将长字符串转换为数据框

Question

这是一个新手问题，但它让我疯狂了。我有一个名为bars.list的字符向量，它是从FTP服务器下载的。向量如下所示：

"\"\",\"times\",\"open\",\"high\",\"low\",\"close\",\"numEvents\",\"volume\"\r\n\"1\",2015-05-18 06:50:00,23.98,23.98,23.5,23.77,421,0\r\n\"2\",2015-05-18 07:50:00,23.77,23.9,23.34,23.6,720,0\r\n\"3\",2015-05-18 08:50:00,23.6,23.6,23.32,23.42,720,0\r\n\"4\",2015-05-18 09:50:00,23.44,23.91,23.43,23.66,720,0\r\n\"5\",2015-05-18 10:50:00,23.67,24.06,23.59,24.02,720,0\r\n\"6\",2015-05-18 11:50:00,24.02,24.04,23.32,23.33,720,0\r\n\"7\",2015-05-18 12:50:00,23.33,23.42,22.74,22.81,720,0\r\n\"8\",2015-05-18 13:50:00,22.79,22.92,22.49,22.69,720,0\r\n\"9\",2015-05-18 14:50:00,22.69,22.7,22.14,22.14,481,0\r\n\"10\",2015-05-19 06:50:00,21.09,21.49,20.82,21.47,421,0\r\n\"11\",2015-05-19 07:50:00,21.48,21.68,21.46,21.51,720,0\r\n\"12\",2015-05-19 08:50:00,21.51,21.93,21.45,21.92,720,0\r\n\"13\",2015-05-19 09:50:00,21.92,21.92,21.55,21.55,720,0\r\n\"

我需要将这个向量转换为可用格式，但是...

> read.table(bars.list, header = TRUE, sep = ",", quote = "", dec = ".")
Error in file(file, "rt") : cannot open the connection
In addition: Warning message:
In file(file, "rt") :
  cannot open file '"","times","open","high","low","close","numEvents","volume"
"1",2015-05-18 06:50:00,23.98,23.98,23.5,23.77,421,0
"2",2015-05-18 07:50:00,23.77,23.9,23.34,23.6,720,0
"3",2015-05-18 08:50:00,23.6,23.6,23.32,23.42,720,0
"4",2015-05-18 09:50:00,23.44,23.91,23.43,23.66,720,0

我不清楚为什么R告诉我某些连接无法打开，因为该对象已经作为函数的参数粘贴。 R显示带有警告标志的输出已经非常接近我所需的内容...

- Pat

Answer 1

这里有两个选项。第一个选项提供了修复当前代码的方法，第二个选项则探讨了一个更加简单高效的替代方案。

选项1：在`read.table（）`中的第一个参数是`file`。你正在从向量中读取而不是从文件中读取，因此需要使用`text`参数，并使用`text = bars.list`。

另外，我们可以先使用`gsub（）`除去所有引号，然后使用`read.csv（）`代替`read.table（）`，因为那里“header = TRUE”和“sep =“，”默认为“，”。

read.csv(text = gsub("\"", "", bars.list), row.names = 1)
#                  times  open  high   low close numEvents volume
# 1  2015-05-18 06:50:00 23.98 23.98 23.50 23.77       421      0
# 2  2015-05-18 07:50:00 23.77 23.90 23.34 23.60       720      0
# 3  2015-05-18 08:50:00 23.60 23.60 23.32 23.42       720      0
# 4  2015-05-18 09:50:00 23.44 23.91 23.43 23.66       720      0
# 5  2015-05-18 10:50:00 23.67 24.06 23.59 24.02       720      0
# 6  2015-05-18 11:50:00 24.02 24.04 23.32 23.33       720      0
# 7  2015-05-18 12:50:00 23.33 23.42 22.74 22.81       720      0
# 8  2015-05-18 13:50:00 22.79 22.92 22.49 22.69       720      0
# 9  2015-05-18 14:50:00 22.69 22.70 22.14 22.14       481      0
# 10 2015-05-19 06:50:00 21.09 21.49 20.82 21.47       421      0
# 11 2015-05-19 07:50:00 21.48 21.68 21.46 21.51       720      0
# 12 2015-05-19 08:50:00 21.51 21.93 21.45 21.92       720      0
# 13 2015-05-19 09:50:00 21.92 21.92 21.55 21.55       720      0

对我而言，使用read.csv()中的quote参数不如这种方法有效。

选项2：data.table包中的fread()也很好用。 它更快且代码更简洁。 不需要使用gsub()。 我们可以直接将bars.list放入其中并删除第一列。

data.table::fread(bars.list, drop = 1)

现在，由于最后的\"引号，您将收到此方法的警告。您可以忽略它或通过删除最后一个引号来获得无警告的结果。

data.table::fread(sub("\"$", "", bars.list), drop = 1)

数据：

bars.list <- "\"\",\"times\",\"open\",\"high\",\"low\",\"close\",\"numEvents\",\"volume\"\r\n\"1\",2015-05-18 06:50:00,23.98,23.98,23.5,23.77,421,0\r\n\"2\",2015-05-18 07:50:00,23.77,23.9,23.34,23.6,720,0\r\n\"3\",2015-05-18 08:50:00,23.6,23.6,23.32,23.42,720,0\r\n\"4\",2015-05-18 09:50:00,23.44,23.91,23.43,23.66,720,0\r\n\"5\",2015-05-18 10:50:00,23.67,24.06,23.59,24.02,720,0\r\n\"6\",2015-05-18 11:50:00,24.02,24.04,23.32,23.33,720,0\r\n\"7\",2015-05-18 12:50:00,23.33,23.42,22.74,22.81,720,0\r\n\"8\",2015-05-18 13:50:00,22.79,22.92,22.49,22.69,720,0\r\n\"9\",2015-05-18 14:50:00,22.69,22.7,22.14,22.14,481,0\r\n\"10\",2015-05-19 06:50:00,21.09,21.49,20.82,21.47,421,0\r\n\"11\",2015-05-19 07:50:00,21.48,21.68,21.46,21.51,720,0\r\n\"12\",2015-05-19 08:50:00,21.51,21.93,21.45,21.92,720,0\r\n\"13\",2015-05-19 09:50:00,21.92,21.92,21.55,21.55,720,0\r\n\""

- Rich Scriven