Repository: spark Updated Branches: refs/heads/master c4c4b07bf -> d14df06c0
[SPARK-6191] [EC2] Generalize ability to download libs Right now we have a method to specifically download boto. This PR generalizes it so it's easy to download additional libraries if we want. For example, adding new external libraries for spark-ec2 is now as simple as: ```python external_libs = [ { "name": "boto", "version": "2.34.0", "md5": "5556223d2d0cc4d06dd4829e671dcecd" }, { "name": "PyYAML", "version": "3.11", "md5": "f50e08ef0fe55178479d3a618efe21db" }, { "name": "argparse", "version": "1.3.0", "md5": "9bcf7f612190885c8c85e30ba41db3c7" } ] ``` Likely use cases: * Downloading PyYAML to allow spark-ec2 configs to be persisted as a YAML file. ([SPARK-925](https://issues.apache.org/jira/browse/SPARK-925)) * Downloading argparse to clean up / modernize our option parsing. First run output, with PyYAML and argparse added just for demonstration purposes: ```shell $ ./spark-ec2 --version Downloading external libraries that spark-ec2 needs from PyPI to /path/to/spark/ec2/lib... This should be a one-time operation. - Downloading boto... - Finished downloading boto. - Downloading PyYAML... - Finished downloading PyYAML. - Downloading argparse... - Finished downloading argparse. spark-ec2 1.2.1 ``` Output thereafter: ```shell $ ./spark-ec2 --version spark-ec2 1.2.1 ``` Author: Nicholas Chammas <nicholas.cham...@gmail.com> Closes #4919 from nchammas/setup-ec2-libs and squashes the following commits: a077955 [Nicholas Chammas] print default region c95fb7d [Nicholas Chammas] to docstring 5448845 [Nicholas Chammas] remove libs added for demo purposes 60d8c23 [Nicholas Chammas] generalize ability to download libs Project: http://git-wip-us.apache.org/repos/asf/spark/repo Commit: http://git-wip-us.apache.org/repos/asf/spark/commit/d14df06c Tree: http://git-wip-us.apache.org/repos/asf/spark/tree/d14df06c Diff: http://git-wip-us.apache.org/repos/asf/spark/diff/d14df06c Branch: refs/heads/master Commit: d14df06c05a6228fd6522914c39aa75898eddfc1 Parents: c4c4b07 Author: Nicholas Chammas <nicholas.cham...@gmail.com> Authored: Tue Mar 10 10:58:31 2015 +0000 Committer: Sean Owen <so...@cloudera.com> Committed: Tue Mar 10 10:58:31 2015 +0000 ---------------------------------------------------------------------- ec2/spark_ec2.py | 82 +++++++++++++++++++++++++++++++++------------------ 1 file changed, 54 insertions(+), 28 deletions(-) ---------------------------------------------------------------------- http://git-wip-us.apache.org/repos/asf/spark/blob/d14df06c/ec2/spark_ec2.py ---------------------------------------------------------------------- diff --git a/ec2/spark_ec2.py b/ec2/spark_ec2.py index b50b381..3acb5fe 100755 --- a/ec2/spark_ec2.py +++ b/ec2/spark_ec2.py @@ -70,34 +70,60 @@ DEFAULT_SPARK_EC2_GITHUB_REPO = "https://github.com/mesos/spark-ec2" DEFAULT_SPARK_EC2_BRANCH = "branch-1.3" -def setup_boto(): - # Download Boto if it's not already present in the SPARK_EC2_DIR/lib folder: - version = "boto-2.34.0" - md5 = "5556223d2d0cc4d06dd4829e671dcecd" - url = "https://pypi.python.org/packages/source/b/boto/%s.tar.gz" % version - lib_dir = os.path.join(SPARK_EC2_DIR, "lib") - if not os.path.exists(lib_dir): - os.mkdir(lib_dir) - boto_lib_dir = os.path.join(lib_dir, version) - if not os.path.isdir(boto_lib_dir): - tgz_file_path = os.path.join(lib_dir, "%s.tar.gz" % version) - print "Downloading Boto from PyPi" - download_stream = urllib2.urlopen(url) - with open(tgz_file_path, "wb") as tgz_file: - tgz_file.write(download_stream.read()) - with open(tgz_file_path) as tar: - if hashlib.md5(tar.read()).hexdigest() != md5: - print >> stderr, "ERROR: Got wrong md5sum for Boto" - sys.exit(1) - tar = tarfile.open(tgz_file_path) - tar.extractall(path=lib_dir) - tar.close() - os.remove(tgz_file_path) - print "Finished downloading Boto" - sys.path.insert(0, boto_lib_dir) +def setup_external_libs(libs): + """ + Download external libraries from PyPI to SPARK_EC2_DIR/lib/ and prepend them to our PATH. + """ + PYPI_URL_PREFIX = "https://pypi.python.org/packages/source" + SPARK_EC2_LIB_DIR = os.path.join(SPARK_EC2_DIR, "lib") + + if not os.path.exists(SPARK_EC2_LIB_DIR): + print "Downloading external libraries that spark-ec2 needs from PyPI to {path}...".format( + path=SPARK_EC2_LIB_DIR + ) + print "This should be a one-time operation." + os.mkdir(SPARK_EC2_LIB_DIR) + + for lib in libs: + versioned_lib_name = "{n}-{v}".format(n=lib["name"], v=lib["version"]) + lib_dir = os.path.join(SPARK_EC2_LIB_DIR, versioned_lib_name) + + if not os.path.isdir(lib_dir): + tgz_file_path = os.path.join(SPARK_EC2_LIB_DIR, versioned_lib_name + ".tar.gz") + print " - Downloading {lib}...".format(lib=lib["name"]) + download_stream = urllib2.urlopen( + "{prefix}/{first_letter}/{lib_name}/{lib_name}-{lib_version}.tar.gz".format( + prefix=PYPI_URL_PREFIX, + first_letter=lib["name"][:1], + lib_name=lib["name"], + lib_version=lib["version"] + ) + ) + with open(tgz_file_path, "wb") as tgz_file: + tgz_file.write(download_stream.read()) + with open(tgz_file_path) as tar: + if hashlib.md5(tar.read()).hexdigest() != lib["md5"]: + print >> stderr, "ERROR: Got wrong md5sum for {lib}.".format(lib=lib["name"]) + sys.exit(1) + tar = tarfile.open(tgz_file_path) + tar.extractall(path=SPARK_EC2_LIB_DIR) + tar.close() + os.remove(tgz_file_path) + print " - Finished downloading {lib}.".format(lib=lib["name"]) + sys.path.insert(1, lib_dir) + + +# Only PyPI libraries are supported. +external_libs = [ + { + "name": "boto", + "version": "2.34.0", + "md5": "5556223d2d0cc4d06dd4829e671dcecd" + } +] +setup_external_libs(external_libs) -setup_boto() import boto from boto.ec2.blockdevicemapping import BlockDeviceMapping, BlockDeviceType, EBSBlockDeviceType from boto import ec2 @@ -136,7 +162,7 @@ def parse_args(): help="Master instance type (leave empty for same as instance-type)") parser.add_option( "-r", "--region", default="us-east-1", - help="EC2 region used to launch instances in, or to find them in") + help="EC2 region used to launch instances in, or to find them in (default: %default)") parser.add_option( "-z", "--zone", default="", help="Availability zone to launch instances in, or 'all' to spread " + @@ -230,7 +256,7 @@ def parse_args(): "(e.g -Dspark.worker.timeout=180)") parser.add_option( "--user-data", type="string", default="", - help="Path to a user-data file (most AMI's interpret this as an initialization script)") + help="Path to a user-data file (most AMIs interpret this as an initialization script)") parser.add_option( "--authorized-address", type="string", default="0.0.0.0/0", help="Address to authorize on created security groups (default: %default)") --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@spark.apache.org For additional commands, e-mail: commits-h...@spark.apache.org