")
    java_options = "-Djava.io.tmpdir={0} -Dlog4j.configurationFile={1}".format(tmp_dir, log4j2_path)
    java_options = java_options + " -Xss4M"
    spark_args = [
        "--conf",
        "spark.driver.extraJavaOptions='{0}'".format(java_options),
        "--conf",
        "spark.executor.extraJavaOptions='{0}'".format(java_options),
        "--conf",
        "spark.sql.warehouse.dir='{0}'".format(metastore_dir),
        "--conf",
        "spark.python.unix.domain.socket.dir={0}".format(sock_dir),
        "pyspark-shell",
    ]

    env["PYSPARK_SUBMIT_ARGS"] = " ".join(spark_args)

    timeout = os.environ.get("PYSPARK_TEST_TIMEOUT")
    if timeout is not None:
        env["PYSPARK_TEST_TIMEOUT"] = timeout
        timeout = int(timeout)

    output_prefix = get_valid_filename(pyspark_python + "__" + test_name + "__").lstrip("_")
    # Delete is always set to False since the cleanup will be either done by removing the
    # whole test dir, or the test output is retained.
    per_test_output = tempfile.NamedTemporaryFile(
        prefix=output_prefix, dir=tmp_dir, suffix=".log", delete=False
    )
    LOGGER.info(
        "Starting test(%s): %s (temp output: %s)", pyspark_python, test_name, per_test_output.name
    )
    cmd = [os.path.join(SPARK_HOME, "bin/pyspark")] + test_name.split()
    start_time = time.time()

    retcode = None
    try:
        retcode = TestRunner(test_name, cmd, env, per_test_output, timeout).run()
        if not keep_test_output:
            # There exists a race condition in Python and it causes flakiness in MacOS
            # https://github.com/python/cpython/issues/73885
            if platform.system() == "Darwin":
                os.system("rm -rf " + tmp_dir)
            else:
                shutil.rmtree(tmp_dir, ignore_errors=True)
    except BaseException:
        LOGGER.exception("Got exception while running %s with %s", test_name, pyspark_python)
        # Here, we use os._exit() instead of sys.exit() in order to force Python to exit even if
        # this code is invoked from a thread other than the main thread.
        os._exit(1)
    duration = time.time() - start_time
    # Exit on the first failure but exclude the code 5 for no test ran, see SPARK-46801.
    if retcode != 0 and retcode != 5:
        try:
            per_test_output.seek(0)
            with FAILURE_REPORTING_LOCK:
                with open(LOG_FILE, "ab") as log_file:
                    log_file.writelines(per_test_output)

            # We don't want the logging lines interleave with the test output, so we read the
            # full file and output with LOGGER which has internal locking.
            per_test_output.seek(0)
            lines = []
            for line in per_test_output:
                line = line.decode("utf-8", "replace")
                if not re.match("[0-9]+", line):
                    lines.append(line)
            LOGGER.error(f"{test_name} with {pyspark_python} failed:\n{''.join(lines)}")
            per_test_output.close()
        except BaseException:
            LOGGER.exception("Got an exception while trying to print failed test output")
        finally:
            print_red("\nHad test failures in %s with %s; see logs." % (test_name, pyspark_python))
            # Here, we use os._exit() instead of sys.exit() in order to force Python to exit even if
            # this code is invoked from a thread other than the main thread.
            os._exit(-1)
    else:
        skipped_counts = 0
        try:
            per_test_output.seek(0)
            # Here expects skipped test output from unittest when verbosity level is
            # 2 (or --verbose option is enabled).
            decoded_lines = map(lambda line: line.decode("utf-8", "replace"), iter(per_test_output))
            skipped_tests = list(
                filter(
                    lambda line: re.search(r"test_.* \(pyspark\..*\) ... (skip|SKIP)", line),
                    decoded_lines,
                )
            )
            skipped_counts = len(skipped_tests)
            if skipped_counts > 0:
                key = (pyspark_python, test_name)
                assert SKIPPED_TESTS is not None
                SKIPPED_TESTS[key] = skipped_tests
            per_test_output.close()
        except BaseException:
            import traceback

            print_red(
                "\nGot an exception while trying to store "
                "skipped test output:\n%s" % traceback.format_exc()
            )
            # Here, we use os._exit() instead of sys.exit() in order to force Python to exit even if
            # this code is invoked from a thread other than the main thread.
            os._exit(-1)
        if skipped_counts != 0:
            LOGGER.info(
                "Finished test(%s): %s (%is) ... %s tests were skipped",
                pyspark_python,
                test_name,
                duration,
                skipped_counts,
            )
        else:
            LOGGER.info("Finished test(%s): %s (%is)", pyspark_python, test_name, duration)


def get_default_python_executables():
    return [sys.executable]


def parse_opts():
    parser = ArgumentParser(prog="run-tests")
    parser.add_argument(
        "--python-executables",
        type=str,
        default=",".join(get_default_python_executables()),
        help="A comma-separated list of Python executables to test against (default: %(default)s)",
    )
    parser.add_argument(
        "--modules",
        type=str,
        default=",".join(sorted(python_modules.keys())),
        help="A comma-separated list of Python modules to test (default: %(default)s)",
    )
    parser.add_argument(
        "-p",
        "--parallelism",
        type=int,
        default=4,
        help="The number of suites to test in parallel (default %(default)d)",
    )
    parser.add_argument("--verbose", action="store_true", help="Enable additional debug logging")

    group = parser.add_argument_group("Developer Options")
    group.add_argument(
        "--testnames",
        type=str,
        default=None,
        help=(
            "A comma-separated list of specific modules, classes and functions of doctest "
            "or unittest to test. "
            "For example, 'pyspark.sql.foo' to run the module as unittests or doctests, "
            "'pyspark.sql.tests FooTests' to run the specific class of unittests, "
            "'pyspark.sql.tests FooTests.test_foo' to run the specific unittest in the class. "
            "'--modules' option is ignored if they are given."
        ),
    )
    group.add_argument(
        "-k",
        "--keep-test-output",
        action="store_true",
        default=False,
        help=(
            "If set to true will retain the temporary test directories. In addition, the "
            "standard output and standard error are redirected to a file in the target "
            "directory."
        ),
    )

    args, unknown = parser.parse_known_args()
    if unknown:
        parser.error("Unsupported arguments: %s" % " ".join(unknown))
    if args.parallelism < 1:
        parser.error("Parallelism cannot be less than 1")
    return args


def _check_coverage(python_exec):
    # Make sure if coverage is installed.
    try:
        subprocess_check_output(
            [python_exec, "-c", "import coverage"], stderr=open(os.devnull, "w")
        )
    except BaseException:
        print_red(
            "Coverage is not installed in Python executable '%s' "
            "but 'COVERAGE_PROCESS_START' environment variable is set, "
            "exiting." % python_exec
        )
        sys.exit(-1)


def main():
    opts = parse_opts()
    if opts.verbose:
        log_level = logging.DEBUG
    else:
        log_level = logging.INFO
    should_test_modules = opts.testnames is None
    logging.basicConfig(stream=sys.stdout, level=log_level, format="%(message)s")
    LOGGER.info("Running PySpark tests. Output is in %s", LOG_FILE)
    if os.path.exists(LOG_FILE):
        os.remove(LOG_FILE)
    python_execs = opts.python_executables.split(",")
    LOGGER.info("Will test against the following Python executables: %s", python_execs)

    if should_test_modules:
        modules_to_test = []
        for module_name in opts.modules.split(","):
            if module_name in python_modules:
                modules_to_test.append(python_modules[module_name])
            else:
                print(
                    "Error: unrecognized module '%s'. Supported modules: %s"
                    % (module_name, ", ".join(python_modules))
                )
                sys.exit(-1)
        LOGGER.info("Will test the following Python modules: %s", [x.name for x in modules_to_test])
    else:
        testnames_to_test = opts.testnames.split(",")
        LOGGER.info("Will test the following Python tests: %s", testnames_to_test)

    task_queue = Queue.PriorityQueue()
    for python_exec in python_execs:
        # Check if the python executable has coverage installed when 'COVERAGE_PROCESS_START'
        # environmental variable is set.
        if "COVERAGE_PROCESS_START" in os.environ:
            _check_coverage(python_exec)

        python_implementation = subprocess_check_output(
            [python_exec, "-c", "import platform; print(platform.python_implementation())"],
            universal_newlines=True,
        ).strip()
        LOGGER.info("%s python_implementation is %s", python_exec, python_implementation)
        LOGGER.info(
            "%s version is: %s",
            python_exec,
            subprocess_check_output(
                [python_exec, "--version"], stderr=subprocess.STDOUT, universal_newlines=True
            ).strip(),
        )
        if should_test_modules:
            for module in modules_to_test:
                if python_implementation not in module.excluded_python_implementations:
                    for test_goal in module.python_test_goals:
                        heavy_tests = [
                            "pyspark.streaming.tests",
                            "pyspark.mllib.tests",
                            "pyspark.tests",
                            "pyspark.sql.tests",
                            "pyspark.ml.tests",
                            "pyspark.pandas.tests",
                        ]
                        if any(map(lambda prefix: test_goal.startswith(prefix), heavy_tests)):
                            priority = 0
                        else:
                            priority = 100
                        task_queue.put((priority, (python_exec, test_goal)))
        else:
            for test_goal in testnames_to_test:
                task_queue.put((0, (python_exec, test_goal)))

    # Create the target directory before starting tasks to avoid races.
    target_dir = os.path.abspath(os.path.join(os.path.dirname(__file__), "target"))
    if not os.path.isdir(target_dir):
        os.mkdir(target_dir)

    def process_queue(task_queue):
        while True:
            try:
                (priority, (python_exec, test_goal)) = task_queue.get_nowait()
            except Queue.Empty:
                break
            try:
                run_individual_python_test(
                    target_dir, test_goal, python_exec, opts.keep_test_output
                )
            finally:
                task_queue.task_done()

    start_time = time.time()
    for _ in range(opts.parallelism):
        worker = Thread(target=process_queue, args=(task_queue,))
        worker.daemon = True
        worker.start()
    try:
        task_queue.join()
    except (KeyboardInterrupt, SystemExit):
        print_red("Exiting due to interrupt")
        sys.exit(-1)
    total_duration = time.time() - start_time
    LOGGER.info("Tests passed in %i seconds", total_duration)

    for key, lines in sorted(SKIPPED_TESTS.items()):
        pyspark_python, test_name = key
        LOGGER.info("\nSkipped tests in %s with %s:" % (test_name, pyspark_python))
        for line in lines:
            LOGGER.info("    %s" % line.rstrip())


if __name__ == "__main__":
    SKIPPED_TESTS = Manager().dict()
    main()