Question

我正在尝试填充一个字典，以便稍后在另一个循环中的系统表的SQL查询生成的循环中使用。当我在嵌套字典中只有2个维度时，我没有遇到任何问题，但我需要再添加2个维度，而且我不确定如何从中填充字典。

我有4个维度：工作室，节点（此数字可能因工作室而异），音量（这是＆＃34;路径＆＃34;在查询中也可能有所不同）和存储（这可能会有所不同）。我希望输出类似于以下内容：

{'studio1': {u'v_studio1_node0001': {u'/vertica/data': {u'CATALOG': n, u'DATA,TEMP': n}, u'/vertica/data2': {u'DATA,TEMP': n}}, u'v_studio1_node0002': {u'/vertica/data': {u'CATALOG': n, u'DATA,TEMP': n}, u'/vertica/data2': {u'DATA,TEMP': n}}, u'v_studio1_node0003': {u'/vertica/data': {u'CATALOG': n, u'DATA,TEMP': n}, u'/vertica/data2': {u'DATA,TEMP': n}}}, 'studio2': {u'v_studio2_node0001': {u'/vertica/data': {u'CATALOG': n, u'DATA,TEMP': n}}, u'v_studio2_node0002': {u'/vertica/data': {u'CATALOG': n, u'DATA,TEMP': n}}}}

以下只有studio和node_name，它是我在字典studio_node_names中使用的2个维度。我添加了studio_node_volume我希望填充所有4个维度，然后将数字作为值。

#Create dictionary to hold studio names and node names for each studio to be used for creating alarms
studio_node_names = {}

# Create dictionary to hold studio names, node names, volume, and storage for each studio to be used for creating alarms
#studio_node_volume = {}

for connection_blob in STUDIO_CONNECTION_INFO:
    studio = connection_blob['studio']
    connection_info = connection_blob['connection_info']

    try:
        conn = vertica_connect(connection_info)

        # Query the v_monitor tables to get metrics for alarms
        try:
            disk_usage_result = conn.query("select su.node_name, su.path, du.storage_usage, su.usage_percent from v_monitor.storage_usage su join v_monitor.disk_storage du on INSTR(CONCAT(du.storage_path, '/'), CONCAT(su.path, '/'), 1, 1) = 1 and du.node_name = su.node_name where su.path not in('', '/') and (du.storage_usage = 'CATALOG' or du.storage_usage like '%DATA%')")
        except (BotoClientError, BotoServerError, Exception) as e:
            print "ERROR: Vertica Query exception message"

        # Update the metrics derived from the row data returned from the query "disk_usage_result".
        for row in disk_usage_result.rows:

            # Update the cluster Disk Usage metric.
            disk_usage = row['usage_percent']
            node_name = row['node_name']
            #volume = row['path']
            #storage = row['storage_usage']

            if studio_node_names.get(studio):  # For looping over to create DiskUsage alarms later
                studio_node_names[studio].append(node_name)
            else:
                studio_node_names[studio] = [node_name]

    except (BotoClientError, BotoServerError, Exception) as e:
        print "ERROR: Vertica Connection exception message"
    finally:
        if conn:
            conn.close()

print "INFO: Done checks of Vertica instances."

for connection_blob in STUDIO_CONNECTION_INFO:
    studio = connection_blob['studio']

    # Create the DiskUsage alarm.
    for node_name in studio_node_names[studio]:

    try:
        # Alarm code here
    except (BotoClientError, BotoServerError, Exception) as e:
        print "ERROR: DiskUsage create_alarm exception message"

＆＃34;创建DiskUsage Alarms＆＃34;的另一个connection_blob循环我将在studio_node_volume上循环，为每个工作室，节点，卷，存储组合创建警报。

Answer 1

所以，我能够使用以下内容创建我期望的字典： if studio not in studio_node_volume: studio_node_volume[studio] = {} if node_name not in studio_node_volume[studio]: studio_node_volume[studio][node_name] = {} if volume not in studio_node_volume[studio][node_name]: studio_node_volume[studio][node_name][volume] = {} if storage not in studio_node_volume[studio][node_name][volume]: studio_node_volume[studio][node_name][volume][storage] = {} studio_node_volume[studio][node_name][volume][storage] = disk_usage

Answer 2

我实际上是用另一种方式创建了字典，所以我可以初始化键。我在for connection_blob in STUDIO_CONNECTION_INFO:循环之前添加了以下内容：

class D(dict): 
    def __missing__(self, key): 
        result = self[key] = D() 
        return result 

studio_node_volume = D()

然后在disk_usage_result.rows中的`for row下：＆＃39;循环并在我拥有的行的四个变量之后：

        studio_node_volume[studio]
        studio_node_volume[studio][node_name][volume]
        studio_node_volume[studio][node_name][volume][storage]
        studio_node_volume[studio][node_name][volume][storage] = disk_usage

在循环内从SQL查询更新嵌套字典

2 个答案: