R-根据另一列中的另一行向下填充一列中的行

Question

I am reading in JSON data in RStudio that is coming from sensors driving around in a city. I am then converting this data to a dataframe with the sofa and jsonlite package. the data basically is sensor data, that consists of location data, and to each location measurement there are some environmental data transmitted and coded in resource paths like (/6/0/0 and /6/0/1 for latitude and longitude information) - due to the JSON data I am reading in, the location information in my R dataframe is in the same "value" column like the environmental data like humidity, CO2 etc. so I loose the location information for the individual observations, as the location info is also treated as value. see converted JSON data.frame below

data.frame

> |------------------------------------------------------- |    
> resourcePath    |     value  |UTC       |lat     |long . |
>   
> /6/0/0          |    48.18   |14:51:43 |  0     | 0      |
> |---------------|------------|---------|--------|--------|   
> /6/0/1          |    16.39   |14:51:43 |  0     | 0      |
> |---------------|------------|---------|--------|--------|   
> /3300/515/5700  |         34 |14:52:43 |  0     | 0      |
> |---------------|------------|---------|--------|--------|   
> /3300/289/5700  |         15 |14:53:43 |  0     | 0      |
> |---------------|------------|---------|--------|--------|   
> /3300/515/5700  |        55  |4:53:47  |  0     | 0      |
> |---------------|------------|---------|--------|--------|   
> /3300/289/5700  |       9004 |14:54:23 |  0     | 0      |
> |---------------|------------|---------|--------|--------|   
> /3304/0/5700    |       367  | 14:54:34| 0      |0       |
> |---------------|------------|---------|--------|--------|   
> /3315/0/5700    |         47 |14:54:54 | 0      |0       |
> |---------------|------------|---------|--------|--------|   
> /6/0/0          |     50.34  |14:57:11 |0       | 0      |
> |---------------|------------|---------|--------|--------|   
> /6/0/1          |     20.52  |14:57:13 |0       | 0      |
> |---------------|------------|---------|--------|--------|   
> /3304/0/5700    |         84 |14:57:34 |0       | 0      |
> |---------------|------------|---------|--------|--------|   
> /3315/0/5700    |         56 |14:57:45    0       0      |

And here a View of the desired dataframe.

[this is the desired df - each row has an associated "lat" and "long" information - depending on the value in the "value" column, as long as there is a new value in the "value" column.][2]

> |------------------------------------------------------- |    
> resourcePath    |     value  |UTC       |lat     |long . |
>   
> /6/0/0          |    48.18   |14:51:43 |  48.18 | 16.39  |
> |---------------|------------|---------|--------|--------|   
> /6/0/1          |    16.39   |14:51:43 |  48.18 | 16.39  |
> |---------------|------------|---------|--------|--------|   
> /3300/515/5700  |         34 |14:52:43 |  48.18 | 16.39  |
> |---------------|------------|---------|--------|--------|   
> /3300/289/5700  |         15 |14:53:43 |  48.18 | 16.39  |
> |---------------|------------|---------|--------|--------|   
> /3300/515/5700  |        55  |4:53:47  |  48.18 | 16.39  |
> |---------------|------------|---------|--------|--------|   
> /3300/289/5700  |       9004 |14:54:23 |  48.18 | 16.39  |
> |---------------|------------|---------|--------|--------|   
> /3304/0/5700    |       367  | 14:54:34| 48.18  |16.39   |
> |---------------|------------|---------|--------|--------|   
> /3315/0/5700    |         47 |14:54:54 | 48.18  |16.39   |
> |---------------|------------|---------|--------|--------|   
> /6/0/0          |     50.34  |14:57:11 |50.34   | 20.52  |
> |---------------|------------|---------|--------|--------|   
> /6/0/1          |     20.52  |14:57:13 |50.34   | 20.52  |
> |---------------|------------|---------|--------|--------|   
> /3304/0/5700    |         84 |14:57:34 |50.34   | 20.52  |
> |---------------|------------|---------|--------|--------|   
> /3315/0/5700    |         56 |14:57:45    50.34   20.52  |

I was looping - using lapply, but currently i am not getting the desired df. Any hints widely appreciated. Thomas

Answer 1

Here is a solution with the use of the tidyr package. This assumes the first row of each grouping of data is the "/6/0/0" row and the second row is "/6/0/1".

df<-structure(list(resourcePath = structure(c(5L, 6L, 2L, 1L, 2L, 
1L, 3L, 4L, 5L, 6L, 3L, 4L), .Label = c("/3300/289/5700", "/3300/515/5700", 
"/3304/0/5700", "/3315/0/5700", "/6/0/0", "/6/0/1"), class = "factor"), 
    value = c(48.18, 16.39, 34, 15, 55, 9004, 367, 47, 50.34, 
    20.52, 84, 56), UTC = structure(c(1L, 1L, 2L, 3L, 4L, 5L, 
    6L, 7L, 8L, 9L, 10L, 11L), .Label = c("14:51:43", "14:52:43", 
    "14:53:43", "14:53:47", "14:54:23", "14:54:34", "14:54:54", 
    "14:57:11", "14:57:13", "14:57:34", "14:57:45"), class = "factor"), 
    lat = c(0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L), 
    long = c(0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L)), .Names = c("resourcePath", 
"value", "UTC", "lat", "long"), class = "data.frame", row.names = c(NA, 
-12L))    

f$resourcePath<-as.character(df$resourcePath)

#reset lat and long columns to NA for the fill command
df$lat<-NA
df$long <- NA

#find rows with the lat resource
#assumes this is the first row of each data grouping
latrows<-which(df$resourcePath=="/6/0/0")
df$lat[latrows]<-df$value[latrows]
df$long[latrows]<-df$value[(latrows+1)]

library(tidyr)  #needed for the fill function
df<-fill(df, lat, long)

Edit note: This is a performance improvement over the initial version, provide the row ordering is consistent in the dataframe.