Exemplo n.º 1
0
 def test_default_logging(self, capsys):
     cfg = ConfigStruct('/home/mycfg', 'options')
     cfg.configure_basic_logging('me')
     main_logger = logging.getLogger('me')
     child_logger = main_logger.getChild('runt')
     other_logger = logging.getLogger('stranger')
     main_logger.info('main info')
     child_logger.info('child info')
     other_logger.info('other info')
     other_logger.warn('other warn')
     out, err = capsys.readouterr()
     assert('main info' in err and
            'child info' in err and
            'other warn' in err and
            not 'other info' in err)
Exemplo n.º 2
0
def main(config_file, region, log_level, log_file, concurrency,
         selection_string, accumulation_string, reduction_string,
         command, s3_uri):
    '''Perform simple listing, collating, or deleting of many S3 objects at the same time.

    Examples:

      \b
      List empty objects:
          s3workers list --select 'size == 0' s3://mybucket/myprefix
      \b
      Report total of all non-empty objects:
          s3workers list --select 'size > 0' --reduce 'accumulator += size' s3://mybucket/myprefix
      \b
      Total size group by MD5:
          s3workers list --accumulator '{}' --reduce 'v=accumulator.get(md5,0)+size; accumulator[md5]=v' s3://mybucket/myprefix
    '''  # noqa: E501

    config = ConfigStruct(config_file, 'options', options=DEFAULTS)
    opts = config.options
    # let command line options have temporary precedence if provided values
    opts.might_prefer(region=region, log_level=log_level, log_file=log_file,
                      concurrency=concurrency)
    config.configure_basic_logging(__name__)
    logger = logging.getLogger(__name__)

    s3_uri = re.sub(r'^(s3:)?/+', '', s3_uri)
    items = s3_uri.split('/', 1)
    bucket_name = items[0]
    prefix = items[1] if len(items) > 1 else ''

    conn = s3.connect_to_region(opts.region) if opts.region else connect_s3()
    bucket = conn.get_bucket(bucket_name)

    progress = S3KeyProgress()
    reducer = None

    if reduction_string:
        reducer = Reducer(reduction_string, accumulation_string)

        def key_dumper(key):
            accumulator = reducer.reduce(key.name, key.size, key.md5, key.last_modified)
            progress.write('%s %10d %s %s => %s',
                           key.last_modified, key.size, key.md5, key.name, accumulator)
    else:
        def key_dumper(key):
            progress.write('%s %10d %s %s', key.last_modified, key.size, key.md5, key.name)

    def key_deleter(key):
        progress.write('DELETING: %s %10d %s %s', key.last_modified, key.size, key.md5, key.name)
        key.delete()

    selector = compile(selection_string, '<select>', 'eval') if selection_string else None
    handler = key_deleter if command == 'delete' else key_dumper

    manager = Manager(opts.concurrency)
    manager.start_workers()

    logger.info('Preparing %d jobs for %d workers', len(SHARDS), manager.worker_count)

    # break up jobs into single char prefix jobs
    for shard in SHARDS:
        manager.add_work(S3ListJob(bucket, prefix + shard, selector, handler, progress.report))

    manager.wait_for_workers()
    progress.finish()

    if reducer:
        click.echo('Final accumulator value: ' + str(reducer.accumulator))