[R] Creating one df from 85 df present in a list

[Rasmus Liland]

On 2020-06-13 01:54 +0200, Rasmus Liland wrote:
> Dear Alejandro,

Sorry, I programmed and wrote that email at 
the same time, changed the “toma_de_muestras” 
perhaps other things, then continued 
programming, thus this might make more 
sense ...

Firstly, it would be easier to help you if 
you provided some example of how 
fusion_de_tablas looks like.

In this first example, I create a small list 
of oddly shaped data.frames which might look 
like your 85-element-long list.  Then, 
determining the unique colnames.  Lastly, 
applying my way through the list again to 
fill in N/A in the columns not there, so the 
do.call function recieves what it expects ... 

	red_tablas_por_punto <-
	  list(
	    "1 - Bv.Artigas y la Rambla (Terminal CUTCSA)" =
	      data.frame("a"=1:3,
	                 "b"=4:6,
	                 "c"=4:6,
	                 'toma_de_muestras'=10001:10003),
	    "10 - Avenida Millán 2515 (Hospital Vilardebó)" =
	      data.frame("d"=4:8,
	                 "b"=8:12,
	                 'toma_de_muestras'=10004:10008),
	    "100 - Fauquet 6358 (Hospital Saint Bois)" =
	      data.frame("e"=100:101,
	                 "a"=85:86,
	                 'toma_de_muestras'=10009:10010)
	  )
	unified.df <- lapply(names(red_tablas_por_punto),
	  function(tabla, cn) {
	    x <- red_tablas_por_punto[[tabla]]
	    x[,cn[!(cn %in% colnames(x))]] <- NA
	    x <- x[,cn]
	    x$tabla <- tabla
	    return(x)
	  }, cn=unique(unlist(lapply(red_tablas_por_punto, colnames))))
	unified.df <- do.call(rbind, unified.df)
	unified.df

yields this:

	    a  b  c toma_de_muestras  d   e                                         tabla
	1   1  4  4            10001 NA  NA  1 - Bv.Artigas y la Rambla (Terminal CUTCSA)
	2   2  5  5            10002 NA  NA  1 - Bv.Artigas y la Rambla (Terminal CUTCSA)
	3   3  6  6            10003 NA  NA  1 - Bv.Artigas y la Rambla (Terminal CUTCSA)
	4  NA  8 NA            10004  4  NA 10 - Avenida Millán 2515 (Hospital Vilardebó)
	5  NA  9 NA            10005  5  NA 10 - Avenida Millán 2515 (Hospital Vilardebó)
	6  NA 10 NA            10006  6  NA 10 - Avenida Millán 2515 (Hospital Vilardebó)
	7  NA 11 NA            10007  7  NA 10 - Avenida Millán 2515 (Hospital Vilardebó)
	8  NA 12 NA            10008  8  NA 10 - Avenida Millán 2515 (Hospital Vilardebó)
	9  85 NA NA            10009 NA 100      100 - Fauquet 6358 (Hospital Saint Bois)
	10 86 NA NA            10010 NA 101      100 - Fauquet 6358 (Hospital Saint Bois)

... right, so you could also use merge with 
Reduce like in that stackoverflow answer [1], 
which might have been what you were looking 
for anyway:

	Reduce(function(x, y)
	  merge(x, y, by='toma_de_muestras', all=T),
	  red_tablas_por_punto)

yields this:

	   toma_de_muestras a.x b.x  c  d b.y   e a.y
	1             10001   1   4  4 NA  NA  NA  NA
	2             10002   2   5  5 NA  NA  NA  NA
	3             10003   3   6  6 NA  NA  NA  NA
	4             10004  NA  NA NA  4   8  NA  NA
	5             10005  NA  NA NA  5   9  NA  NA
	6             10006  NA  NA NA  6  10  NA  NA
	7             10007  NA  NA NA  7  11  NA  NA
	8             10008  NA  NA NA  8  12  NA  NA
	9             10009  NA  NA NA NA  NA 100  85
	10            10010  NA  NA NA NA  NA 101  86

Best,
Rasmus

[1] https://stackoverflow.com/questions/22644780/merging-multiple-csv-files-in-r-using-do-call

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: not available
URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20200613/4defe719/attachment.sig>