예제 #1
0
def print_version(output_format=None):
    """
    Print web server version information in the given format.
    """
    context = webserver_context.get_context()

    server_versions = ['{0}.{1}'.format(major, minor) for
                       major, minor in version.SUPPORTED_VERSIONS.items()]

    if output_format != 'json':
        server_versions = ', '.join(server_versions)

    rows = [
        ("Base package version", context.version),
        ("Package build date", context.package_build_date),
        ("Git commit ID (hash)", context.package_git_hash),
        ("Git tag information", context.package_git_tag),
        ("Configuration schema", str(context.product_db_version_info)),
        ("Database schema", str(context.run_db_version_info)),
        ("Server supported API (Thrift)", server_versions),
        ("Client API (Thrift)", version.CLIENT_API)
    ]

    if output_format != "json":
        print(output_formatters.twodim_to_str(output_format,
                                              ["Kind", "Version"],
                                              rows))
    elif output_format == "json":
        # Use a special JSON format here, instead of
        # [ {"kind": "something", "version": "0.0.0"}, {"kind": "foo", ... } ]
        # do
        # { "something": "0.0.0", "foo": ... }
        print(json.dumps(dict(rows)))
예제 #2
0
def print_version(output_format=None):
    """
    Print web server version information in the given format.
    """
    context = webserver_context.get_context()

    server_versions = [
        '{0}.{1}'.format(major, minor)
        for major, minor in version.SUPPORTED_VERSIONS.items()
    ]

    if output_format != 'json':
        server_versions = ', '.join(server_versions)

    rows = [("Base package version", context.version),
            ("Package build date", context.package_build_date),
            ("Git commit ID (hash)", context.package_git_hash),
            ("Git tag information", context.package_git_tag),
            ("Server supported API (Thrift)", server_versions),
            ("Client API (Thrift)", version.CLIENT_API)]

    if output_format != "json":
        print(
            output_formatters.twodim_to_str(output_format, ["Kind", "Version"],
                                            rows))
    elif output_format == "json":
        # Use a special JSON format here, instead of
        # [ {"kind": "something", "version": "0.0.0"}, {"kind": "foo", ... } ]
        # do
        # { "something": "0.0.0", "foo": ... }
        print(json.dumps(dict(rows)))
예제 #3
0
    def __init__(self):
        context = webserver_context.get_context()

        self.server_versions = [
            f'{major}.{minor}'
            for major, minor in version.SUPPORTED_VERSIONS.items()
        ]

        self.version = context.version
        self.build_date = context.package_build_date
        self.git_hash = context.package_git_hash
        self.git_tag = context.package_git_tag
        self.client_api = version.CLIENT_API
예제 #4
0
def upgrade():
    op.add_column(
        'products',
        sa.Column('latest_storage_date', sa.DateTime(), nullable=True))

    op.add_column(
        'products',
        sa.Column('num_of_runs',
                  sa.Integer(),
                  server_default="0",
                  nullable=True))

    try:
        product_con = op.get_bind()
        products = product_con.execute(
            "SELECT id, connection from products").fetchall()

        context = webserver_context.get_context()
        for id, connection in products:
            sql_server = database.SQLServer.from_connection_string(
                connection, RUN_META, context.run_migration_root)

            engine = sa.create_engine(sql_server.get_connection_string())
            conn = engine.connect()

            run_info = \
                conn.execute("SELECT count(*), max(date) from runs").fetchone()

            values = [f"num_of_runs={run_info[0]}"]
            if run_info[1]:
                values.append(f"latest_storage_date='{run_info[1]}'")

            product_con.execute(f"""
                UPDATE products
                SET {', '.join(values)}
                WHERE id={id}
            """)
    except Exception as ex:
        LOG.error(
            "Failed to fill product detail columns (num_of_runs, "
            "latest_storage_date): %s", ex)
        pass
예제 #5
0
def server_init_start(args):
    """
    Start or manage a CodeChecker report server.
    """

    if 'list' in args or 'stop' in args or 'stop_all' in args:
        __instance_management(args)
        sys.exit(0)

    if 'reload' in args:
        __reload_config(args)
        sys.exit(0)

    # Actual server starting from this point.
    if not host_check.check_zlib():
        raise Exception("zlib is not available on the system!")

    # WARNING
    # In case of SQLite args.dbaddress default value is used
    # for which the is_localhost should return true.
    if util.is_localhost(args.dbaddress) and \
            not os.path.exists(args.config_directory):
        os.makedirs(args.config_directory)

    # Make sure the SQLite file can be created if it not exists.
    if 'sqlite' in args and \
            not os.path.isdir(os.path.dirname(args.sqlite)):
        os.makedirs(os.path.dirname(args.sqlite))

    if 'reset_root' in args:
        try:
            os.remove(os.path.join(args.config_directory, 'root.user'))
            LOG.info("Master superuser (root) credentials invalidated and "
                     "deleted. New ones will be generated...")
        except OSError:
            # File doesn't exist.
            pass

    if 'force_auth' in args:
        LOG.info("'--force-authentication' was passed as a command-line "
                 "option. The server will ask for users to authenticate!")

    context = webserver_context.get_context()
    context.codechecker_workspace = args.config_directory
    context.db_username = args.dbusername

    check_env = get_check_env(context.path_env_extra,
                              context.ld_lib_path_extra)

    cfg_sql_server = database.SQLServer.from_cmdline_args(
        vars(args),
        CONFIG_META,
        context.config_migration_root,
        interactive=True,
        env=check_env)

    LOG.info("Checking configuration database ...")
    db_status = cfg_sql_server.connect()
    db_status_msg = database_status.db_status_msg.get(db_status)
    LOG.info(db_status_msg)

    if db_status == DBStatus.SCHEMA_MISSING:
        LOG.debug("Config database schema is missing, initializing new.")
        db_status = cfg_sql_server.connect(init=True)
        if db_status != DBStatus.OK:
            LOG.error("Config database initialization failed!")
            LOG.error("Please check debug logs.")
            sys.exit(1)

    if db_status == DBStatus.SCHEMA_MISMATCH_NO:
        LOG.debug("Configuration database schema mismatch.")
        LOG.debug("No schema upgrade is possible.")
        sys.exit(1)

    force_upgrade = True if 'force_upgrade' in args else False

    if db_status == DBStatus.SCHEMA_MISMATCH_OK:
        LOG.debug("Configuration database schema mismatch.")
        LOG.debug("Schema upgrade is possible.")
        LOG.warning("Please note after migration only "
                    "newer CodeChecker versions can be used "
                    "to start the server")
        LOG.warning("It is advised to make a full backup of your "
                    "configuration database")

        LOG.warning(cfg_sql_server.get_db_location())

        question = 'Do you want to upgrade to the new schema?' \
                   ' Y(es)/n(o) '
        if force_upgrade or util.get_user_input(question):
            print("Upgrading schema ...")
            ret = cfg_sql_server.upgrade()
            msg = database_status.db_status_msg.get(ret,
                                                    'Unknown database status')
            print(msg)
            if ret != DBStatus.OK:
                LOG.error("Schema migration failed")
                sys.exit(ret)
        else:
            LOG.info("No schema migration was done.")
            sys.exit(0)

    if db_status == DBStatus.MISSING:
        LOG.error("Missing configuration database.")
        LOG.error("Server can not be started.")
        sys.exit(1)

    # Configuration database setup and check is needed before database
    # statuses can be checked.
    try:
        if args.status:
            ret = __db_status_check(cfg_sql_server, context, args.status)
            sys.exit(ret)
    except AttributeError:
        LOG.debug('Status was not in the arguments.')

    try:
        if args.product_to_upgrade:
            ret = __db_migration(cfg_sql_server, context,
                                 args.product_to_upgrade, force_upgrade)
            sys.exit(ret)
    except AttributeError:
        LOG.debug('Product upgrade was not in the arguments.')

    # Create the main database link from the arguments passed over the
    # command line.
    cfg_dir = os.path.abspath(args.config_directory)
    default_product_path = os.path.join(cfg_dir, 'Default.sqlite')
    create_default_product = 'sqlite' in args and \
                             not os.path.exists(default_product_path)

    if create_default_product:
        # Create a default product and add it to the configuration database.

        LOG.debug("Create default product...")
        LOG.debug("Configuring schema and migration...")

        prod_server = database.SQLiteDatabase(default_product_path, RUN_META,
                                              context.run_migration_root,
                                              check_env)

        LOG.debug("Checking 'Default' product database.")
        db_status = prod_server.connect()
        if db_status != DBStatus.MISSING:
            db_status = prod_server.connect(init=True)
            LOG.debug(database_status.db_status_msg.get(db_status))
            if db_status != DBStatus.OK:
                LOG.error("Failed to configure default product")
                sys.exit(1)

        product_conn_string = prod_server.get_connection_string()

        server.add_initial_run_database(cfg_sql_server, product_conn_string)

        LOG.info("Product 'Default' at '%s' created and set up.",
                 default_product_path)

    prod_statuses = check_product_db_status(cfg_sql_server, context)

    upgrade_available = {}
    for k, v in prod_statuses.items():
        db_status, _, _, _ = v
        if db_status == DBStatus.SCHEMA_MISMATCH_OK or \
                db_status == DBStatus.SCHEMA_MISSING:
            upgrade_available[k] = v

    if upgrade_available:
        print_prod_status(prod_statuses)
        LOG.warning("Multiple products can be upgraded, make a backup!")
        __db_migration(cfg_sql_server, context, 'all', force_upgrade)

    prod_statuses = check_product_db_status(cfg_sql_server, context)
    print_prod_status(prod_statuses)

    non_ok_db = False
    for k, v in prod_statuses.items():
        db_status, _, _, _ = v
        if db_status != DBStatus.OK:
            non_ok_db = True
        break

    if non_ok_db:
        msg = "There are some database issues. " \
              "Do you want to start the " \
              "server? Y(es)/n(o) "
        if not util.get_user_input(msg):
            sys.exit(1)

    # Start database viewer.
    checker_md_docs = os.path.join(context.doc_root, 'checker_md_docs')
    checker_md_docs_map = os.path.join(checker_md_docs, 'checker_doc_map.json')
    checker_md_docs_map = util.load_json_or_empty(checker_md_docs_map, {})

    package_data = {
        'www_root': context.www_root,
        'doc_root': context.doc_root,
        'checker_md_docs': checker_md_docs,
        'checker_md_docs_map': checker_md_docs_map,
        'version': context.package_git_tag
    }

    suppr_handler = suppress_handler. \
        GenericSuppressHandler(None, False)

    try:
        server.start_server(args.config_directory, package_data,
                            args.view_port, cfg_sql_server, suppr_handler,
                            args.listen_address, 'force_auth' in args,
                            args.skip_db_cleanup, context, check_env)
    except socket.error as err:
        if err.errno == errno.EADDRINUSE:
            LOG.error(
                "Server can't be started, maybe the given port number "
                "(%s) is already used. Check the connection "
                "parameters.", args.view_port)
            sys.exit(1)
        else:
            raise
예제 #6
0
def main(args):
    """
    Store the defect results in the specified input list as bug reports in the
    database.
    """
    logger.setup_logger(args.verbose if 'verbose' in args else None)

    try:
        cmd_config.check_config_file(args)
    except FileNotFoundError as fnerr:
        LOG.error(fnerr)
        sys.exit(1)

    if not host_check.check_zlib():
        raise Exception("zlib is not available on the system!")

    # To ensure the help message prints the default folder properly,
    # the 'default' for 'args.input' is a string, not a list.
    # But we need lists for the foreach here to work.
    if isinstance(args.input, str):
        args.input = [args.input]

    if 'name' not in args:
        LOG.debug("Generating name for analysis...")
        generated = __get_run_name(args.input)
        if generated:
            setattr(args, 'name', generated)
        else:
            LOG.error("No suitable name was found in the inputs for the "
                      "analysis run. Please specify one by passing argument "
                      "--name run_name in the invocation.")
            sys.exit(2)  # argparse returns error code 2 for bad invocations.

    LOG.info("Storing analysis results for run '" + args.name + "'")

    if 'force' in args:
        LOG.info("argument --force was specified: the run with name '" +
                 args.name + "' will be deleted.")

    # Setup connection to the remote server.
    client = libclient.setup_client(args.product_url)

    _, zip_file = tempfile.mkstemp('.zip')
    LOG.debug("Will write mass store ZIP to '%s'...", zip_file)

    try:
        LOG.debug("Assembling zip file.")
        try:
            assemble_zip(args.input, zip_file, client)
        except Exception as ex:
            print(ex)
            import traceback
            traceback.print_stack()

        zip_size = os.stat(zip_file).st_size
        LOG.debug("Assembling zip done, size is %s",
                  sizeof_fmt(zip_size))

        if zip_size > MAX_UPLOAD_SIZE:
            LOG.error("The result list to upload is too big (max: %s).",
                      sizeof_fmt(MAX_UPLOAD_SIZE))
            sys.exit(1)

        with open(zip_file, 'rb') as zf:
            b64zip = base64.b64encode(zf.read()).decode("utf-8")

        context = webserver_context.get_context()

        trim_path_prefixes = args.trim_path_prefix if \
            'trim_path_prefix' in args else None

        description = args.description if 'description' in args else None

        LOG.info("Storing results to the server...")
        client.massStoreRun(args.name,
                            args.tag if 'tag' in args else None,
                            str(context.version),
                            b64zip,
                            'force' in args,
                            trim_path_prefixes,
                            description)

        # Storing analysis statistics if the server allows them.
        if client.allowsStoringAnalysisStatistics():
            storing_analysis_statistics(client, args.input, args.name)

        LOG.info("Storage finished successfully.")
    except RequestFailed as reqfail:
        if reqfail.errorCode == ErrorCode.SOURCE_FILE:
            header = ['File', 'Line', 'Checker name']
            table = twodim_to_str('table',
                                  header,
                                  [c.split('|') for c in reqfail.extraInfo])
            LOG.warning("Setting the review statuses for some reports failed "
                        "because of non valid source code comments: "
                        "%s\n %s", reqfail.message, table)
        sys.exit(1)
    except Exception as ex:
        import traceback
        traceback.print_stack()
        LOG.info("Storage failed: %s", str(ex))
        sys.exit(1)
    finally:
        os.remove(zip_file)
예제 #7
0
def main(args):
    """
    Store the defect results in the specified input list as bug reports in the
    database.
    """
    logger.setup_logger(args.verbose if 'verbose' in args else None)

    if not host_check.check_zlib():
        raise Exception("zlib is not available on the system!")

    # To ensure the help message prints the default folder properly,
    # the 'default' for 'args.input' is a string, not a list.
    # But we need lists for the foreach here to work.
    if isinstance(args.input, str):
        args.input = [args.input]

    if 'name' not in args:
        LOG.debug("Generating name for analysis...")
        generated = __get_run_name(args.input)
        if generated:
            setattr(args, 'name', generated)
        else:
            LOG.error("No suitable name was found in the inputs for the "
                      "analysis run. Please specify one by passing argument "
                      "--name run_name in the invocation.")
            sys.exit(2)  # argparse returns error code 2 for bad invocations.

    LOG.info("Storing analysis results for run '" + args.name + "'")

    if 'force' in args:
        LOG.info("argument --force was specified: the run with name '" +
                 args.name + "' will be deleted.")

    protocol, host, port, product_name = split_product_url(args.product_url)

    # Before any transmission happens, check if we have the PRODUCT_STORE
    # permission to prevent a possibly long ZIP operation only to get an
    # error later on.
    product_client = libclient.setup_product_client(protocol,
                                                    host, port, product_name)
    product_id = product_client.getCurrentProduct().id

    auth_client, _ = libclient.setup_auth_client(protocol, host, port)
    has_perm = libclient.check_permission(
        auth_client, Permission.PRODUCT_STORE, {'productID': product_id})
    if not has_perm:
        LOG.error("You are not authorised to store analysis results in "
                  "product '%s'", product_name)
        sys.exit(1)

    # Setup connection to the remote server.
    client = libclient.setup_client(args.product_url, product_client=False)

    LOG.debug("Initializing client connecting to %s:%d/%s done.",
              host, port, product_name)

    _, zip_file = tempfile.mkstemp('.zip')
    LOG.debug("Will write mass store ZIP to '%s'...", zip_file)

    try:
        assemble_zip(args.input, zip_file, client)

        if os.stat(zip_file).st_size > MAX_UPLOAD_SIZE:
            LOG.error("The result list to upload is too big (max: %s).",
                      sizeof_fmt(MAX_UPLOAD_SIZE))
            sys.exit(1)

        with open(zip_file, 'rb') as zf:
            b64zip = base64.b64encode(zf.read())

        context = webserver_context.get_context()

        trim_path_prefixes = args.trim_path_prefix if \
            'trim_path_prefix' in args else None

        client.massStoreRun(args.name,
                            args.tag if 'tag' in args else None,
                            str(context.version),
                            b64zip,
                            'force' in args,
                            trim_path_prefixes)

        # Storing analysis statistics if the server allows them.
        if client.allowsStoringAnalysisStatistics():
            storing_analysis_statistics(client, args.input, args.name)

        LOG.info("Storage finished successfully.")
    except RequestFailed as reqfail:
        if reqfail.errorCode == ErrorCode.SOURCE_FILE:
            header = ['File', 'Line', 'Checker name']
            table = twodim_to_str('table',
                                  header,
                                  [c.split('|') for c in reqfail.extraInfo])
            LOG.warning("Setting the review statuses for some reports failed "
                        "because of non valid source code comments: "
                        "%s\n %s", reqfail.message, table)
        sys.exit(1)
    except Exception as ex:
        LOG.info("Storage failed: %s", str(ex))
        sys.exit(1)
    finally:
        os.remove(zip_file)
예제 #8
0
def handle_diff_results(args):

    init_logger(args.verbose if 'verbose' in args else None)
    check_deprecated_arg_usage(args)

    f_severities, f_checkers, f_file_path, _, _ = check_filter_values(args)

    context = webserver_context.get_context()

    def skip_report_dir_result(report):
        """
        Returns True if the report should be skipped from the results based on
        the given filter set.
        """
        if f_severities:
            severity_name = context.severity_map.get(report.main['check_name'])
            if severity_name.lower() not in map(str.lower, f_severities):
                return True

        if f_checkers:
            checker_name = report.main['check_name']
            if not any([
                    re.match(r'^' + c.replace("*", ".*") + '$', checker_name,
                             re.IGNORECASE) for c in f_checkers
            ]):
                return True

        if f_file_path:
            file_path = report.files[int(report.main['location']['file'])]
            if not any([
                    re.match(r'^' + f.replace("*", ".*") + '$', file_path,
                             re.IGNORECASE) for f in f_file_path
            ]):
                return True

        if 'checker_msg' in args:
            checker_msg = report.main['description']
            if not any([
                    re.match(r'^' + c.replace("*", ".*") + '$', checker_msg,
                             re.IGNORECASE) for c in args.checker_msg
            ]):
                return True

        return False

    def get_report_dir_results(reportdir):
        all_reports = []
        processed_path_hashes = set()
        for filename in os.listdir(reportdir):
            if filename.endswith(".plist"):
                file_path = os.path.join(reportdir, filename)
                LOG.debug("Parsing: %s", file_path)
                try:
                    files, reports = plist_parser.parse_plist(file_path)
                    for report in reports:
                        path_hash = get_report_path_hash(report, files)
                        if path_hash in processed_path_hashes:
                            LOG.debug("Not showing report because it is a "
                                      "deduplication of an already processed "
                                      "report!")
                            LOG.debug("Path hash: %s", path_hash)
                            LOG.debug(report)
                            continue

                        if skip_report_dir_result(report):
                            continue

                        processed_path_hashes.add(path_hash)
                        report.main['location']['file_name'] = \
                            files[int(report.main['location']['file'])]
                        all_reports.append(report)

                except Exception as ex:
                    LOG.error('The generated plist is not valid!')
                    LOG.error(ex)
        return all_reports

    def get_line_from_file(filename, lineno):
        with open(filename, 'r') as f:
            i = 1
            for line in f:
                if i == lineno:
                    return line
                i += 1
        return ""

    def get_diff_base_results(client, baseids, base_hashes, suppressed_hashes):
        base_results = []
        report_filter = ttypes.ReportFilter()
        add_filter_conditions(client, report_filter, args)

        sort_mode = [(ttypes.SortMode(ttypes.SortType.FILENAME,
                                      ttypes.Order.ASC))]
        limit = constants.MAX_QUERY_SIZE
        offset = 0

        report_filter.reportHash = base_hashes + suppressed_hashes
        results = client.getRunResults(baseids, limit, offset, sort_mode,
                                       report_filter, None)
        while results:
            base_results.extend(results)
            offset += limit
            results = client.getRunResults(baseids, limit, offset, sort_mode,
                                           report_filter, None)
        return base_results

    def get_suppressed_reports(reports):
        """
        Returns suppressed reports.
        """
        suppressed_in_code = []
        for rep in reports:
            bughash = rep.report_hash
            source_file = rep.main['location']['file_name']
            bug_line = rep.main['location']['line']
            checker_name = rep.main['check_name']

            sc_handler = SourceCodeCommentHandler(source_file)
            src_comment_data = sc_handler.filter_source_line_comments(
                bug_line, checker_name)

            if len(src_comment_data) == 1:
                suppressed_in_code.append(bughash)
                LOG.debug("Bug %s is suppressed in code. file: %s Line %s",
                          bughash, source_file, bug_line)
            elif len(src_comment_data) > 1:
                LOG.warning(
                    "Multiple source code comment can be found "
                    "for '%s' checker in '%s' at line %s. "
                    "This bug will not be suppressed!", checker_name,
                    source_file, bug_line)
        return suppressed_in_code

    def get_diff_type():
        """
        Returns Thrift DiffType value by processing the arguments.
        """
        if 'new' in args:
            return ttypes.DiffType.NEW

        if 'unresolved' in args:
            return ttypes.DiffType.UNRESOLVED

        if 'resolved' in args:
            return ttypes.DiffType.RESOLVED

        return None

    def get_diff_local_dir_remote_run(client, report_dir, run_name):
        """
        Compares a local report directory with a remote run.
        """
        filtered_reports = []
        report_dir_results = get_report_dir_results(
            os.path.abspath(report_dir))
        suppressed_in_code = get_suppressed_reports(report_dir_results)

        diff_type = get_diff_type()
        run_ids, run_names, _ = process_run_arg(run_name)
        local_report_hashes = set([r.report_hash for r in report_dir_results])

        if diff_type == ttypes.DiffType.NEW:
            # Get report hashes which can be found only in the remote runs.
            remote_hashes = \
                client.getDiffResultsHash(run_ids,
                                          local_report_hashes,
                                          ttypes.DiffType.RESOLVED)

            results = get_diff_base_results(client, run_ids, remote_hashes,
                                            suppressed_in_code)
            for result in results:
                filtered_reports.append(result)
        elif diff_type == ttypes.DiffType.UNRESOLVED:
            # Get remote hashes which can be found in the remote run and in the
            # local report directory.
            remote_hashes = \
                client.getDiffResultsHash(run_ids,
                                          local_report_hashes,
                                          ttypes.DiffType.UNRESOLVED)
            for result in report_dir_results:
                rep_h = result.report_hash
                if rep_h in remote_hashes and rep_h not in suppressed_in_code:
                    filtered_reports.append(result)
        elif diff_type == ttypes.DiffType.RESOLVED:
            # Get remote hashes which can be found in the remote run and in the
            # local report directory.
            remote_hashes = \
                client.getDiffResultsHash(run_ids,
                                          local_report_hashes,
                                          ttypes.DiffType.UNRESOLVED)
            for result in report_dir_results:
                if result.report_hash not in remote_hashes:
                    filtered_reports.append(result)

        return filtered_reports, run_names

    def get_diff_remote_run_local_dir(client, run_name, report_dir):
        """
        Compares a remote run with a local report directory.
        """
        filtered_reports = []
        report_dir_results = get_report_dir_results(
            os.path.abspath(report_dir))
        suppressed_in_code = get_suppressed_reports(report_dir_results)

        diff_type = get_diff_type()
        run_ids, run_names, _ = process_run_arg(run_name)
        local_report_hashes = set([r.report_hash for r in report_dir_results])

        remote_hashes = client.getDiffResultsHash(run_ids, local_report_hashes,
                                                  diff_type)

        if diff_type in [ttypes.DiffType.NEW, ttypes.DiffType.UNRESOLVED]:
            # Shows reports from the report dir which are not present in
            # the baseline (NEW reports) or appear in both side (UNRESOLVED
            # reports) and not suppressed in the code.
            for result in report_dir_results:
                rep_h = result.report_hash
                if rep_h in remote_hashes and rep_h not in suppressed_in_code:
                    filtered_reports.append(result)
        elif diff_type == ttypes.DiffType.RESOLVED:
            # Show bugs in the baseline (server) which are not present in
            # the report dir or suppressed.
            results = get_diff_base_results(client, run_ids, remote_hashes,
                                            suppressed_in_code)
            for result in results:
                filtered_reports.append(result)

        return filtered_reports, run_names

    def get_diff_remote_runs(client, basename, newname):
        """
        Compares two remote runs and returns the filtered results.
        """
        report_filter = ttypes.ReportFilter()
        add_filter_conditions(client, report_filter, args)

        base_ids, base_run_names, base_run_tags = process_run_arg(basename)
        report_filter.runTag = base_run_tags

        cmp_data = ttypes.CompareData()
        cmp_data.diffType = get_diff_type()

        new_ids, new_run_names, new_run_tags = process_run_arg(newname)
        cmp_data.runIds = new_ids
        cmp_data.runTag = new_run_tags

        # Do not show resolved bugs in compare mode new.
        if cmp_data.diffType == ttypes.DiffType.NEW:
            report_filter.detectionStatus = [
                ttypes.DetectionStatus.NEW, ttypes.DetectionStatus.UNRESOLVED,
                ttypes.DetectionStatus.REOPENED
            ]

        sort_mode = [(ttypes.SortMode(ttypes.SortType.FILENAME,
                                      ttypes.Order.ASC))]
        limit = constants.MAX_QUERY_SIZE
        offset = 0

        all_results = []
        results = client.getRunResults(base_ids, limit, offset, sort_mode,
                                       report_filter, cmp_data)

        while results:
            all_results.extend(results)
            offset += limit
            results = client.getRunResults(base_ids, limit, offset, sort_mode,
                                           report_filter, cmp_data)
        return all_results, base_run_names, new_run_names

    def get_diff_local_dirs(basename, newname):
        """
        Compares two report directories and returns the filtered results.
        """
        filtered_reports = []
        base_results = get_report_dir_results(os.path.abspath(basename))
        new_results = get_report_dir_results(os.path.abspath(newname))

        base_hashes = set([res.report_hash for res in base_results])
        new_hashes = set([res.report_hash for res in new_results])

        diff_type = get_diff_type()
        if diff_type == ttypes.DiffType.NEW:
            for res in new_results:
                if res.report_hash not in base_hashes:
                    filtered_reports.append(res)
        if diff_type == ttypes.DiffType.UNRESOLVED:
            for res in new_results:
                if res.report_hash in base_hashes:
                    filtered_reports.append(res)
        elif diff_type == ttypes.DiffType.RESOLVED:
            for res in base_results:
                if res.report_hash not in new_hashes:
                    filtered_reports.append(res)

        return filtered_reports

    def cached_report_file_lookup(file_cache, file_id):
        """
        Get source file data for the given file and caches it in a file cache
        if file data is not found in the cache. Finally, it returns the source
        file data from the cache.
        """
        if file_id not in file_cache:
            source = client.getSourceFileData(file_id, True,
                                              ttypes.Encoding.BASE64)
            file_content = base64.b64decode(source.fileContent)
            file_cache[file_id] = {
                'id': file_id,
                'path': source.filePath,
                'content': file_content
            }

        return file_cache[file_id]

    def get_report_data(client, reports, file_cache):
        """
        Returns necessary report files and report data events for the HTML
        plist parser.
        """
        file_sources = {}
        report_data = []

        for report in reports:
            file_sources[report.fileId] = cached_report_file_lookup(
                file_cache, report.fileId)

            details = client.getReportDetails(report.reportId)
            events = []
            for event in details.pathEvents:
                file_sources[event.fileId] = cached_report_file_lookup(
                    file_cache, event.fileId)

                location = {
                    'line': event.startLine,
                    'col': event.startCol,
                    'file': event.fileId
                }

                events.append({'location': location, 'message': event.msg})

            # Get extended data.
            macros = []
            notes = []
            for extended_data in details.extendedData:
                file_sources[extended_data.fileId] = cached_report_file_lookup(
                    file_cache, extended_data.fileId)

                location = {
                    'line': extended_data.startLine,
                    'col': extended_data.startCol,
                    'file': extended_data.fileId
                }

                if extended_data.type == ttypes.ExtendedReportDataType.MACRO:
                    macros.append({
                        'location': location,
                        'expansion': event.msg
                    })
                elif extended_data.type == ttypes.ExtendedReportDataType.NOTE:
                    notes.append({'location': location, 'message': event.msg})

            report_data.append({
                'events': events,
                'macros': macros,
                'notes': notes,
                'path': report.checkedFile,
                'reportHash': report.bugHash,
                'checkerName': report.checkerId
            })

        return {'files': file_sources, 'reports': report_data}

    def reports_to_report_data(reports):
        """
        Converts reports from Report class from one plist file
        to report data events for the HTML plist parser.
        """
        file_sources = {}
        report_data = []

        for report in reports:
            # Not all report in this list may refer to the same files
            # thus we need to create a single file list with
            # all files from all reports.
            for file_index, file_path in enumerate(report.files):
                if file_index not in file_sources:
                    try:
                        with io.open(file_path,
                                     'r',
                                     encoding='UTF-8',
                                     errors='ignore') as source_data:
                            content = source_data.read()
                    except (OSError, IOError):
                        content = file_path + " NOT FOUND."
                    file_sources[file_index] = {
                        'id': file_index,
                        'path': file_path,
                        'content': content
                    }

            events = []
            for element in report.bug_path:
                kind = element['kind']
                if kind == 'event':
                    events.append({
                        'location': element['location'],
                        'message': element['message']
                    })

            macros = []
            for macro in report.macro_expansions:
                macros.append({
                    'location': macro['location'],
                    'expansion': macro['expansion'],
                    'name': macro['name']
                })

            notes = []
            for note in report.notes:
                notes.append({
                    'location': note['location'],
                    'message': note['message']
                })

            report_hash = report.main['issue_hash_content_of_line_in_context']
            report_data.append({
                'events': events,
                'macros': macros,
                'notes': notes,
                'path': report.main['location']['file_name'],
                'reportHash': report_hash,
                'checkerName': report.main['check_name']
            })

        return {'files': file_sources, 'reports': report_data}

    def report_to_html(client, reports, output_dir):
        """
        Generate HTML output files for the given reports in the given output
        directory by using the Plist To HTML parser.
        """
        html_builder = PlistToHtml.HtmlBuilder(context.path_plist_to_html_dist,
                                               context.severity_map)

        file_report_map = defaultdict(list)
        for report in reports:
            if isinstance(report, Report):
                file_path = report.main['location']['file_name']
            else:
                file_path = report.checkedFile
            file_report_map[file_path].append(report)

        file_cache = {}
        for file_path, file_reports in file_report_map.items():
            checked_file = file_path
            filename = os.path.basename(checked_file)
            h = int(hashlib.md5(file_path).hexdigest(), 16) % (10**8)

            if isinstance(file_reports[0], Report):
                report_data = reports_to_report_data(file_reports)
            else:
                report_data = get_report_data(client, file_reports, file_cache)

            output_path = os.path.join(output_dir,
                                       filename + '_' + str(h) + '.html')
            html_builder.create(output_path, report_data)
            print('Html file was generated for file://{0}: file://{1}'.format(
                checked_file, output_path))

        html_builder.create_index_html(output_dir)

    def print_reports(client, reports, output_format):
        output_dir = args.export_dir if 'export_dir' in args else None
        if 'clean' in args and os.path.isdir(output_dir):
            print("Previous analysis results in '{0}' have been removed, "
                  "overwriting with current results.".format(output_dir))
            shutil.rmtree(output_dir)

        if output_format == 'json':
            output = []
            for report in reports:
                if isinstance(report, Report):
                    output.append(report.main)
                else:
                    output.append(report)
            print(CmdLineOutputEncoder().encode(output))
            return

        if output_format == 'html':
            output_dir = args.export_dir
            if not os.path.exists(output_dir):
                os.makedirs(output_dir)

            print("Generating HTML output files to file://{0} directory:\n".
                  format(output_dir))

            report_to_html(client, reports, output_dir)

            print('\nTo view the results in a browser run:\n'
                  '  $ firefox {0}'.format(
                      os.path.join(args.export_dir, 'index.html')))
            return

        header = ['File', 'Checker', 'Severity', 'Msg', 'Source']
        rows = []

        source_lines = defaultdict(set)
        for report in reports:
            if not isinstance(report, Report) and report.line is not None:
                source_lines[report.fileId].add(report.line)

        lines_in_files_requested = []
        for key in source_lines:
            lines_in_files_requested.append(
                ttypes.LinesInFilesRequested(fileId=key,
                                             lines=source_lines[key]))

        for report in reports:
            source_line = ''
            if isinstance(report, Report):
                # report is coming from a plist file.
                bug_line = report.main['location']['line']
                bug_col = report.main['location']['col']
                checked_file = report.main['location']['file_name']\
                    + ':' + str(bug_line) + ":" + str(bug_col)
                check_name = report.main['check_name']
                sev = context.severity_map.get(check_name)
                check_msg = report.main['description']
                source_line =\
                    get_line_from_file(report.main['location']['file_name'],
                                       bug_line)
            else:
                # report is of ReportData type coming from CodeChecker server.
                bug_line = report.line
                bug_col = report.column
                sev = ttypes.Severity._VALUES_TO_NAMES[report.severity]
                checked_file = report.checkedFile
                if bug_line is not None:
                    checked_file += ':' + str(bug_line) + ":" + str(bug_col)

                check_name = report.checkerId
                check_msg = report.checkerMsg

                if lines_in_files_requested:
                    source_line_contents = client.getLinesInSourceFileContents(
                        lines_in_files_requested, ttypes.Encoding.BASE64)
                    source_line = base64.b64decode(
                        source_line_contents[report.fileId][bug_line])
            rows.append(
                (sev, checked_file, check_msg, check_name, source_line))
        if output_format == 'plaintext':
            for row in rows:
                print("[{0}] {1}: {2} [{3}]\n{4}\n".format(
                    row[0], row[1], row[2], row[3], row[4]))
        else:
            print(twodim_to_str(output_format, header, rows))

    def get_run_tag(client, run_ids, tag_name):
        """
        Returns run tag information for the given tag name in the given runs.
        """
        run_history_filter = ttypes.RunHistoryFilter()
        run_history_filter.tagNames = [tag_name]
        run_histories = client.getRunHistory(run_ids, None, None,
                                             run_history_filter)

        return run_histories[0] if len(run_histories) else None

    def process_run_arg(run_arg_with_tag):
        """
        Process the argument and returns run ids a run tag ids.
        The argument has the following format: <run_name>:<run_tag>
        """
        run_with_tag = run_arg_with_tag.split(':')
        run_name = run_with_tag[0]

        runs = get_runs(client, [run_name])
        run_ids = map(lambda run: run.runId, runs)
        run_names = map(lambda run: run.name, runs)

        # Set base run tag if it is available.
        run_tag_name = run_with_tag[1] if len(run_with_tag) > 1 else None
        run_tags = None
        if run_tag_name:
            tag = get_run_tag(client, run_ids, run_tag_name)
            run_tags = [tag.id] if tag else None

        if not run_ids:
            LOG.warning("No run names match the given pattern: %s",
                        run_arg_with_tag)
            sys.exit(1)

        LOG.info("Matching runs: %s", ', '.join(map(lambda run: run.name,
                                                    runs)))

        return run_ids, run_names, run_tags

    def print_diff_results(reports):
        """
        Print the results.
        """
        if reports:
            print_reports(client, reports, args.output_format)
        else:
            LOG.info("No results.")

    client = None

    # We set up the client if we are not comparing two local report directory.
    if not os.path.isdir(args.basename) or not os.path.isdir(args.newname):
        client = setup_client(args.product_url)

    if os.path.isdir(args.basename) and os.path.isdir(args.newname):
        reports = get_diff_local_dirs(args.basename, args.newname)
        print_diff_results(reports)
        LOG.info("Compared two local report directories %s and %s",
                 os.path.abspath(args.basename), os.path.abspath(args.newname))
    elif os.path.isdir(args.newname):
        reports, base_run_names = \
            get_diff_remote_run_local_dir(client,
                                          args.basename,
                                          os.path.abspath(args.newname))
        print_diff_results(reports)
        LOG.info(
            "Compared remote run(s) %s (matching: %s) and local report "
            "directory %s", args.basename, ', '.join(base_run_names),
            os.path.abspath(args.newname))
    elif os.path.isdir(args.basename):
        reports, new_run_names = \
            get_diff_local_dir_remote_run(client,
                                          os.path.abspath(args.basename),
                                          args.newname)
        print_diff_results(reports)
        LOG.info(
            "Compared local report directory %s and remote run(s) %s "
            "(matching: %s).", os.path.abspath(args.basename), args.newname,
            ', '.join(new_run_names))
    else:
        reports, base_run_names, new_run_names = \
            get_diff_remote_runs(client, args.basename, args.newname)
        print_diff_results(reports)
        LOG.info(
            "Compared multiple remote runs %s (matching: %s) and %s "
            "(matching: %s)", args.basename, ', '.join(base_run_names),
            args.newname, ', '.join(new_run_names))
예제 #9
0
 def setup_class(cls):
     ctx = webserver_context.get_context()
     cls.__git_commit_urls = ctx.git_commit_urls
예제 #10
0
def handle_diff_results(args):

    init_logger(args.verbose if 'verbose' in args else None)
    check_deprecated_arg_usage(args)

    f_severities, f_checkers, f_file_path, _, _ = check_filter_values(args)

    context = webserver_context.get_context()

    def skip_report_dir_result(report):
        """
        Returns True if the report should be skipped from the results based on
        the given filter set.
        """
        if f_severities:
            severity_name = context.severity_map.get(report.main['check_name'])
            if severity_name.lower() not in map(str.lower, f_severities):
                return True

        if f_checkers:
            checker_name = report.main['check_name']
            if not any([re.match(r'^' + c.replace("*", ".*") + '$',
                                 checker_name, re.IGNORECASE)
                        for c in f_checkers]):
                return True

        if f_file_path:
            file_path = report.files[int(report.main['location']['file'])]
            if not any([re.match(r'^' + f.replace("*", ".*") + '$',
                                 file_path, re.IGNORECASE)
                        for f in f_file_path]):
                return True

        if 'checker_msg' in args:
            checker_msg = report.main['description']
            if not any([re.match(r'^' + c.replace("*", ".*") + '$',
                                 checker_msg, re.IGNORECASE)
                        for c in args.checker_msg]):
                return True

        return False

    def get_report_dir_results(reportdir):
        all_reports = []
        processed_path_hashes = set()
        for filename in os.listdir(reportdir):
            if filename.endswith(".plist"):
                file_path = os.path.join(reportdir, filename)
                LOG.debug("Parsing: %s", file_path)
                try:
                    files, reports = plist_parser.parse_plist(file_path)
                    for report in reports:
                        path_hash = get_report_path_hash(report, files)
                        if path_hash in processed_path_hashes:
                            LOG.debug("Not showing report because it is a "
                                      "deduplication of an already processed "
                                      "report!")
                            LOG.debug("Path hash: %s", path_hash)
                            LOG.debug(report)
                            continue

                        if skip_report_dir_result(report):
                            continue

                        processed_path_hashes.add(path_hash)
                        report.main['location']['file_name'] = \
                            files[int(report.main['location']['file'])]
                        all_reports.append(report)

                except Exception as ex:
                    LOG.error('The generated plist is not valid!')
                    LOG.error(ex)
        return all_reports

    def get_line_from_file(filename, lineno):
        with open(filename, 'r') as f:
            i = 1
            for line in f:
                if i == lineno:
                    return line
                i += 1
        return ""

    def get_diff_base_results(client, baseids, base_hashes, suppressed_hashes):
        base_results = []
        report_filter = ttypes.ReportFilter()
        add_filter_conditions(client, report_filter, args)

        sort_mode = [(ttypes.SortMode(
            ttypes.SortType.FILENAME,
            ttypes.Order.ASC))]
        limit = constants.MAX_QUERY_SIZE
        offset = 0

        report_filter.reportHash = base_hashes + suppressed_hashes
        results = client.getRunResults(baseids,
                                       limit,
                                       offset,
                                       sort_mode,
                                       report_filter,
                                       None)
        while results:
            base_results.extend(results)
            offset += limit
            results = client.getRunResults(baseids,
                                           limit,
                                           offset,
                                           sort_mode,
                                           report_filter,
                                           None)
        return base_results

    def get_suppressed_reports(reports):
        """
        Returns suppressed reports.
        """
        suppressed_in_code = []
        for rep in reports:
            bughash = rep.report_hash
            source_file = rep.main['location']['file_name']
            bug_line = rep.main['location']['line']
            checker_name = rep.main['check_name']

            sc_handler = SourceCodeCommentHandler(source_file)
            src_comment_data = sc_handler.filter_source_line_comments(
                bug_line,
                checker_name)

            if len(src_comment_data) == 1:
                suppressed_in_code.append(bughash)
                LOG.debug("Bug %s is suppressed in code. file: %s Line %s",
                          bughash, source_file, bug_line)
            elif len(src_comment_data) > 1:
                LOG.warning(
                    "Multiple source code comment can be found "
                    "for '%s' checker in '%s' at line %s. "
                    "This bug will not be suppressed!",
                    checker_name, source_file, bug_line)
        return suppressed_in_code

    def get_diff_type():
        """
        Returns Thrift DiffType value by processing the arguments.
        """
        if 'new' in args:
            return ttypes.DiffType.NEW

        if 'unresolved' in args:
            return ttypes.DiffType.UNRESOLVED

        if 'resolved' in args:
            return ttypes.DiffType.RESOLVED

        return None

    def get_diff_local_dir_remote_run(client, report_dir, run_name):
        """
        Compares a local report directory with a remote run.
        """
        filtered_reports = []
        report_dir_results = get_report_dir_results(
            os.path.abspath(report_dir))
        suppressed_in_code = get_suppressed_reports(report_dir_results)

        diff_type = get_diff_type()
        run_ids, run_names, _ = process_run_arg(run_name)
        local_report_hashes = set([r.report_hash for r in report_dir_results])

        if diff_type == ttypes.DiffType.NEW:
            # Get report hashes which can be found only in the remote runs.
            remote_hashes = \
                client.getDiffResultsHash(run_ids,
                                          local_report_hashes,
                                          ttypes.DiffType.RESOLVED)

            results = get_diff_base_results(client, run_ids,
                                            remote_hashes,
                                            suppressed_in_code)
            for result in results:
                filtered_reports.append(result)
        elif diff_type == ttypes.DiffType.UNRESOLVED:
            # Get remote hashes which can be found in the remote run and in the
            # local report directory.
            remote_hashes = \
                client.getDiffResultsHash(run_ids,
                                          local_report_hashes,
                                          ttypes.DiffType.UNRESOLVED)
            for result in report_dir_results:
                rep_h = result.report_hash
                if rep_h in remote_hashes and rep_h not in suppressed_in_code:
                    filtered_reports.append(result)
        elif diff_type == ttypes.DiffType.RESOLVED:
            # Get remote hashes which can be found in the remote run and in the
            # local report directory.
            remote_hashes = \
                client.getDiffResultsHash(run_ids,
                                          local_report_hashes,
                                          ttypes.DiffType.UNRESOLVED)
            for result in report_dir_results:
                if result.report_hash not in remote_hashes:
                    filtered_reports.append(result)

        return filtered_reports, run_names

    def get_diff_remote_run_local_dir(client, run_name, report_dir):
        """
        Compares a remote run with a local report directory.
        """
        filtered_reports = []
        report_dir_results = get_report_dir_results(
            os.path.abspath(report_dir))
        suppressed_in_code = get_suppressed_reports(report_dir_results)

        diff_type = get_diff_type()
        run_ids, run_names, _ = process_run_arg(run_name)
        local_report_hashes = set([r.report_hash for r in report_dir_results])

        remote_hashes = client.getDiffResultsHash(run_ids,
                                                  local_report_hashes,
                                                  diff_type)

        if diff_type in [ttypes.DiffType.NEW, ttypes.DiffType.UNRESOLVED]:
            # Shows reports from the report dir which are not present in
            # the baseline (NEW reports) or appear in both side (UNRESOLVED
            # reports) and not suppressed in the code.
            for result in report_dir_results:
                rep_h = result.report_hash
                if rep_h in remote_hashes and rep_h not in suppressed_in_code:
                    filtered_reports.append(result)
        elif diff_type == ttypes.DiffType.RESOLVED:
            # Show bugs in the baseline (server) which are not present in
            # the report dir or suppressed.
            results = get_diff_base_results(client,
                                            run_ids,
                                            remote_hashes,
                                            suppressed_in_code)
            for result in results:
                filtered_reports.append(result)

        return filtered_reports, run_names

    def get_diff_remote_runs(client, basename, newname):
        """
        Compares two remote runs and returns the filtered results.
        """
        report_filter = ttypes.ReportFilter()
        add_filter_conditions(client, report_filter, args)

        base_ids, base_run_names, base_run_tags = process_run_arg(basename)
        report_filter.runTag = base_run_tags

        cmp_data = ttypes.CompareData()
        cmp_data.diffType = get_diff_type()

        new_ids, new_run_names, new_run_tags = process_run_arg(newname)
        cmp_data.runIds = new_ids
        cmp_data.runTag = new_run_tags

        # Do not show resolved bugs in compare mode new.
        if cmp_data.diffType == ttypes.DiffType.NEW:
            report_filter.detectionStatus = [
                ttypes.DetectionStatus.NEW,
                ttypes.DetectionStatus.UNRESOLVED,
                ttypes.DetectionStatus.REOPENED]

        sort_mode = [(ttypes.SortMode(
            ttypes.SortType.FILENAME,
            ttypes.Order.ASC))]
        limit = constants.MAX_QUERY_SIZE
        offset = 0

        all_results = []
        results = client.getRunResults(base_ids,
                                       limit,
                                       offset,
                                       sort_mode,
                                       report_filter,
                                       cmp_data)

        while results:
            all_results.extend(results)
            offset += limit
            results = client.getRunResults(base_ids,
                                           limit,
                                           offset,
                                           sort_mode,
                                           report_filter,
                                           cmp_data)
        return all_results, base_run_names, new_run_names

    def get_diff_local_dirs(basename, newname):
        """
        Compares two report directories and returns the filtered results.
        """
        filtered_reports = []
        base_results = get_report_dir_results(os.path.abspath(basename))
        new_results = get_report_dir_results(os.path.abspath(newname))

        base_hashes = set([res.report_hash for res in base_results])
        new_hashes = set([res.report_hash for res in new_results])

        diff_type = get_diff_type()
        if diff_type == ttypes.DiffType.NEW:
            for res in new_results:
                if res.report_hash not in base_hashes:
                    filtered_reports.append(res)
        if diff_type == ttypes.DiffType.UNRESOLVED:
            for res in new_results:
                if res.report_hash in base_hashes:
                    filtered_reports.append(res)
        elif diff_type == ttypes.DiffType.RESOLVED:
            for res in base_results:
                if res.report_hash not in new_hashes:
                    filtered_reports.append(res)

        return filtered_reports

    def cached_report_file_lookup(file_cache, file_id):
        """
        Get source file data for the given file and caches it in a file cache
        if file data is not found in the cache. Finally, it returns the source
        file data from the cache.
        """
        if file_id not in file_cache:
            source = client.getSourceFileData(file_id, True,
                                              ttypes.Encoding.BASE64)
            file_content = base64.b64decode(source.fileContent)
            file_cache[file_id] = {'id': file_id,
                                   'path': source.filePath,
                                   'content': file_content}

        return file_cache[file_id]

    def get_report_data(client, reports, file_cache):
        """
        Returns necessary report files and report data events for the HTML
        plist parser.
        """
        file_sources = {}
        report_data = []

        for report in reports:
            file_sources[report.fileId] = cached_report_file_lookup(
                file_cache, report.fileId)

            details = client.getReportDetails(report.reportId)
            events = []
            for event in details.pathEvents:
                file_sources[event.fileId] = cached_report_file_lookup(
                    file_cache, event.fileId)

                location = {'line': event.startLine,
                            'col': event.startCol,
                            'file': event.fileId}

                events.append({'location': location,
                               'message': event.msg})

            # Get extended data.
            macros = []
            notes = []
            for extended_data in details.extendedData:
                file_sources[extended_data.fileId] = cached_report_file_lookup(
                    file_cache, extended_data.fileId)

                location = {'line': extended_data.startLine,
                            'col': extended_data.startCol,
                            'file': extended_data.fileId}

                if extended_data.type == ttypes.ExtendedReportDataType.MACRO:
                    macros.append({'location': location,
                                   'expansion': event.msg})
                elif extended_data.type == ttypes.ExtendedReportDataType.NOTE:
                    notes.append({'location': location,
                                  'message': event.msg})

            report_data.append({
                'events': events,
                'macros': macros,
                'notes': notes,
                'path': report.checkedFile,
                'reportHash': report.bugHash,
                'checkerName': report.checkerId})

        return {'files': file_sources,
                'reports': report_data}

    def reports_to_report_data(reports):
        """
        Converts reports from Report class from one plist file
        to report data events for the HTML plist parser.
        """
        file_sources = {}
        report_data = []

        for report in reports:
            # Not all report in this list may refer to the same files
            # thus we need to create a single file list with
            # all files from all reports.
            for file_index, file_path in enumerate(report.files):
                if file_index not in file_sources:
                    try:
                        with io.open(file_path, 'r', encoding='UTF-8',
                                     errors='ignore') as source_data:
                            content = source_data.read()
                    except (OSError, IOError):
                        content = file_path + " NOT FOUND."
                    file_sources[file_index] = {'id': file_index,
                                                'path': file_path,
                                                'content': content}

            events = []
            for element in report.bug_path:
                kind = element['kind']
                if kind == 'event':
                    events.append({'location': element['location'],
                                   'message':  element['message']})

            macros = []
            for macro in report.macro_expansions:
                macros.append({'location': macro['location'],
                               'expansion': macro['expansion'],
                               'name': macro['name']})

            notes = []
            for note in report.notes:
                notes.append({'location': note['location'],
                              'message': note['message']})

            report_hash = report.main['issue_hash_content_of_line_in_context']
            report_data.append({
                'events': events,
                'macros': macros,
                'notes': notes,
                'path': report.main['location']['file_name'],
                'reportHash': report_hash,
                'checkerName': report.main['check_name']})

        return {'files': file_sources,
                'reports': report_data}

    def report_to_html(client, reports, output_dir):
        """
        Generate HTML output files for the given reports in the given output
        directory by using the Plist To HTML parser.
        """
        html_builder = PlistToHtml.HtmlBuilder(
            context.path_plist_to_html_dist,
            context.severity_map)

        file_report_map = defaultdict(list)
        for report in reports:
            if isinstance(report, Report):
                file_path = report.main['location']['file_name']
            else:
                file_path = report.checkedFile
            file_report_map[file_path].append(report)

        file_cache = {}
        for file_path, file_reports in file_report_map.items():
            checked_file = file_path
            filename = os.path.basename(checked_file)
            h = int(hashlib.md5(file_path).hexdigest(), 16) % (10 ** 8)

            if isinstance(file_reports[0], Report):
                report_data = reports_to_report_data(file_reports)
            else:
                report_data = get_report_data(client, file_reports, file_cache)

            output_path = os.path.join(output_dir,
                                       filename + '_' + str(h) + '.html')
            html_builder.create(output_path, report_data)
            print('Html file was generated for file://{0}: file://{1}'.format(
                checked_file, output_path))

        html_builder.create_index_html(output_dir)

    def print_reports(client, reports, output_format):
        output_dir = args.export_dir if 'export_dir' in args else None
        if 'clean' in args and os.path.isdir(output_dir):
            print("Previous analysis results in '{0}' have been removed, "
                  "overwriting with current results.".format(output_dir))
            shutil.rmtree(output_dir)

        if output_format == 'json':
            output = []
            for report in reports:
                if isinstance(report, Report):
                    output.append(report.main)
                else:
                    output.append(report)
            print(CmdLineOutputEncoder().encode(output))
            return

        if output_format == 'html':
            output_dir = args.export_dir
            if not os.path.exists(output_dir):
                os.makedirs(output_dir)

            print("Generating HTML output files to file://{0} directory:\n"
                  .format(output_dir))

            report_to_html(client, reports, output_dir)

            print('\nTo view the results in a browser run:\n'
                  '  $ firefox {0}'.format(os.path.join(args.export_dir,
                                                        'index.html')))
            return

        header = ['File', 'Checker', 'Severity', 'Msg', 'Source']
        rows = []

        source_lines = defaultdict(set)
        for report in reports:
            if not isinstance(report, Report) and report.line is not None:
                source_lines[report.fileId].add(report.line)

        lines_in_files_requested = []
        for key in source_lines:
            lines_in_files_requested.append(
                ttypes.LinesInFilesRequested(fileId=key,
                                             lines=source_lines[key]))

        for report in reports:
            source_line = ''
            if isinstance(report, Report):
                # report is coming from a plist file.
                bug_line = report.main['location']['line']
                bug_col = report.main['location']['col']
                checked_file = report.main['location']['file_name']\
                    + ':' + str(bug_line) + ":" + str(bug_col)
                check_name = report.main['check_name']
                sev = context.severity_map.get(check_name)
                check_msg = report.main['description']
                source_line =\
                    get_line_from_file(report.main['location']['file_name'],
                                       bug_line)
            else:
                # report is of ReportData type coming from CodeChecker server.
                bug_line = report.line
                bug_col = report.column
                sev = ttypes.Severity._VALUES_TO_NAMES[report.severity]
                checked_file = report.checkedFile
                if bug_line is not None:
                    checked_file += ':' + str(bug_line) + ":" + str(bug_col)

                check_name = report.checkerId
                check_msg = report.checkerMsg

                if lines_in_files_requested:
                    source_line_contents = client.getLinesInSourceFileContents(
                        lines_in_files_requested, ttypes.Encoding.BASE64)
                    source_line = base64.b64decode(
                        source_line_contents[report.fileId][bug_line])
            rows.append(
                (sev, checked_file, check_msg, check_name, source_line))
        if output_format == 'plaintext':
            for row in rows:
                print("[{0}] {1}: {2} [{3}]\n{4}\n".format(
                    row[0], row[1], row[2], row[3], row[4]))
        else:
            print(twodim_to_str(output_format, header, rows))

    def get_run_tag(client, run_ids, tag_name):
        """
        Returns run tag information for the given tag name in the given runs.
        """
        run_history_filter = ttypes.RunHistoryFilter()
        run_history_filter.tagNames = [tag_name]
        run_histories = client.getRunHistory(run_ids, None, None,
                                             run_history_filter)

        return run_histories[0] if len(run_histories) else None

    def process_run_arg(run_arg_with_tag):
        """
        Process the argument and returns run ids a run tag ids.
        The argument has the following format: <run_name>:<run_tag>
        """
        run_with_tag = run_arg_with_tag.split(':')
        run_name = run_with_tag[0]

        runs = get_runs(client, [run_name])
        run_ids = map(lambda run: run.runId, runs)
        run_names = map(lambda run: run.name, runs)

        # Set base run tag if it is available.
        run_tag_name = run_with_tag[1] if len(run_with_tag) > 1 else None
        run_tags = None
        if run_tag_name:
            tag = get_run_tag(client, run_ids, run_tag_name)
            run_tags = [tag.id] if tag else None

        if not run_ids:
            LOG.warning(
                "No run names match the given pattern: %s", run_arg_with_tag)
            sys.exit(1)

        LOG.info("Matching runs: %s",
                 ', '.join(map(lambda run: run.name, runs)))

        return run_ids, run_names, run_tags

    def print_diff_results(reports):
        """
        Print the results.
        """
        if reports:
            print_reports(client, reports, args.output_format)
        else:
            LOG.info("No results.")

    client = None

    # We set up the client if we are not comparing two local report directory.
    if not os.path.isdir(args.basename) or not os.path.isdir(args.newname):
        client = setup_client(args.product_url)

    if os.path.isdir(args.basename) and os.path.isdir(args.newname):
        reports = get_diff_local_dirs(args.basename, args.newname)
        print_diff_results(reports)
        LOG.info("Compared two local report directories %s and %s",
                 os.path.abspath(args.basename),
                 os.path.abspath(args.newname))
    elif os.path.isdir(args.newname):
        reports, base_run_names = \
            get_diff_remote_run_local_dir(client,
                                          args.basename,
                                          os.path.abspath(args.newname))
        print_diff_results(reports)
        LOG.info("Compared remote run(s) %s (matching: %s) and local report "
                 "directory %s",
                 args.basename,
                 ', '.join(base_run_names),
                 os.path.abspath(args.newname))
    elif os.path.isdir(args.basename):
        reports, new_run_names = \
            get_diff_local_dir_remote_run(client,
                                          os.path.abspath(args.basename),
                                          args.newname)
        print_diff_results(reports)
        LOG.info("Compared local report directory %s and remote run(s) %s "
                 "(matching: %s).",
                 os.path.abspath(args.basename),
                 args.newname,
                 ', '.join(new_run_names))
    else:
        reports, base_run_names, new_run_names = \
            get_diff_remote_runs(client, args.basename, args.newname)
        print_diff_results(reports)
        LOG.info("Compared multiple remote runs %s (matching: %s) and %s "
                 "(matching: %s)",
                 args.basename,
                 ', '.join(base_run_names),
                 args.newname,
                 ', '.join(new_run_names))
예제 #11
0
def server_init_start(args):
    """
    Start or manage a CodeChecker report server.
    """

    if 'list' in args or 'stop' in args or 'stop_all' in args:
        __instance_management(args)
        sys.exit(0)

    if 'reload' in args:
        __reload_config(args)
        sys.exit(0)

    # Actual server starting from this point.
    if not host_check.check_zlib():
        raise Exception("zlib is not available on the system!")

    # WARNING
    # In case of SQLite args.dbaddress default value is used
    # for which the is_localhost should return true.
    if util.is_localhost(args.dbaddress) and \
            not os.path.exists(args.config_directory):
        os.makedirs(args.config_directory)

    # Make sure the SQLite file can be created if it not exists.
    if 'sqlite' in args and \
            not os.path.isdir(os.path.dirname(args.sqlite)):
        os.makedirs(os.path.dirname(args.sqlite))

    if 'reset_root' in args:
        try:
            os.remove(os.path.join(args.config_directory, 'root.user'))
            LOG.info("Master superuser (root) credentials invalidated and "
                     "deleted. New ones will be generated...")
        except OSError:
            # File doesn't exist.
            pass

    if 'force_auth' in args:
        LOG.info("'--force-authentication' was passed as a command-line "
                 "option. The server will ask for users to authenticate!")

    context = webserver_context.get_context()
    context.codechecker_workspace = args.config_directory
    context.db_username = args.dbusername

    check_env = get_check_env(context.path_env_extra,
                              context.ld_lib_path_extra)

    cfg_sql_server = database.SQLServer.from_cmdline_args(
        vars(args), CONFIG_META, context.config_migration_root,
        interactive=True, env=check_env)

    LOG.info("Checking configuration database ...")
    db_status = cfg_sql_server.connect()
    db_status_msg = database_status.db_status_msg.get(db_status)
    LOG.info(db_status_msg)

    if db_status == DBStatus.SCHEMA_MISSING:
        LOG.debug("Config database schema is missing, initializing new.")
        db_status = cfg_sql_server.connect(init=True)
        if db_status != DBStatus.OK:
            LOG.error("Config database initialization failed!")
            LOG.error("Please check debug logs.")
            sys.exit(1)

    if db_status == DBStatus.SCHEMA_MISMATCH_NO:
        LOG.debug("Configuration database schema mismatch.")
        LOG.debug("No schema upgrade is possible.")
        sys.exit(1)

    force_upgrade = True if 'force_upgrade' in args else False

    if db_status == DBStatus.SCHEMA_MISMATCH_OK:
        LOG.debug("Configuration database schema mismatch.")
        LOG.debug("Schema upgrade is possible.")
        LOG.warning("Please note after migration only "
                    "newer CodeChecker versions can be used "
                    "to start the server")
        LOG.warning("It is advised to make a full backup of your "
                    "configuration database")

        LOG.warning(cfg_sql_server.get_db_location())

        question = 'Do you want to upgrade to the new schema?' \
                   ' Y(es)/n(o) '
        if force_upgrade or util.get_user_input(question):
            print("Upgrading schema ...")
            ret = cfg_sql_server.upgrade()
            msg = database_status.db_status_msg.get(
                ret, 'Unknown database status')
            print(msg)
            if ret != DBStatus.OK:
                LOG.error("Schema migration failed")
                sys.exit(ret)
        else:
            LOG.info("No schema migration was done.")
            sys.exit(0)

    if db_status == DBStatus.MISSING:
        LOG.error("Missing configuration database.")
        LOG.error("Server can not be started.")
        sys.exit(1)

    # Configuration database setup and check is needed before database
    # statuses can be checked.
    try:
        if args.status:
            ret = __db_status_check(cfg_sql_server, context, args.status)
            sys.exit(ret)
    except AttributeError:
        LOG.debug('Status was not in the arguments.')

    try:
        if args.product_to_upgrade:
            ret = __db_migration(cfg_sql_server, context,
                                 args.product_to_upgrade, force_upgrade)
            sys.exit(ret)
    except AttributeError:
        LOG.debug('Product upgrade was not in the arguments.')

    # Create the main database link from the arguments passed over the
    # command line.
    cfg_dir = os.path.abspath(args.config_directory)
    default_product_path = os.path.join(cfg_dir, 'Default.sqlite')
    create_default_product = 'sqlite' in args and \
                             not os.path.exists(default_product_path)

    if create_default_product:
        # Create a default product and add it to the configuration database.

        LOG.debug("Create default product...")
        LOG.debug("Configuring schema and migration...")

        prod_server = database.SQLiteDatabase(
            default_product_path, RUN_META,
            context.run_migration_root, check_env)

        LOG.debug("Checking 'Default' product database.")
        db_status = prod_server.connect()
        if db_status != DBStatus.MISSING:
            db_status = prod_server.connect(init=True)
            LOG.debug(database_status.db_status_msg.get(db_status))
            if db_status != DBStatus.OK:
                LOG.error("Failed to configure default product")
                sys.exit(1)

        product_conn_string = prod_server.get_connection_string()

        server.add_initial_run_database(
            cfg_sql_server, product_conn_string)

        LOG.info("Product 'Default' at '%s' created and set up.",
                 default_product_path)

    prod_statuses = check_product_db_status(cfg_sql_server, context)

    upgrade_available = {}
    for k, v in prod_statuses.items():
        db_status, _, _, _ = v
        if db_status == DBStatus.SCHEMA_MISMATCH_OK or \
                db_status == DBStatus.SCHEMA_MISSING:
            upgrade_available[k] = v

    if upgrade_available:
        print_prod_status(prod_statuses)
        LOG.warning("Multiple products can be upgraded, make a backup!")
        __db_migration(cfg_sql_server, context, 'all', force_upgrade)

    prod_statuses = check_product_db_status(cfg_sql_server, context)
    print_prod_status(prod_statuses)

    non_ok_db = False
    for k, v in prod_statuses.items():
        db_status, _, _, _ = v
        if db_status != DBStatus.OK:
            non_ok_db = True
        break

    if non_ok_db:
        msg = "There are some database issues. " \
              "Do you want to start the " \
              "server? Y(es)/n(o) "
        if not util.get_user_input(msg):
            sys.exit(1)

    # Start database viewer.
    checker_md_docs = os.path.join(context.doc_root, 'checker_md_docs')
    checker_md_docs_map = os.path.join(checker_md_docs,
                                       'checker_doc_map.json')
    checker_md_docs_map = util.load_json_or_empty(checker_md_docs_map, {})

    package_data = {'www_root': context.www_root,
                    'doc_root': context.doc_root,
                    'checker_md_docs': checker_md_docs,
                    'checker_md_docs_map': checker_md_docs_map,
                    'version': context.package_git_tag}

    suppr_handler = suppress_handler. \
        GenericSuppressHandler(None, False)

    try:
        server.start_server(args.config_directory,
                            package_data,
                            args.view_port,
                            cfg_sql_server,
                            suppr_handler,
                            args.listen_address,
                            'force_auth' in args,
                            args.skip_db_cleanup,
                            context,
                            check_env)
    except socket.error as err:
        if err.errno == errno.EADDRINUSE:
            LOG.error("Server can't be started, maybe the given port number "
                      "(%s) is already used. Check the connection "
                      "parameters.", args.view_port)
            sys.exit(1)
        else:
            raise