个人中心
文章发布
退出
作者热门文章
- html - 出于某种原因，IE8 对我的 Sass 文件中继承的 html5 CSS 不友好？
- JMeter 在响应断言中使用 span 标签的问题
- html - 在 :hover and :active? 上具有不同效果的 CSS 动画
- html - 相对于居中的 html 内容固定的 CSS 重复背景？
29
4
我正在开发一种同步机制，将数据从 aws redshift 移动到 aurora。为了降低网络 I/O 的负载，我正在转换 redshift 上的查询并为它们添加一个校验和查询，这样我将只导出已更改的记录。我用另一个 Select * 查询包装基本查询并添加校验和函数。
我尝试删除一些内部子查询，之后新查询开始运行，但我无法更改查询，因为它是由其他平台提供的。
基本查询:
select x.player_id as playerid,
p.player_nickname,
r.region_code,
s.title as season_name,
x.rating as ranking_score,
x.rank_no as rank_no,
x.rank_no_change as rank_no_change,
x.game_mode,
to_char(ga.total_score, 'FM9D00') as gyo_perf_total_score,
to_char(agressive_score*100/aggresive_weight, 'FM9D000') as gyo_perf_aggressive_score,
to_char(defensive_score*100/defensive_weight, 'FM9D000') as gyo_perf_defensive_score,
to_char(survival_score*100/survival_weight, 'FM9D000') as gyo_perf_survival_score,
ga.match_place_avg,
rounds_played,
kills,
(kills*1.0)/rounds_played as avg_kills_per_round,
assists,
(assists*1.0)/rounds_played as avg_assists_per_round,
headshot_kills,
top10s,
(top10s*1.0)/rounds_played as top10s_ratio,
wins,
(wins*1.0)/rounds_played as win_ratio,
case when top10s = 0 then 0 else (wins*1.0)/(top10s*1.0) end as win_to_top10_ratio,
losses,
ga.match_group as ___matchgroup,
ga.last_update_id as ___lastupdateid_ga,
___lastupdateid_st,
x.last_update_id as ___lastupdateid_rn
from hd_stats.calc_player_ranking x
inner join hd_core.core_player p on x.player_id = p.player_id
and p.player_game_id =
(select g.game_id
from hd_core.core_game g
where lower(g.game_short_title) = 'xxxx')
left join hd_core.core_region r on x.region_id = r.region_id
left join hd_core.core_season s on x.season_id = s.season_id
--gyo perf
left join hd_stats.calc_pubg_gyo_average ga on
ga.player_id = x.player_id
and nvl(x.season_id,0) = nvl(ga.season_id,0)
and nvl(x.region_id,0) = nvl(ga.region_id,0)
and nvl(x.game_mode,'') = nvl(ga.match_mode,'')
left join (select player_id, region_id,
season_id, match_mode,
rounds_played as rounds_played,
kills as kills,
assists as assists,
headshot_kills as headshot_kills,
top10s as top10s,
wins as wins,
losses as losses,
lastupdate as ___lastupdateid_st
from hd_stats.calc_pubg_player_season_stats s
) as y
on x.player_id = y.player_id
and nvl(x.season_id,0) = nvl(y.season_id,0)
and nvl(x.region_id,0) = nvl(y.region_id,0)
and nvl(x.game_mode,'') = nvl(y.match_mode,'')
order by ___lastupdateid_rn
失败的转换查询:
Select top 100 *
,func_sha1(
''
)
as synch_checksum
from
(
select x.player_id as playerid,
p.player_nickname,
r.region_code,
s.title as season_name,
x.rating as ranking_score,
x.rank_no as rank_no,
x.rank_no_change as rank_no_change,
x.game_mode,
to_char(ga.total_score, 'FM9D00') as gyo_perf_total_score,
to_char(agressive_score*100/aggresive_weight, 'FM9D000') as gyo_perf_aggressive_score,
to_char(defensive_score*100/defensive_weight, 'FM9D000') as gyo_perf_defensive_score,
to_char(survival_score*100/survival_weight, 'FM9D000') as gyo_perf_survival_score,
ga.match_place_avg,
rounds_played,
kills,
(kills*1.0)/rounds_played as avg_kills_per_round,
assists,
(assists*1.0)/rounds_played as avg_assists_per_round,
headshot_kills,
top10s,
(top10s*1.0)/rounds_played as top10s_ratio,
wins,
(wins*1.0)/rounds_played as win_ratio,
case when top10s = 0 then 0 else (wins*1.0)/(top10s*1.0) end as win_to_top10_ratio,
losses,
ga.match_group as ___matchgroup,
ga.last_update_id as ___lastupdateid_ga,
___lastupdateid_st,
x.last_update_id as ___lastupdateid_rn
from hd_stats.calc_player_ranking x
inner join hd_core.core_player p on x.player_id = p.player_id
and p.player_game_id =
(select g.game_id
from hd_core.core_game g
where lower(g.game_short_title) = 'xxxx')
left join hd_core.core_region r on x.region_id = r.region_id
left join hd_core.core_season s on x.season_id = s.season_id
--gyo perf
left join hd_stats.calc_pubg_gyo_average ga on
ga.player_id = x.player_id
and nvl(x.season_id,0) = nvl(ga.season_id,0)
and nvl(x.region_id,0) = nvl(ga.region_id,0)
and nvl(x.game_mode,'') = nvl(ga.match_mode,'')
left join (select player_id, region_id,
season_id, match_mode,
rounds_played as rounds_played,
kills as kills,
assists as assists,
headshot_kills as headshot_kills,
top10s as top10s,
wins as wins,
losses as losses,
lastupdate as ___lastupdateid_st
from hd_stats.calc_pubg_player_season_stats s
) as y
on x.player_id = y.player_id
and nvl(x.season_id,0) = nvl(y.season_id,0)
and nvl(x.region_id,0) = nvl(y.region_id,0)
and nvl(x.game_mode,'') = nvl(y.match_mode,'')
order by ___lastupdateid_rn
)
有效的转换查询:
Select top 100 *
,func_sha1(
''
)
as synch_checksum
from
(
select x.player_id as playerid,
p.player_nickname,
r.region_code,
s.title as season_name,
x.rating as ranking_score,
x.rank_no as rank_no,
x.rank_no_change as rank_no_change,
x.game_mode,
to_char(ga.total_score, 'FM9D00') as gyo_perf_total_score,
to_char(agressive_score*100/aggresive_weight, 'FM9D000') as gyo_perf_aggressive_score,
to_char(defensive_score*100/defensive_weight, 'FM9D000') as gyo_perf_defensive_score,
to_char(survival_score*100/survival_weight, 'FM9D000') as gyo_perf_survival_score,
ga.match_place_avg,
rounds_played,
kills,
(kills*1.0)/rounds_played as avg_kills_per_round,
assists,
(assists*1.0)/rounds_played as avg_assists_per_round,
headshot_kills,
top10s,
(top10s*1.0)/rounds_played as top10s_ratio,
wins,
(wins*1.0)/rounds_played as win_ratio,
case when top10s = 0 then 0 else (wins*1.0)/(top10s*1.0) end as win_to_top10_ratio,
losses,
ga.match_group as ___matchgroup,
ga.last_update_id as ___lastupdateid_ga,
___lastupdateid_st,
x.last_update_id as ___lastupdateid_rn
from hd_stats.calc_player_ranking x
inner join hd_core.core_player p on x.player_id = p.player_id
and p.player_game_id = 2
--(select g.game_id
-- from hd_core.core_game g
-- where lower(g.game_short_title) = 'xxxx')
left join hd_core.core_region r on x.region_id = r.region_id
left join hd_core.core_season s on x.season_id = s.season_id
--gyo perf
left join hd_stats.calc_pubg_gyo_average ga on
ga.player_id = x.player_id
and nvl(x.season_id,0) = nvl(ga.season_id,0)
and nvl(x.region_id,0) = nvl(ga.region_id,0)
and nvl(x.game_mode,'') = nvl(ga.match_mode,'')
left join (select player_id, region_id,
season_id, match_mode,
rounds_played as rounds_played,
kills as kills,
assists as assists,
headshot_kills as headshot_kills,
top10s as top10s,
wins as wins,
losses as losses,
lastupdate as ___lastupdateid_st
from hd_stats.calc_pubg_player_season_stats s
) as y
on x.player_id = y.player_id
and nvl(x.season_id,0) = nvl(y.season_id,0)
and nvl(x.region_id,0) = nvl(y.region_id,0)
and nvl(x.game_mode,'') = nvl(y.match_mode,'')
order by ___lastupdateid_rn
)
预期获取基本查询中的所有记录和列以及校验和的另一列。
相反，我得到一个错误:
亚马逊无效操作:无法识别的节点类型:407； [SQL 状态=XX000，数据库错误代码=500310]
最佳答案
省略 ORDER BY 子句。这减轻了我的错误。
关于sql - redshift 无法识别的节点类型 407，我们在Stack Overflow上找到一个类似的问题： https://stackoverflow.com/questions/57856121/
29
4
0
我正在使用 JavaFX 8 创建一个应用程序。我使用拖/放动态更改网格 Pane 的内容。我希望每行或每行/列迭代 GridPane 内容。JavaFX 允许通过指定行和列在 GridPane 中添
我正在尝试将图像拖放到div上。图像没有被拖到div上并给出以下错误 Uncaught TypeError: Failed to execute 'appendChild' on 'Node': pa
我正在 android studio 中创建内部构建 AR 导航。我正在寻找一种方法将 anchor 与其他 anchor 或 anchor 节点/节点“连接”起来。我不确定使用哪一个。基于我将强制用
我在 Hive 上运行一些作业:首先是 4 节点，然后是 2 节点。令我惊讶的是，我的 2 节点性能比我的 4 节点更好。 首先，我在一个 4 节点(4 个事件节点)上运行查询，然后关闭 2 个节点(
我有 Node* current ，我在其中存储指向列表“顶部”当前节点的指针。当我将一个新节点设置为当前节点时，出现错误: '=' : cannot convert from 'CircularDo
我是 dcos Mesos 的新手，在本地 Ubuntu 机器上安装了 dc os。 我可以查看 dcos 仪表板。 但我无法使用 dcos node ssh --master-proxy --lea
在 JavaFX 中，是否有类似 setLayout(); 的东西？或 setBounds(); ? 例如，我想将按钮定位到我想要的位置。 最佳答案 JavaFX 场景图上的所有内容都是 Node .
我正在开发一个 JavaFX 应用程序，其中我开发的类(从 javafx.scene.Parent 扩展)是根据用户在 ListView 控件中单击的条目动态创建的。 只是要清楚这个节点，它不是使用像
我正在尝试为节点-边缘关系创建一个类图，因为它可以在有向图中找到。我想传达的是，Nodes 引用了 Edges，Edges 也引用了 Nodes。每个 Edge 都恰好需要两个 Node(源和目标)。
在mapreduce作业期间，单个任务将在随机节点上运行，是否有任何方法限制应在其中运行任务的节点？ 最佳答案 Hadoop不会选择节点来随机运行任务。考虑到数据局部性，否则将有很多网络开销。 任务与
有什么区别: a) nodetool 重建 b) nodetool 修复 [-pr] 换句话来说，各个命令到底是做什么的？ 最佳答案 nodetool重建:类似于引导过程(当您向集群添加新节点时)，但
我已将第一个 OneToMany 关系添加到我的 hibernate 3.6.10 项目中。这是一个类: /** * */ package com.heavyweightsoftware.leal
是否有可能找到正在监听触发当前函数的事件的元素？ 在下面的代码中，event.target 返回 #xScrollPane 和 event.currentTarget 和 event 的最低子节点.f
我正在尝试覆盖我数据库中的一些数据。结构很简单，就是: recipes { user_1{ recipe_1{data} recipe_2{data} } user_2{
我使用 setInterval 来运行该函数，但它会多次执行函数 2... 如何在输入中插入一个值后执行函数 第一个输入与其余输入的距离不同 如何在插入 val(tab 选项)后将插入从 1 个输入移
我不知道代码有什么问题，但在 visual studio 中不断收到这些错误消息。 Error 18 error C1903: unable to recover from previous e
我正在尝试从其类中获取 SharePoint 搜索导航节点的对象。 var nodes = $("div.ms-qSuggest-listItem"); 我正在获取节点对象，现在想要获取“_promp
D:\nodeP>node main.js module.js:327 抛出错误； ^ 错误:在 Function.Module 的 Function.Module._resolveFilename
struct node{ int key, prior, cnt, val; node *l, *r; node(){} node(int nkey) : key(nkey),
我有以下代码使用迭代器将项目插入双链表。这就是我们被要求这样做的方式。代码有效，但问题是我有 24 字节的绝对内存泄漏。 NodeIterator insert(NodeIterator & itrP
我是一名优秀的程序员,十分优秀！