diff options
Diffstat (limited to 'resources/tools/presentation/input_data_files.py')
| -rw-r--r-- | resources/tools/presentation/input_data_files.py | 394 |
1 files changed, 185 insertions, 209 deletions
diff --git a/resources/tools/presentation/input_data_files.py b/resources/tools/presentation/input_data_files.py index d81f64fbe6..cde6d1acc4 100644 --- a/resources/tools/presentation/input_data_files.py +++ b/resources/tools/presentation/input_data_files.py @@ -1,4 +1,4 @@ -# Copyright (c) 2017 Cisco and/or its affiliates. +# Copyright (c) 2018 Cisco and/or its affiliates. # Licensed 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: @@ -16,12 +16,9 @@ Download all data. """ import re -import gzip -import logging -from os import rename, remove -from os.path import join, getsize -from shutil import move +from os import rename, mkdir +from os.path import join from zipfile import ZipFile, is_zipfile, BadZipfile from httplib import responses from requests import get, codes, RequestException, Timeout, TooManyRedirects, \ @@ -39,216 +36,195 @@ SEPARATOR = "__" REGEX_RELEASE = re.compile(r'(\D*)(\d{4}|master)(\D*)') -def download_data_files(spec): - """Download all data specified in the specification file in the section - type: input --> builds. +def _download_file(url, file_name, log): + """Download a file with input data. - :param spec: Specification. + :param url: URL to the file to download. + :param file_name: Name of file to download. + :param log: List of log messages. + :type url: str + :type file_name: str + :type log: list of tuples (severity, msg) + :returns: True if the download was successful, otherwise False. + :rtype: bool + """ + + success = False + try: + log.append(("INFO", " Connecting to '{0}' ...".format(url))) + + response = get(url, stream=True) + code = response.status_code + + log.append(("INFO", " {0}: {1}".format(code, responses[code]))) + + if code != codes["OK"]: + return False + + log.append(("INFO", " Downloading the file '{0}' to '{1}' ...". + format(url, file_name))) + + file_handle = open(file_name, "wb") + for chunk in response.iter_content(chunk_size=CHUNK_SIZE): + if chunk: + file_handle.write(chunk) + file_handle.close() + success = True + except ConnectionError as err: + log.append(("ERROR", "Not possible to connect to '{0}'.".format(url))) + log.append(("DEBUG", str(err))) + except HTTPError as err: + log.append(("ERROR", "Invalid HTTP response from '{0}'.".format(url))) + log.append(("DEBUG", str(err))) + except TooManyRedirects as err: + log.append(("ERROR", "Request exceeded the configured number " + "of maximum re-directions.")) + log.append(("DEBUG", str(err))) + except Timeout as err: + log.append(("ERROR", "Request timed out.")) + log.append(("DEBUG", str(err))) + except RequestException as err: + log.append(("ERROR", "Unexpected HTTP request exception.")) + log.append(("DEBUG", str(err))) + except (IOError, ValueError, KeyError) as err: + log.append(("ERROR", "Download failed.")) + log.append(("DEBUG", str(err))) + + log.append(("INFO", " Download finished.")) + return success + + +def _unzip_file(spec, build, pid, log): + """Unzip downloaded source file. + + :param spec: Specification read form the specification file. + :param build: Information about the build. + :param log: List of log messages. :type spec: Specification - :raises: PresentationError if there is no url defined for the job. + :type build: dict + :type log: list of tuples (severity, msg) + :returns: True if the download was successful, otherwise False. + :rtype: bool """ - for job, builds in spec.builds.items(): - for build in builds: - if job.startswith("csit-"): - if spec.input["file-name"].endswith(".zip"): - url = spec.environment["urls"]["URL[JENKINS,CSIT]"] - elif spec.input["file-name"].endswith(".gz"): - url = spec.environment["urls"]["URL[NEXUS,LOG]"] - else: - logging.error("Not supported file format.") - continue - elif job.startswith("hc2vpp-"): - url = spec.environment["urls"]["URL[JENKINS,HC]"] - else: - raise PresentationError("No url defined for the job '{}'.". - format(job)) - file_name = spec.input["file-name"] - full_name = spec.input["download-path"].\ - format(job=job, build=build["build"], filename=file_name) - url = "{0}/{1}".format(url, full_name) - new_name = join( - spec.environment["paths"]["DIR[WORKING,DATA]"], - "{job}{sep}{build}{sep}{name}".format(job=job, - sep=SEPARATOR, - build=build["build"], - name=file_name)) - logging.info( - "Downloading the file '{0}' to '{1}' ...".format(url, new_name)) - - status = "failed" - try: - response = get(url, stream=True) - code = response.status_code - - if code != codes["OK"]: - logging.warning( - "Jenkins: {0}: {1}.".format(code, responses[code])) - logging.info("Trying to download from Nexus:") - spec.set_input_state(job, build["build"], "not found") - if code == codes["not_found"]: - release = re.search(REGEX_RELEASE, job).group(2) - nexus_file_name = "{job}{sep}{build}{sep}{name}".\ - format(job=job, sep=SEPARATOR, build=build["build"], - name=file_name) - try: - release = "rls{0}".format(int(release)) - except ValueError: - pass - url = "{url}/{release}/{dir}/{file}".\ - format(url=spec.environment["urls"]["URL[NEXUS]"], - release=release, - dir=spec.environment["urls"]["DIR[NEXUS]"], - file=nexus_file_name) - logging.info("Downloading the file '{0}' to '{1}' ...". - format(url, new_name)) - response = get(url, stream=True) - code = response.status_code - if code != codes["OK"]: - logging.error( - "Nexus: {0}: {1}".format(code, responses[code])) - spec.set_input_state( - job, build["build"], "not found") - continue - - file_handle = open(new_name, "wb") - for chunk in response.iter_content(chunk_size=CHUNK_SIZE): - if chunk: - file_handle.write(chunk) - file_handle.close() - - if spec.input["file-name"].endswith(".zip"): - expected_length = None - try: - expected_length = int(response. - headers["Content-Length"]) - logging.debug(" Expected file size: {0}B". - format(expected_length)) - except KeyError: - logging.debug(" No information about expected size.") - - real_length = getsize(new_name) - logging.debug(" Downloaded size: {0}B".format(real_length)) - - if expected_length: - if real_length == expected_length: - status = "downloaded" - logging.info("{0}: {1}".format(code, - responses[code])) - else: - logging.error("The file size differs from the " - "expected size.") - else: - status = "downloaded" - logging.info("{0}: {1}".format(code, responses[code])) - - elif spec.input["file-name"].endswith(".gz"): - if "docs.fd.io" in url: - execute_command("gzip --decompress --keep --force {0}". - format(new_name)) - else: - rename(new_name, new_name[:-3]) - execute_command("gzip --keep {0}".format(new_name[:-3])) - new_name = new_name[:-3] - status = "downloaded" - logging.info("{0}: {1}".format(code, responses[code])) - - except ConnectionError as err: - logging.error("Not possible to connect to '{0}'.".format(url)) - logging.debug(err) - except HTTPError as err: - logging.error("Invalid HTTP response from '{0}'.".format(url)) - logging.debug(err) - except TooManyRedirects as err: - logging.error("Request exceeded the configured number " - "of maximum re-directions.") - logging.debug(err) - except Timeout as err: - logging.error("Request timed out.") - logging.debug(err) - except RequestException as err: - logging.error("Unexpected HTTP request exception.") - logging.debug(err) - except (IOError, ValueError, KeyError) as err: - logging.error("Download failed.") - logging.debug("Reason: {0}".format(err)) - - spec.set_input_state(job, build["build"], status) - spec.set_input_file_name(job, build["build"], new_name) - - if status == "failed": - logging.info("Removing the file '{0}'".format(new_name)) - try: - remove(new_name) - except OSError as err: - logging.warning(str(err)) - spec.set_input_file_name(job, build["build"], None) - - unzip_files(spec) - - -def unzip_files(spec): - """Unzip downloaded zip files - - :param spec: Specification. + data_file = spec.input["extract"] + file_name = build["file-name"] + directory = spec.environment["paths"]["DIR[WORKING,DATA]"] + tmp_dir = join(directory, str(pid)) + try: + mkdir(tmp_dir) + except OSError: + pass + new_name = "{0}{1}{2}".format(file_name.rsplit('.')[-2], + SEPARATOR, + data_file.split("/")[-1]) + + log.append(("INFO", " Unzipping: '{0}' from '{1}'.". + format(data_file, file_name))) + try: + with ZipFile(file_name, 'r') as zip_file: + zip_file.extract(data_file, tmp_dir) + log.append(("INFO", " Renaming the file '{0}' to '{1}'". + format(join(tmp_dir, data_file), new_name))) + rename(join(tmp_dir, data_file), new_name) + build["file-name"] = new_name + return True + except (BadZipfile, RuntimeError) as err: + log.append(("ERROR", "Failed to unzip the file '{0}': {1}.". + format(file_name, str(err)))) + return False + except OSError as err: + log.append(("ERROR", "Failed to rename the file '{0}': {1}.". + format(data_file, str(err)))) + return False + + +def download_and_unzip_data_file(spec, job, build, pid, log): + """Download and unzip a source file. + + :param spec: Specification read form the specification file. + :param job: Name of the Jenkins job. + :param build: Information about the build. + :param pid: PID of the process executing this method. + :param log: List of log messages. :type spec: Specification - :raises: PresentationError if the zip file does not exist or it is not a - zip file. + :type job: str + :type build: dict + :type pid: int + :type log: list of tuples (severity, msg) + :returns: True if the download was successful, otherwise False. + :rtype: bool """ - if spec.is_debug: - data_file = spec.debug["extract"] + if job.startswith("csit-"): + if spec.input["file-name"].endswith(".zip"): + url = spec.environment["urls"]["URL[JENKINS,CSIT]"] + elif spec.input["file-name"].endswith(".gz"): + url = spec.environment["urls"]["URL[NEXUS,LOG]"] + else: + log.append(("ERROR", "Not supported file format.")) + return False + elif job.startswith("hc2vpp-"): + url = spec.environment["urls"]["URL[JENKINS,HC]"] else: - data_file = spec.input["extract"] - - for job, builds in spec.builds.items(): - for build in builds: - if build["status"] == "failed" or build["status"] == "not found": - continue + raise PresentationError("No url defined for the job '{}'.". + format(job)) + file_name = spec.input["file-name"] + full_name = spec.input["download-path"]. \ + format(job=job, build=build["build"], filename=file_name) + url = "{0}/{1}".format(url, full_name) + new_name = join(spec.environment["paths"]["DIR[WORKING,DATA]"], + "{job}{sep}{build}{sep}{name}". + format(job=job, sep=SEPARATOR, build=build["build"], + name=file_name)) + # Download the file from the defined source (Jenkins, logs.fd.io): + success = _download_file(url, new_name, log) + + if success and new_name.endswith(".zip"): + if not is_zipfile(new_name): + success = False + + # If not successful, download from docs.fd.io: + if not success: + log.append(("INFO", " Trying to download from https://docs.fd.io:")) + release = re.search(REGEX_RELEASE, job).group(2) + for rls in (release, "master"): + nexus_file_name = "{job}{sep}{build}{sep}{name}". \ + format(job=job, sep=SEPARATOR, build=build["build"], + name=file_name) try: - status = "failed" - directory = spec.environment["paths"]["DIR[WORKING,DATA]"] - file_name = join(build["file-name"]) - - if build["status"] == "downloaded": - logging.info("Unziping: '{0}' from '{1}'.". - format(data_file, file_name)) - new_name = "{0}{1}{2}".format(file_name.rsplit('.')[-2], - SEPARATOR, - data_file.split("/")[-1]) - try: - if is_zipfile(file_name): - with ZipFile(file_name, 'r') as zip_file: - zip_file.extract(data_file, directory) - logging.info("Moving {0} to {1} ...". - format(join(directory, data_file), - directory)) - move(join(directory, data_file), directory) - logging.info("Renaming the file '{0}' to '{1}'". - format(join(directory, - data_file.split("/")[-1]), - new_name)) - rename(join(directory, data_file.split("/")[-1]), - new_name) - spec.set_input_file_name(job, build["build"], - new_name) - status = "unzipped" - spec.set_input_state(job, build["build"], status) - except (BadZipfile, RuntimeError) as err: - logging.error("Failed to unzip the file '{0}': {1}.". - format(file_name, str(err))) - except OSError as err: - logging.error("Failed to rename the file '{0}': {1}.". - format(data_file, str(err))) - finally: - if status == "failed": - spec.set_input_file_name(job, build["build"], None) - else: - raise PresentationError("The file '{0}' does not exist or " - "it is not a zip file". - format(file_name)) - - spec.set_input_state(job, build["build"], status) - - except KeyError: + rls = "rls{0}".format(int(rls)) + except ValueError: pass + url = "{url}/{release}/{dir}/{file}". \ + format(url=spec.environment["urls"]["URL[NEXUS]"], + release=rls, + dir=spec.environment["urls"]["DIR[NEXUS]"], + file=nexus_file_name) + success = _download_file(url, new_name, log) + if success: + break + + if success: + build["file-name"] = new_name + else: + return False + + if spec.input["file-name"].endswith(".gz"): + if "docs.fd.io" in url: + execute_command("gzip --decompress --keep --force {0}". + format(new_name)) + else: + rename(new_name, new_name[:-3]) + execute_command("gzip --keep {0}".format(new_name[:-3])) + build["file-name"] = new_name[:-3] + + if new_name.endswith(".zip"): + if is_zipfile(new_name): + return _unzip_file(spec, build, pid, log) + else: + log.append(("ERROR", + "Zip file '{0}' is corrupted.".format(new_name))) + return False + else: + return True |