温馨提示:本文翻译自stackoverflow.com，查看原文请点击：r - Grouping dates and times
datetime grouping lubridate r iso8601

r - 分组日期和时间

发布于 2020-04-04 10:20:58

是否有一个R包可以为所有典型时间单位（秒，分钟等）提供日期和时间分组？

或换种说法：我将如何截断特定时间单位的倍数？

背景

我经常需要将对HTTP请求的响应分为15秒间隔。

功能lubridate::floor_date()和lubridate::ceiling_date()让我开始，但

我想出了一些基于模（%%）的东西，但是感觉就像我在重新发明有关原子时间组件的轮子。

功能定义

library(magrittr)

group_date <- function(
  x,
  interval = 15,
  unit = c(
    "seconds",
    "minutes",
    "hours",
    "days",
    "weeks",
    "months",
    "bimonths",
    "quarters",
    "seasons",
    "halfyears",
    "years"
  )
) {
  # Validate units:
  unit <- match.arg(unit)

  # Possibly base units on "basic units" as {lubridate} does:
  # parsed_unit <- lubridate:::parse_period_unit(unit)
  # n <- parsed_unit$n
  # basic_unit <- lubridate:::standardise_period_names(parsed_unit$unit)

  if (unit %in% c("bimonths", "halfyears", "season")) {
    stop(stringr::str_glue("Unit '{unit}' not supported yet"))
  }
  # No clue how these would need to be handled yet

  # Extract unit value:
  unit_value <- dplyr::case_when(
    unit == "seconds" ~ as.numeric(lubridate::second(x)),
    unit == "minutes" ~ as.numeric(lubridate::minute(x)),
    unit == "hours" ~ as.numeric(lubridate::hour(x)),
    unit == "days" ~ as.numeric(lubridate::day(x)),
    unit == "weeks" ~ as.numeric(lubridate::isoweek(x)),
    unit == "months" ~ as.numeric(lubridate::month(x)),
    unit == "quarters" ~ as.numeric(lubridate::quarter(x)),
    unit == "year" ~ as.numeric(lubridate::year(x))
  )

  offset_factor <- dplyr::case_when(
    unit == "seconds" ~ 1,
    unit == "minutes" ~ 60,
    unit == "hours" ~ 60 * 60,
    unit == "days" ~ 60 * 60 * 24,
    unit == "weeks" ~ NA_real_, # Seconds per week -> no clue how to do that,
    unit == "months" ~ NA_real_, # Seconds per month -> no clue how to do that
    unit == "quarters" ~ NA_real_, # Seconds per quarter -> no clue how to do that
    unit == "year" ~ NA_real_ # Seconds per year -> no clue how to do that
  )

  # Calculate time offset to lower group boundary:
  time_offset <- unit_value %% interval

  # Apply offset:
  x - (time_offset * offset_factor)
}

分组秒

x <- c(
  "2020-01-31 13:01:14",
  "2020-01-31 13:01:15",
  "2020-01-31 13:01:16",
  "2020-01-31 13:01:29",
  "2020-01-31 13:01:30",
  "2020-01-31 13:01:31",
  "2020-01-31 13:01:44",
  "2020-01-31 13:01:45",
  "2020-01-31 13:01:46",
  "2020-01-31 13:01:59",
  "2020-01-31 13:02:00",
  "2020-01-31 13:02:01"
) %>%
  lubridate::ymd_hms()

x %>% group_date()
#>  [1] "2020-01-31 13:01:00 UTC" "2020-01-31 13:01:15 UTC"
#>  [3] "2020-01-31 13:01:15 UTC" "2020-01-31 13:01:15 UTC"
#>  [5] "2020-01-31 13:01:30 UTC" "2020-01-31 13:01:30 UTC"
#>  [7] "2020-01-31 13:01:30 UTC" "2020-01-31 13:01:45 UTC"
#>  [9] "2020-01-31 13:01:45 UTC" "2020-01-31 13:01:45 UTC"
#> [11] "2020-01-31 13:02:00 UTC" "2020-01-31 13:02:00 UTC"

x %>% group_date(30)
#>  [1] "2020-01-31 13:01:00 UTC" "2020-01-31 13:01:00 UTC"
#>  [3] "2020-01-31 13:01:00 UTC" "2020-01-31 13:01:00 UTC"
#>  [5] "2020-01-31 13:01:30 UTC" "2020-01-31 13:01:30 UTC"
#>  [7] "2020-01-31 13:01:30 UTC" "2020-01-31 13:01:30 UTC"
#>  [9] "2020-01-31 13:01:30 UTC" "2020-01-31 13:01:30 UTC"
#> [11] "2020-01-31 13:02:00 UTC" "2020-01-31 13:02:00 UTC"

x %>% group_date(45)
#>  [1] "2020-01-31 13:01:00 UTC" "2020-01-31 13:01:00 UTC"
#>  [3] "2020-01-31 13:01:00 UTC" "2020-01-31 13:01:00 UTC"
#>  [5] "2020-01-31 13:01:00 UTC" "2020-01-31 13:01:00 UTC"
#>  [7] "2020-01-31 13:01:00 UTC" "2020-01-31 13:01:45 UTC"
#>  [9] "2020-01-31 13:01:45 UTC" "2020-01-31 13:01:45 UTC"
#> [11] "2020-01-31 13:02:00 UTC" "2020-01-31 13:02:00 UTC"

x %>% group_date(60)
#>  [1] "2020-01-31 13:01:00 UTC" "2020-01-31 13:01:00 UTC"
#>  [3] "2020-01-31 13:01:00 UTC" "2020-01-31 13:01:00 UTC"
#>  [5] "2020-01-31 13:01:00 UTC" "2020-01-31 13:01:00 UTC"
#>  [7] "2020-01-31 13:01:00 UTC" "2020-01-31 13:01:00 UTC"
#>  [9] "2020-01-31 13:01:00 UTC" "2020-01-31 13:01:00 UTC"
#> [11] "2020-01-31 13:02:00 UTC" "2020-01-31 13:02:00 UTC"

分组时间

x <- c(
  "2020-01-31 13:04:00",
  "2020-01-31 13:05:00",
  "2020-01-31 13:06:00",
  "2020-01-31 13:29:00",
  "2020-01-31 13:30:00",
  "2020-01-31 13:31:00",
  "2020-01-31 13:44:00",
  "2020-01-31 13:45:00",
  "2020-01-31 13:46:00"
) %>%
  lubridate::ymd_hms()

x %>% group_date(15, "minutes")
#> [1] "2020-01-31 13:00:00 UTC" "2020-01-31 13:00:00 UTC"
#> [3] "2020-01-31 13:00:00 UTC" "2020-01-31 13:15:00 UTC"
#> [5] "2020-01-31 13:30:00 UTC" "2020-01-31 13:30:00 UTC"
#> [7] "2020-01-31 13:30:00 UTC" "2020-01-31 13:45:00 UTC"
#> [9] "2020-01-31 13:45:00 UTC"

x %>% group_date(30, "minutes")
#> [1] "2020-01-31 13:00:00 UTC" "2020-01-31 13:00:00 UTC"
#> [3] "2020-01-31 13:00:00 UTC" "2020-01-31 13:00:00 UTC"
#> [5] "2020-01-31 13:30:00 UTC" "2020-01-31 13:30:00 UTC"
#> [7] "2020-01-31 13:30:00 UTC" "2020-01-31 13:30:00 UTC"
#> [9] "2020-01-31 13:30:00 UTC"

x %>% group_date(45, "minutes")
#> [1] "2020-01-31 13:00:00 UTC" "2020-01-31 13:00:00 UTC"
#> [3] "2020-01-31 13:00:00 UTC" "2020-01-31 13:00:00 UTC"
#> [5] "2020-01-31 13:00:00 UTC" "2020-01-31 13:00:00 UTC"
#> [7] "2020-01-31 13:00:00 UTC" "2020-01-31 13:45:00 UTC"
#> [9] "2020-01-31 13:45:00 UTC"

编辑

刚遇到hms::trunc_hms()。

似乎给我什么，我需要secs = 15和secs = 30，所以它会解决我的眼前的问题。但是我看不到秒以外的时间单位如何工作：

library(magrittr)

x <- c(
  "2020-01-31 13:01:14",
  "2020-01-31 13:01:15",
  "2020-01-31 13:01:16",
  "2020-01-31 13:01:29",
  "2020-01-31 13:01:30",
  "2020-01-31 13:01:31",
  "2020-01-31 13:01:44",
  "2020-01-31 13:01:45",
  "2020-01-31 13:01:46",
  "2020-01-31 13:01:59",
  "2020-01-31 13:02:00",
  "2020-01-31 13:02:01"
) %>%
  lubridate::ymd_hms()

x %>% hms::trunc_hms(15)
#>  [1] "2020-01-31 13:01:00 UTC" "2020-01-31 13:01:15 UTC"
#>  [3] "2020-01-31 13:01:15 UTC" "2020-01-31 13:01:15 UTC"
#>  [5] "2020-01-31 13:01:30 UTC" "2020-01-31 13:01:30 UTC"
#>  [7] "2020-01-31 13:01:30 UTC" "2020-01-31 13:01:45 UTC"
#>  [9] "2020-01-31 13:01:45 UTC" "2020-01-31 13:01:45 UTC"
#> [11] "2020-01-31 13:02:00 UTC" "2020-01-31 13:02:00 UTC"

x %>% hms::trunc_hms(30)
#>  [1] "2020-01-31 13:01:00 UTC" "2020-01-31 13:01:00 UTC"
#>  [3] "2020-01-31 13:01:00 UTC" "2020-01-31 13:01:00 UTC"
#>  [5] "2020-01-31 13:01:30 UTC" "2020-01-31 13:01:30 UTC"
#>  [7] "2020-01-31 13:01:30 UTC" "2020-01-31 13:01:30 UTC"
#>  [9] "2020-01-31 13:01:30 UTC" "2020-01-31 13:01:30 UTC"
#> [11] "2020-01-31 13:02:00 UTC" "2020-01-31 13:02:00 UTC"

x %>% hms::trunc_hms(45)
#>  [1] "2020-01-31 13:00:45 UTC" "2020-01-31 13:00:45 UTC"
#>  [3] "2020-01-31 13:00:45 UTC" "2020-01-31 13:00:45 UTC"
#>  [5] "2020-01-31 13:01:30 UTC" "2020-01-31 13:01:30 UTC"
#>  [7] "2020-01-31 13:01:30 UTC" "2020-01-31 13:01:30 UTC"
#>  [9] "2020-01-31 13:01:30 UTC" "2020-01-31 13:01:30 UTC"
#> [11] "2020-01-31 13:01:30 UTC" "2020-01-31 13:01:30 UTC"

x %>% hms::trunc_hms(60)
#>  [1] "2020-01-31 13:01:00 UTC" "2020-01-31 13:01:00 UTC"
#>  [3] "2020-01-31 13:01:00 UTC" "2020-01-31 13:01:00 UTC"
#>  [5] "2020-01-31 13:01:00 UTC" "2020-01-31 13:01:00 UTC"
#>  [7] "2020-01-31 13:01:00 UTC" "2020-01-31 13:01:00 UTC"
#>  [9] "2020-01-31 13:01:00 UTC" "2020-01-31 13:01:00 UTC"
#> [11] "2020-01-31 13:02:00 UTC" "2020-01-31 13:02:00 UTC"

编辑2

调查使用的建议 cut()

x %>%
  tibble::enframe() %>%
  dplyr::mutate(
    grp = value %>% group_date(15)
  ) %>% 
  dplyr::group_by(
    grp_2 = cut(as.POSIXct(value, '%Y-%m-%d %H:%M:%S'), '15 secs')
  )
# # A tibble: 12 x 4
# # Groups:   grp [4]
# name value                   grp_2                   grp                
# <int> <dttm>                  <dttm>                  <fct>              
#   1     1 2020-01-31 13:01:14.000 2020-01-31 13:01:00.000 2020-01-31 13:01:14
# 2     2 2020-01-31 13:01:15.000 2020-01-31 13:01:15.000 2020-01-31 13:01:14
# 3     3 2020-01-31 13:01:16.000 2020-01-31 13:01:15.000 2020-01-31 13:01:14
# 4     4 2020-01-31 13:01:29.000 2020-01-31 13:01:15.000 2020-01-31 13:01:29
# 5     5 2020-01-31 13:01:30.000 2020-01-31 13:01:30.000 2020-01-31 13:01:29
# 6     6 2020-01-31 13:01:31.000 2020-01-31 13:01:30.000 2020-01-31 13:01:29
# 7     7 2020-01-31 13:01:44.000 2020-01-31 13:01:30.000 2020-01-31 13:01:44
# 8     8 2020-01-31 13:01:45.000 2020-01-31 13:01:45.000 2020-01-31 13:01:44
# 9     9 2020-01-31 13:01:46.000 2020-01-31 13:01:45.000 2020-01-31 13:01:44
# 10    10 2020-01-31 13:01:59.000 2020-01-31 13:01:45.000 2020-01-31 13:01:59
# 11    11 2020-01-31 13:02:00.000 2020-01-31 13:02:00.000 2020-01-31 13:01:59
# 12    12 2020-01-31 13:02:01.000 2020-01-31 13:02:00.000 2020-01-31 13:01:59

由reprex软件包（v0.3.0）创建于2020-01-31

查看更多

提问者
Rappster
被浏览
71
分享
Allan Cameron 2020-01-31 22:07

如何将此简单函数舍入到给定的秒数，分钟数，小时数，天数或周数？如果要在特定的日期和时间启动块，则有一个可选的默认原始时间。“ units”的参数是匹配的，因此可以缩写。默认为秒

time_group <- function(times, intervals, since = as.POSIXct("2000-01-01"), 
                       units = c("secs", "mins", "hours", "days", "weeks"))
{
  all_units <- c("secs", "mins", "hours", "days", "weeks")
  units     <- match.arg(units, all_units)
  intervals <- intervals * c(1, 60, 3600, 86400, 604800)[match(units, all_units)]
  cuts      <- intervals * floor(as.numeric(difftime(times, since, units = "secs"))/intervals)

  return(as.POSIXct(cuts, origin = since))
}

这使您可以执行以下操作：

# Units default to seconds so this groups by 15 seconds at a time
time_group(x, 15)
#>  [1] "2020-01-31 13:01:00 GMT" "2020-01-31 13:01:15 GMT" "2020-01-31 13:01:15 GMT"
#>  [4] "2020-01-31 13:01:15 GMT" "2020-01-31 13:01:30 GMT" "2020-01-31 13:01:30 GMT"
#>  [7] "2020-01-31 13:01:30 GMT" "2020-01-31 13:01:45 GMT" "2020-01-31 13:01:45 GMT"
#> [10] "2020-01-31 13:01:45 GMT" "2020-01-31 13:02:00 GMT" "2020-01-31 13:02:00 GMT"

# We have used argument matching so we can abbreviate minutes to "m"
time_group(x, 1, units = "m")
#>  [1] "2020-01-31 13:01:00 GMT" "2020-01-31 13:01:00 GMT" "2020-01-31 13:01:00 GMT"
#>  [4] "2020-01-31 13:01:00 GMT" "2020-01-31 13:01:00 GMT" "2020-01-31 13:01:00 GMT"
#>  [7] "2020-01-31 13:01:00 GMT" "2020-01-31 13:01:00 GMT" "2020-01-31 13:01:00 GMT"
#> [10] "2020-01-31 13:01:00 GMT" "2020-01-31 13:02:00 GMT" "2020-01-31 13:02:00 GMT"
Rappster 2020-01-31 21:45:08

是的，我认为使用的是相同的基本方法hms::trunc_hms()：trunc(as.numeric(x)/secs) * secs

Rappster 2020-01-31 21:47:49

但是，当我希望将其分组/截断为该时间单位的倍数时，如何将其推广到其他时间单位？

Allan Cameron 2020-01-31 22:00:01

@Rappster看到我的更新

Rappster 2020-01-31 22:05:08

太好了，非常感谢

热门帖子

1
程序员想开发漂亮的个人网站是不是用 react 会比 vue 简单一些？
2
有关于上海居住证 120 积分
3
窗户隔音方案讨论。
4
急！二手房过户了房东不愿搬走怎么解决？
5
[送会员🐯] 全职开发的一款背单词软件
6
华为 MateBook X Pro 2024 上手吐槽一下这个触摸板
7
排除自带和所谓的 Rime，你们 iPhone 上用啥输入法？
8
寻找靠谱的服务器运维人员
9
用过 discuz! 的技术人/个人站长孩子早就满地跑了吧？
10
黑豆 黑米 黑芝麻 桑葚对秃顶到底有没有用?

相关问题

1
如何计算两个日期之间的年数？
2
TypeError：无法将类似datetime的日期从[datetime64 [ns]]分配为[timedelta64 [D]]
3
如何使用MySQL中的Between子句加快count（distinct）
4
如何使用Ruby on Rails查找今天，昨天等的记录？
5
对一组对象进行分组并添加时间属性
6
时间转换器
7
获取日期时间和恒定时间变量 pandas 之间的时差
8
Python：TypeError：dtype = datetime64 [ns]和日期之间的比较无效
9
C＃DateTime + TimeSpan =额外的12小时？
10
如何将字符串转换为像UTC这样的DateTime就这么简单

热门github

1
Implementation of paper - YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information (翻译：论文实现——YOLOv9: Learning What You Want to Learn using Programmable Gradient Information（利用可编程梯度信息学习您想学的内容）)
2
A Windows and Office activator using HWID / Ohook / KMS38 / Online KMS activation methods, with a focus on open-source code and fewer antivirus detections. (翻译：使用 HWID / KMS38 / Online KMS 激活方法激活 Microsoft 产品的脚本集合，重点是开源代码、较少的防病毒检测和用户友好性。)
3
Get up and running with Llama 2, Mistral, Gemma, and other large language models. (翻译：启动并运行 Llama 2、Mistral、Gemma 和其他大型语言模型。)
4
该项目可以让你通过订阅的方式使用Cloudflare WARP+，自动获取流量。This project enables you to use Cloudflare WARP+ through subscription, automatically acquiring traffic.
5
Multi functional app to find duplicates, empty folders, similar images etc. (翻译：多功能应用程序可查找重复项、空文件夹、相似图像等。)
6
Xray panel supporting multi-protocol multi-user expire day & traffic & ip limit (Vmess & Vless & Trojan & ShadowSocks & Wireguard) (翻译：Xray面板支持多协议多用户到期日&流量&IP限制（Vmess & Vless & Trojan & ShadowSocks & Wireguard)
7
The Free Software Media System (翻译：Jellyfin 是一个免费软件媒体系统，可让您控制管理和流式传输您的媒体。它是专有 Emby 和 Plex 的替代方案，通过多个应用程序从专用服务器向最终用户设备提供媒体。)
8
lightweight, standalone C++ inference engine for Google's Gemma models. (翻译：适用于 Google Gemma 模型的轻量级独立 C++ 推理引擎。)
9
📚 Freely available programming books (翻译：📚 免费提供的编程书籍)
10
A collective list of free APIs (翻译：免费 API 的集合列表)
11
1️⃣🐝🏎️ The One Billion Row Challenge -- A fun exploration of how quickly 1B rows from a text file can be aggregated with Java (翻译：十亿行挑战 —— 使用 Java 对文本文件中的 10 亿行数据进行聚合的有趣探索)
12
🎓 Path to a free self-taught education in Computer Science! (翻译：🎓计算机科学免费自学教程！)
13
Curso para aprender el lenguaje de programación Python desde cero y para principiantes. 75 clases, 37 horas en vídeo, código, proyectos y grupo de chat. Fundamentos, frontend, backend, testing, IA... (翻译：从零开始学习 Python 编程语言的课程，适合初学者)
14
This repository contains System Design resources which are useful while preparing for interviews and learning Distributed Systems (翻译：该存储库包含系统设计资源，在准备面试和学习分布式系统时非常有用)
15
Mamba is a new state space model architecture showing promising performance on information-dense data such as language modeling, where previous subquadratic models fall short of Transformers. It is based on the line of progress on structured state space models, with an efficient hardware-aware design and implementation in the spirit of FlashAttention. (翻译：Mamba 是一种新的状态空间模型架构，在信息密集型数据（例如语言建模）上显示出良好的性能，而之前的二次模型在 Transformers 方面存在不足。它基于结构化状态空间模型的进展，并本着FlashAttention的精神进行高效的硬件感知设计和实现。)