如何`data.frame`具有不同的行数但相关（不是`by`）

Question

Here is the sample of the XML format in my dataset.

<info>
    <a>1990-01-02T06:58:12+08:00</a>
    <b>120.980</b>
    <c>23.786</c>
    <d>18.7</d>
    <e>2</e>
</info>
<info>
    <a>1990-02-02T06:58:12+08:00</a>
    <b>120.804</b>
    <c>23.790</c>
</info>

But the numbers of tag is not same as tag , for example there are 4000 rows tag a, b, c, and only 3950 rows for tag d, e

Here is my code in R

library(xml2)

data.frame(Time = xml_text(xml_find_all(xml_data, ".//a")),
           Num = xml_text(xml_find_all(xml_data, ".//b")),
           Dist = xml_text(xml_find_all(xml_data, ".//c")),
           Gap = xml_text(xml_find_all(xml_data, ".//d")),
           Type = xml_text(xml_find_all(xml_data, ".//e")),
           stringsAsFactors = F)
}) -> df

The error message is: (I knew this will happened)

arguments imply differing number of rows

The output I want will be like the table below:

Time                       Num      Dist   Gap   Type
1990-01-02T06:58:12+08:00  120.980  23.786 18.7  2
1990-02-02T06:58:12+08:00  120.804  23.790 <NA>  <NA>
...
1993-03-03T08:42:15+08:00  120.412  23.523 <NA>  1

Which function or library should I try for this?
Thanks for helping me !!

I have tried some another method like map_if

Answer 1

Finally I found the solution!!

Once we are using the xml file, be sure to get the root node of the records at first.

Here I will show you how it works.

Take this xml file for example: (name it to test.xml)

<dataset>
  <dataset_info>
    <data_count>2</data_count>
    <status>Actual</status>
  </dataset_info>
  <data>
    <time>2019-06-01</time>
    <event>event1</event>
    <describe>describe for event1</describe>
  </data>
  <data>
    <time>2019-06-02</time>
    <event>event2</event>
  </data>
</dataset>

We know that there is a tag describe missing in event2, but we hope to make data frame by this xml data. I was taught to use the function xml2::xml_find_all to get the value in the selected tag. By the R code like this:

# library import
library(xml) #require(xml2)

# file reading
xml <- read_xml("path/where/the/file/is/test.xml")


data.frame(Time = xml_text(xml_find_all(xml, ".//time"))
           Event = xml_text(xml_find_all(xml, ".//event"))
           Describe = xml_text(xml_find_all(xml, ".//describe"))
           )

Then we will get error message arguments imply differing number of rows

So what we need to do is get the root of records first!! As the code below:

# library import
library(xml) #require(xml2)

# file reading
xml <- read_xml("path/where/the/file/is/test.xml")
record <- xml_find_all(xml, ".//data")


data.frame(Time = xml_text(xml_find_all(record, ".//time"))
           Event = xml_text(xml_find_all(record, ".//event"))
           Describe = xml_text(xml_find_all(record, ".//describe"))
           )

After adding record <- xml_find_all(xml, ".//data") , we will no longer get the error cause by different counting of the results.

Hope this can help !!