blocks|key|1915058|text|+P+<-+function(t,+s)+{
++ep+<-+.parseISO8601(s)

++if(grepl('T[0-9]{2}:[0-9]{2}:[0-9]{2}/T[0-9]{2}:[0-9]{2}:[0-9]{2}',+s)){
++++first.time+<-+as.double(ep$first.time)
++++last.time+<-+as.double(ep$last.time)-31449600
++++SecOfDay+<-+as.double(t)+%25%25+86400
++++return(SecOfDay+>=+first.time+&+SecOfDay+<=+last.time+)

++}+else+{
++++return(t+>=+ep$first.time+&+t+<=+ep$last.time)++++
++}

}

F+<-+function(t,+st,+et)+{
++time+<-+as.ITime(t)+
++return(time>=as.ITime(st)+&+time<=as.ITime(et))
}


+Sys.setenv(TZ='GMT')
+N+=+1e7;
+set.seed(1);

+DT+<-+data.table(dts+=+.POSIXct(1e5*rnorm(N),+tz="GMT"))


+system.time(resP+<-+DT[P(dts,+'T00:00:00/T12:00:00'),+])
##+++user++system+elapsed+
##+++1.11++++0.11++++1.22+
+system.time(resF+<-+DT[F(dts,'00:00:00','12:00:00')])
##+++user++system+elapsed+
##+++2.22++++0.29++++2.51+

+resP
##+++++++++++++++++++++++++dts
##++++++1:+1969-12-31+06:35:54
##++++++2:+1970-01-01+05:06:04
##++++++3:+1969-12-31+00:47:17
##++++++4:+1970-01-01+09:09:10
##++++++5:+1969-12-31+01:12:33
##+++++---++++++++++++++++++++
##5000672:+1970-01-01+06:08:15
##5000673:+1970-01-01+05:02:27
##5000674:+1969-12-31+02:25:24
##5000675:+1970-01-03+00:37:00
##5000676:+1969-12-30+08:30:23
+resF
##+++++++++++++++++++++++++dts
##++++++1:+1969-12-31+06:35:54
##++++++2:+1970-01-01+05:06:04
##++++++3:+1969-12-31+00:47:17
##++++++4:+1970-01-01+09:09:10
##++++++5:+1969-12-31+01:12:33
##+++++---++++++++++++++++++++
##5000672:+1970-01-01+06:08:15
##5000673:+1970-01-01+05:02:27
##5000674:+1969-12-31+02:25:24
##5000675:+1970-01-03+00:37:00
##5000676:+1969-12-30+08:30:23

+#Check+the+correctness
+resP[,list(mindts=max(dts)),by=list(as.Date(dts))]
##+++++++as.Date++++++++++++++mindts
##+1:+1969-12-31+1969-12-31+12:00:00
##+2:+1970-01-01+1970-01-01+12:00:00
##+3:+1969-12-29+1969-12-29+12:00:00
##+4:+1970-01-02+1970-01-02+12:00:00
##+5:+1969-12-30+1969-12-30+12:00:00
##+6:+1970-01-03+1970-01-03+12:00:00
##+7:+1970-01-04+1970-01-04+11:59:59
##+8:+1970-01-05+1970-01-05+11:59:45
##+9:+1969-12-28+1969-12-28+12:00:00
##10:+1969-12-27+1969-12-27+11:59:21
##11:+1970-01-06+1970-01-06+10:53:21
##12:+1969-12-26+1969-12-26+10:15:03
##13:+1970-01-07+1970-01-07+08:21:55
+resF[,list(mindts=max(dts)),by=list(as.Date(dts))]
##+++++++as.Date++++++++++++++mindts
##+1:+1969-12-31+1969-12-31+12:00:00
##+2:+1970-01-01+1970-01-01+12:00:00
##+3:+1969-12-29+1969-12-29+12:00:00
##+4:+1970-01-02+1970-01-02+12:00:00
##+5:+1969-12-30+1969-12-30+12:00:00
##+6:+1970-01-03+1970-01-03+12:00:00
##+7:+1970-01-04+1970-01-04+11:59:59
##+8:+1970-01-05+1970-01-05+11:59:45
##+9:+1969-12-28+1969-12-28+12:00:00
##10:+1969-12-27+1969-12-27+11:59:21
##11:+1970-01-06+1970-01-06+10:53:21
##12:+1969-12-26+1969-12-26+10:15:03
##13:+1970-01-07+1970-01-07+08:21:55|type|code-block|depth|inlineStyleRanges|entityRanges|data|syntax|javascript|1915059|下面是一些不错的xts样式子设置的演示|unstyled|offset|length|style|CODE|1915060|+DT[P(dts,+'1970')]
##+++++++++++++++++++++++++dts
##++++++1:+1970-01-01+05:06:04
##++++++2:+1970-01-02+20:18:48
##++++++3:+1970-01-01+09:09:10
##++++++4:+1970-01-01+13:32:22
##++++++5:+1970-01-01+20:30:32
##+++++---++++++++++++++++++++
##5001741:+1970-01-02+15:51:12
##5001742:+1970-01-03+01:41:31
##5001743:+1970-01-01+06:08:15
##5001744:+1970-01-01+05:02:27
##5001745:+1970-01-03+00:37:00
+DT[P(dts,+'197001')]
##+++++++++++++++++++++++++dts
##++++++1:+1970-01-01+05:06:04
##++++++2:+1970-01-02+20:18:48
##++++++3:+1970-01-01+09:09:10
##++++++4:+1970-01-01+13:32:22
##++++++5:+1970-01-01+20:30:32
##+++++---++++++++++++++++++++
##5001741:+1970-01-02+15:51:12
##5001742:+1970-01-03+01:41:31
##5001743:+1970-01-01+06:08:15
##5001744:+1970-01-01+05:02:27
##5001745:+1970-01-03+00:37:00
+DT[P(dts,+'19700102')]
##+++++++++++++++++++++++++dts
##++++++1:+1970-01-02+20:18:48
##++++++2:+1970-01-02+17:59:38
##++++++3:+1970-01-02+07:14:53
##++++++4:+1970-01-02+02:13:03
##++++++5:+1970-01-02+01:31:37
##+++++---++++++++++++++++++++
##1519426:+1970-01-02+11:25:24
##1519427:+1970-01-02+10:00:21
##1519428:+1970-01-02+05:21:25
##1519429:+1970-01-02+05:11:26
##1519430:+1970-01-02+15:51:12
+DT[P(dts,+'19700102+00:00:00/19700103+12:00:00')]
##+++++++++++++++++++++++++dts
##++++++1:+1970-01-02+20:18:48
##++++++2:+1970-01-02+17:59:38
##++++++3:+1970-01-02+07:14:53
##++++++4:+1970-01-02+02:13:03
##++++++5:+1970-01-02+01:31:37
##+++++---++++++++++++++++++++
##1785762:+1970-01-02+05:21:25
##1785763:+1970-01-02+05:11:26
##1785764:+1970-01-02+15:51:12
##1785765:+1970-01-03+01:41:31
##1785766:+1970-01-03+00:37:00

+#Check+the+correctness+again
+DT[P(dts,+'19700102+00:00:00/19700103+12:00:00'),+max(dts)]
##[1]+"1970-01-03+12:00:00+GMT"
+DT[P(dts,+'19700102+00:00:00/19700103+12:00:00'),+min(dts)]
##[1]+"1970-01-02+00:00:00+GMT"|1915061|entityMap^0|0|8|3|0|0^^$0|@$1|2|3|4|5|6|7|O|8|@]|9|@]|A|$B|C]]|$1|D|3|E|5|F|7|P|8|@$G|Q|H|R|I|J]]|9|@]|A|$]]|$1|K|3|L|5|6|7|S|8|@]|9|@]|A|$B|C]]|$1|M|3|-4|5|F|7|T|8|@]|9|@]|A|$]]]|N|$]]

<pre><code> P &lt;- function(t, s) {
 ep &lt;- .parseISO8601(s)

 if(grepl('T[0-9]{2}:[0-9]{2}:[0-9]{2}/T[0-9]{2}:[0-9]{2}:[0-9]{2}', s)){
 first.time &lt;- as.double(ep$first.time)
 last.time &lt;- as.double(ep$last.time)-31449600
 SecOfDay &lt;- as.double(t) %% 86400
 return(SecOfDay &gt;= first.time &amp; SecOfDay &lt;= last.time )

 } else {
 return(t &gt;= ep$first.time &amp; t &lt;= ep$last.time) 
 }

}

F &lt;- function(t, st, et) {
 time &lt;- as.ITime(t) 
 return(time&gt;=as.ITime(st) &amp; time&lt;=as.ITime(et))
}


 Sys.setenv(TZ='GMT')
 N = 1e7;
 set.seed(1);

 DT &lt;- data.table(dts = .POSIXct(1e5*rnorm(N), tz="GMT"))


 system.time(resP &lt;- DT[P(dts, 'T00:00:00/T12:00:00'), ])
## user system elapsed 
## 1.11 0.11 1.22 
 system.time(resF &lt;- DT[F(dts,'00:00:00','12:00:00')])
## user system elapsed 
## 2.22 0.29 2.51 

 resP
## dts
## 1: 1969-12-31 06:35:54
## 2: 1970-01-01 05:06:04
## 3: 1969-12-31 00:47:17
## 4: 1970-01-01 09:09:10
## 5: 1969-12-31 01:12:33
## --- 
##5000672: 1970-01-01 06:08:15
##5000673: 1970-01-01 05:02:27
##5000674: 1969-12-31 02:25:24
##5000675: 1970-01-03 00:37:00
##5000676: 1969-12-30 08:30:23
 resF
## dts
## 1: 1969-12-31 06:35:54
## 2: 1970-01-01 05:06:04
## 3: 1969-12-31 00:47:17
## 4: 1970-01-01 09:09:10
## 5: 1969-12-31 01:12:33
## --- 
##5000672: 1970-01-01 06:08:15
##5000673: 1970-01-01 05:02:27
##5000674: 1969-12-31 02:25:24
##5000675: 1970-01-03 00:37:00
##5000676: 1969-12-30 08:30:23

 #Check the correctness
 resP[,list(mindts=max(dts)),by=list(as.Date(dts))]
## as.Date mindts
## 1: 1969-12-31 1969-12-31 12:00:00
## 2: 1970-01-01 1970-01-01 12:00:00
## 3: 1969-12-29 1969-12-29 12:00:00
## 4: 1970-01-02 1970-01-02 12:00:00
## 5: 1969-12-30 1969-12-30 12:00:00
## 6: 1970-01-03 1970-01-03 12:00:00
## 7: 1970-01-04 1970-01-04 11:59:59
## 8: 1970-01-05 1970-01-05 11:59:45
## 9: 1969-12-28 1969-12-28 12:00:00
##10: 1969-12-27 1969-12-27 11:59:21
##11: 1970-01-06 1970-01-06 10:53:21
##12: 1969-12-26 1969-12-26 10:15:03
##13: 1970-01-07 1970-01-07 08:21:55
 resF[,list(mindts=max(dts)),by=list(as.Date(dts))]
## as.Date mindts
## 1: 1969-12-31 1969-12-31 12:00:00
## 2: 1970-01-01 1970-01-01 12:00:00
## 3: 1969-12-29 1969-12-29 12:00:00
## 4: 1970-01-02 1970-01-02 12:00:00
## 5: 1969-12-30 1969-12-30 12:00:00
## 6: 1970-01-03 1970-01-03 12:00:00
## 7: 1970-01-04 1970-01-04 11:59:59
## 8: 1970-01-05 1970-01-05 11:59:45
## 9: 1969-12-28 1969-12-28 12:00:00
##10: 1969-12-27 1969-12-27 11:59:21
##11: 1970-01-06 1970-01-06 10:53:21
##12: 1969-12-26 1969-12-26 10:15:03
##13: 1970-01-07 1970-01-07 08:21:55
</code></pre>

Now some demo of nice <code>xts</code> style subsetting

<pre><code> DT[P(dts, '1970')]
## dts
## 1: 1970-01-01 05:06:04
## 2: 1970-01-02 20:18:48
## 3: 1970-01-01 09:09:10
## 4: 1970-01-01 13:32:22
## 5: 1970-01-01 20:30:32
## --- 
##5001741: 1970-01-02 15:51:12
##5001742: 1970-01-03 01:41:31
##5001743: 1970-01-01 06:08:15
##5001744: 1970-01-01 05:02:27
##5001745: 1970-01-03 00:37:00
 DT[P(dts, '197001')]
## dts
## 1: 1970-01-01 05:06:04
## 2: 1970-01-02 20:18:48
## 3: 1970-01-01 09:09:10
## 4: 1970-01-01 13:32:22
## 5: 1970-01-01 20:30:32
## --- 
##5001741: 1970-01-02 15:51:12
##5001742: 1970-01-03 01:41:31
##5001743: 1970-01-01 06:08:15
##5001744: 1970-01-01 05:02:27
##5001745: 1970-01-03 00:37:00
 DT[P(dts, '19700102')]
## dts
## 1: 1970-01-02 20:18:48
## 2: 1970-01-02 17:59:38
## 3: 1970-01-02 07:14:53
## 4: 1970-01-02 02:13:03
## 5: 1970-01-02 01:31:37
## --- 
##1519426: 1970-01-02 11:25:24
##1519427: 1970-01-02 10:00:21
##1519428: 1970-01-02 05:21:25
##1519429: 1970-01-02 05:11:26
##1519430: 1970-01-02 15:51:12
 DT[P(dts, '19700102 00:00:00/19700103 12:00:00')]
## dts
## 1: 1970-01-02 20:18:48
## 2: 1970-01-02 17:59:38
## 3: 1970-01-02 07:14:53
## 4: 1970-01-02 02:13:03
## 5: 1970-01-02 01:31:37
## --- 
##1785762: 1970-01-02 05:21:25
##1785763: 1970-01-02 05:11:26
##1785764: 1970-01-02 15:51:12
##1785765: 1970-01-03 01:41:31
##1785766: 1970-01-03 00:37:00

 #Check the correctness again
 DT[P(dts, '19700102 00:00:00/19700103 12:00:00'), max(dts)]
##[1] "1970-01-03 12:00:00 GMT"
 DT[P(dts, '19700102 00:00:00/19700103 12:00:00'), min(dts)]
##[1] "1970-01-02 00:00:00 GMT"
</code></pre>

blocks|key|1915072|text|执行此操作的规范方法是转换为POSIXlt并提取hour组件。|type|unstyled|depth|inlineStyleRanges|offset|length|style|CODE|entityRanges|data|1915073|hour(as.POSIXlt(DT$dts,+"GMT"))+<+12|code-block|syntax|javascript|1915074|这在性能上似乎可以与所讨论的其他技术相媲美(并且更容易理解)。|1915075|entityMap^0|E|7|0|0|0^^$0|@$1|2|3|4|5|6|7|O|8|@$9|P|A|Q|B|C]]|D|@]|E|$]]|$1|F|3|G|5|H|7|R|8|@]|D|@]|E|$I|J]]|$1|K|3|L|5|6|7|S|8|@]|D|@]|E|$]]|$1|M|3|-4|5|6|7|T|8|@]|D|@]|E|$]]]|N|$]]

The canonical way of doing this is to convert to <code>POSIXlt</code> and extract the hour component.

<pre><code>hour(as.POSIXlt(DT$dts, "GMT")) &lt; 12
</code></pre>

This seems to be comparable in performance to the other techniques discussed (and is easier to understand).

blocks|key|1701596|text|这是一个较晚的条目，但我认为as.POSIXlt解决方案将创建一个已命名的向量列表，其中您只需要一个小时|type|unstyled|depth|inlineStyleRanges|offset|length|style|CODE|entityRanges|data|1701597|我会键入一个ITime列，然后使用二进制搜索来对这些时间进行子集，直到中午12点|1701598|12+up之前有60*60+*12+-+1秒数，因此seq_len(43199)将返回12+up之前(但不包括在内)的所有内容|1701599|#+create+IDate+and+ITime+columns+and+key+by+time
setkey(DT[,+c('Date','Time')+:=+IDateTime(dts)],Time)

#+subset+times+before+12pm
DT[.(seq_len(43199))]|code-block|syntax|javascript|1701600|entityMap^0|E|A|0|6|5|0|8|D|Q|E|0|0^^$0|@$1|2|3|4|5|6|7|Q|8|@$9|R|A|S|B|C]]|D|@]|E|$]]|$1|F|3|G|5|6|7|T|8|@$9|U|A|V|B|C]]|D|@]|E|$]]|$1|H|3|I|5|6|7|W|8|@$9|X|A|Y|B|C]|$9|Z|A|10|B|C]]|D|@]|E|$]]|$1|J|3|K|5|L|7|11|8|@]|D|@]|E|$M|N]]|$1|O|3|-4|5|6|7|12|8|@]|D|@]|E|$]]]|P|$]]

A late entry, but I think that the <code>as.POSIXlt</code> solution will create a named list of vectors, of which you only want the hour

I would key by an <code>ITime</code> column and then use a binary search to subset those times before 12pm

There are <code>60*60 *12 - 1</code> seconds before 12pm so <code>seq_len(43199)</code> will return everything up to (but not including) 12pm

<pre><code># create IDate and ITime columns and key by time
setkey(DT[, c('Date','Time') := IDateTime(dts)],Time)

# subset times before 12pm
DT[.(seq_len(43199))]
</code></pre>

Hello I am looking for an efficient way of selecting <code>POSIXct</code> rows from a <code>data.table</code> such that the time of day is less than say <code>12:00:00</code> (NOTE that millisecond is NOT required, so we can use <code>ITime</code> for example)

<pre><code>set.seed(1); N = 1e7;
DT = data.table(dts = .POSIXct(1e5*rnorm(N), tz="GMT"))
DT
 dts
# 1: 1969-12-31 06:35:54.618925
# 2: 1970-01-01 05:06:04.332422
# --- 
# 9999999: 1970-01-03 00:37:00.035565
#10000000: 1969-12-30 08:30:23.624506
</code></pre>

One solution (the problem here is that the cast could be costly if N is big)

<pre><code>f &lt;- function(t, st, et) {time &lt;- as.ITime(t); return(time&gt;=as.ITime(st) &amp; time&lt;=as.ITime(et))}
P &lt;- function(t, s) { #geekTrader solution
 ep &lt;- .parseISO8601(s) 
 if(grepl('T[0-9]{2}:[0-9]{2}:[0-9]{2}/T[0-9]{2}:[0-9]{2}:[0-9]{2}', s)){
 first.time &lt;- as.double(ep$first.time)
 last.time &lt;- as.double(ep$last.time)-31449600
 SecOfDay &lt;- as.double(t) %% 86400
 return(SecOfDay &gt;= first.time &amp; SecOfDay &lt;= last.time )
 } else {
 return(t &gt;= ep$first.time &amp; t &lt;= ep$last.time) 
 }
}
</code></pre>

Quick look about the perf

<pre><code>system.time(resf &lt;- DT[f(dts,'00:00:00','11:59:59')])
 user system elapsed 
 1.01 0.28 1.29
system.time(resP &lt;- DT[P(dts,'T00:00:00/T11:59:59')])
 user system elapsed 
 0.64 0.13 0.76 

identical(resf,resP)
[1] TRUE
</code></pre>

Efficient comparison of POSIXct in data.table

翻译质量差，导致语言生硬或混乱。

没有提供实际的解决方法或示例。

解答不清晰，无法理解或解决问题。

页面排版不美观，阅读体验差。

文章

问答

视频

学习中心

腾讯云实验室

直播

竞赛

腾讯云代码分析专区

腾讯iOA零信任安全管理系统专区

腾讯云架构师技术同盟交流圈

腾讯云数据库专区

腾讯云顾问专区

腾讯云原生专区

腾讯混元专区

腾讯云TCE专区

腾讯云Lighthouse专区

腾讯云HAI专区

腾讯云Edgeone专区

腾讯云存储专区

腾讯云智能专区

腾讯轻联专区 

腾讯云开发专区

TAPD专区

腾讯轻量云游戏服专区

腾讯云最具价值专家

腾讯云架构师技术同盟

腾讯云创作之星

腾讯云开发者先锋

腾讯云代码助手

云原生构建

TAPD 敏捷项目管理

Cloud Studio

SDK中心

API中心

命令行工具

涵盖代码开发、场景应用、自动测试全流程，助你从零构建专属AI助手

一站式MCP教程库，解锁AI应用新玩法

你好，我正在寻找一种从data.table中选择POSIXct行的有效方法，以便一天中的时间小于12:00:00 (注意，毫秒不是必需的，因此我们可以使用ITime )set.seed(1); N = 1e7;DT = data.table(dts = .POSIXct(1e5*rnorm(N), tz="GMT"))...

问data.table中POSIXct的高效比较
EN

社区

活动

圈层

关于

腾讯云开发者

热门产品

热门推荐

更多推荐

问data.table中POSIXct的高效比较EN

社区

活动

圈层

关于

腾讯云开发者

热门产品

热门推荐

更多推荐

问data.table中POSIXct的高效比较
EN