Question

我尝试过各种不同的操作，但我的基本问题是：

url<- "http://www.ref.org.uk/fuel/tablebysp.php?valdate=2015-03-08"
data <- readHTMLTable(url,header = TRUE,as.data.frame =TRUE,which=2)
typeof(data)

我的数据看起来很棒，但我无法将其强制转换为数据框。我不知道是什么阻止了我。

Answer 1

正如帖子评论中所指出的，您的代码实际上运行正常。您可以使用字符串而不是因子来获取字符串：

url<- "http://www.ref.org.uk/fuel/tablebysp.php?valdate=2015-03-08"
data <- readHTMLTable(url, header=TRUE, as.data.frame=TRUE, which=2, 
                      stringsAsFactors=FALSE)

下面是如何使用rvest包来实现这一点，它真的很适合这一点，特别是当有多个表或奇怪的嵌套表时。并且，有一些dplyr

在这种情况下，有多个表，第二个表是您想要的。值得庆幸的是，它的形式非常好。下面的代码从页面中提取所有表格（使用CSS选择器），然后使用方便的magrittr extract2来避免使用奇怪/丑陋的[[]]。

管道习惯用法（以magrittr开头，现在用于大部分hadleyverse）从左到右“推”或“流”数据与从嵌套括号调用中“弹出”数据。

library(rvest)
library(magrittr)
library(dplyr)

pg <- html("http://www.ref.org.uk/fuel/tablebysp.php?valdate=2015-03-08")
dat <- pg %>% html_nodes("table") %>% extract2(2) %>% html_table(header=TRUE)
glimpse(dat)

## Observations: 48
## Variables:
## $ SD         (chr) "2015-03-08", "2015-03-08", "2015-03-08", "2015-03-08", "2015-03-08", "2015-03-08", "...
## $ SP         (int) 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24...
## $ Gas        (chr) "3,467", "3,522", "3,594", "3,529", "2,811", "2,538", "2,520", "2,489", "2,498", "2,5...
## $ Coal       (chr) "8,261", "8,062", "7,876", "7,437", "6,751", "6,799", "6,621", "6,428", "6,586", "6,2...
## $ Nuclear    (chr) "7,495", "7,553", "7,641", "7,676", "7,674", "7,672", "7,676", "7,677", "7,672", "7,6...
## $ Hydro      (int) 737, 729, 666, 651, 646, 647, 645, 648, 658, 729, 734, 736, 740, 738, 740, 741, 751, ...
## $ Net Pumped (chr) "-438", "-84", "-504", "-860", "-1,092", "-1,118", "-1,396", "-1,700", "-1,606", "-1,...
## $ Wind       (chr) "4,675", "4,795", "4,623", "4,572", "4,647", "4,570", "4,377", "4,445", "4,602", "4,5...
## $ OCGT       (int) 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0...
## $ Oil        (int) 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0...
## $ Biomass    (chr) "1,078", "1,079", "1,081", "1,048", "1,005", "1,022", "1,086", "1,035", "1,072", "1,0...
## $ French Int (chr) "480", "480", "680", "678", "1,614", "1,626", "1,532", "1,536", "772", "772", "504", ...
## $ Dutch Int  (chr) "860", "852", "874", "838", "850", "866", "848", "830", "830", "866", "862", "862", "...
## $ NI Int     (int) 22, -72, 2, -16, -30, -50, 4, 4, -122, -138, -108, -114, -2, 16, 24, 24, 28, -30, -42...
## $ Eire Int   (int) 170, 190, 142, 142, 142, 114, 114, 114, 114, 112, 88, 50, 16, 16, 16, 42, 18, -72, -1...
## $ Net Supply (chr) "26,807", "27,106", "26,675", "25,695", "25,018", "24,686", "24,027", "23,506", "23,0...

您也可以这样做：

html_table(extract2(html_nodes(pg, "table"), 2), header=TRUE)

如果您不喜欢或一般使用烟斗。

然后，您可以对列进行一些基本清理，以获得有用的数字/日期值：

dat %>% 
  mutate(SD=as.Date(SD),
         Gas=as.numeric(gsub(",", "", Gas)),
         Coal=as.numeric(gsub(",", "", Coal)),
         Nuclear=as.numeric(gsub(",", "", Nuclear)),
         `Net Pumped`=as.numeric(gsub(",", "", `Net Pumped`)),
         `Wind`=as.numeric(gsub(",", "", `Wind`)),
         Biomass=as.numeric(gsub(",", "", Biomass)),
         `French Int`=as.numeric(gsub(",", "", `French Int`)),
         `Dutch Int`=as.numeric(gsub(",", "", `Dutch Int`)),
         `Net Supply`=as.numeric(gsub(",", "", `Net Supply`))) -> dat

glimpse(dat)

## Observations: 48
## Variables:
## $ SD         (date) 2015-03-08, 2015-03-08, 2015-03-08, 2015-03-08, 2015-03-08, 2015-03-08, 2015-03-08, ...
## $ SP         (int) 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24...
## $ Gas        (dbl) 3467, 3522, 3594, 3529, 2811, 2538, 2520, 2489, 2498, 2543, 2531, 2522, 2627, 2729, 2...
## $ Coal       (dbl) 8261, 8062, 7876, 7437, 6751, 6799, 6621, 6428, 6586, 6229, 6194, 6299, 6455, 6639, 6...
## $ Nuclear    (dbl) 7495, 7553, 7641, 7676, 7674, 7672, 7676, 7677, 7672, 7670, 7673, 7677, 7677, 7681, 7...
## $ Hydro      (int) 737, 729, 666, 651, 646, 647, 645, 648, 658, 729, 734, 736, 740, 738, 740, 741, 751, ...
## $ Net Pumped (dbl) -438, -84, -504, -860, -1092, -1118, -1396, -1700, -1606, -1632, -1344, -1052, -1342,...
## $ Wind       (dbl) 4675, 4795, 4623, 4572, 4647, 4570, 4377, 4445, 4602, 4570, 4529, 4512, 4312, 3976, 3...
## $ OCGT       (int) 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0...
## $ Oil        (int) 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0...
## $ Biomass    (dbl) 1078, 1079, 1081, 1048, 1005, 1022, 1086, 1035, 1072, 1086, 1084, 1085, 1086, 1082, 1...
## $ French Int (dbl) 480, 480, 680, 678, 1614, 1626, 1532, 1536, 772, 772, 504, 502, 1598, 1602, 1878, 188...
## $ Dutch Int  (dbl) 860, 852, 874, 838, 850, 866, 848, 830, 830, 866, 862, 862, 884, 846, 942, 914, 1032,...
## $ NI Int     (int) 22, -72, 2, -16, -30, -50, 4, 4, -122, -138, -108, -114, -2, 16, 24, 24, 28, -30, -42...
## $ Eire Int   (int) 170, 190, 142, 142, 142, 114, 114, 114, 114, 112, 88, 50, 16, 16, 16, 42, 18, -72, -1...
## $ Net Supply (dbl) 26807, 27106, 26675, 25695, 25018, 24686, 24027, 23506, 23076, 22807, 22747, 23079, 2...

使用readHTMLTable从URL中抓取数据后，如何将结果转换为数据框？

1 个答案: