lookup modifies data in place

[krlmlr]

Something I would not expect from R. I really had to look twice, but unfortunately data.table doesn't seem to behave well here. #3 is related.

The code creates a lookup table from a larger data frame and checks that after lookup the order of the rows is unchanged. It isn't.

Find script in comment below.

[krlmlr]

There's a workaround: create a physical copy of subkey by e.g. coercing it to a matrix.

[krlmlr]

My bad -- the script is faulty. Will reopen as necessary.

[krlmlr]

This one's better. If you uncomment the line with %l+%, no error is shown. It looks like y is scrambled in-place.

library(magrittr)
library(qdapTools)

set.seed(123)
N <- 100

key <- data.frame(x = sample.int(N, N), y = 1:N, z = 1:N)
key$w <- key$x
key %>% head()

##    x y z  w
## 1 29 1 1 29
## 2 79 2 2 79
## 3 41 3 3 41
## 4 86 4 4 86
## 5 91 5 5 91
## 6  5 6 6  5

terms <- sample.int(2 * N, 1e2, replace = TRUE)

subkey <- key[c("x", "y")]

terms %l+% subkey %>% head

## [1] 120  75  12 191  41 179

stopifnot(key$x == key$w)
stopifnot(key$y == key$z)

## Error: key$y == key$z are not all TRUE

key$y

##   [1]  74  35  51  18   6  54  86  46  15  56  30  88  47  38  40  60  87
##  [18]  99  72  39  17  44  71  43  42  55  19  67   1  65  84  36  58  94
##  [35]  90  98  89  59  53  68   3  10  96  28  77  69  33  97  37   7   9
##  [52]  70  92  26  22 100  52  95  48  13  63  73  32  83  82  85  75  31
##  [69]  49  57  64  21  50  66  34  80  16  20   2  93  25  61   8  29  24
##  [86]   4  11  45  27  23   5  76  62  14  81  78  41  12  79  91

key$z

##   [1]   1   2   3   4   5   6   7   8   9  10  11  12  13  14  15  16  17
##  [18]  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34
##  [35]  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51
##  [52]  52  53  54  55  56  57  58  59  60  61  62  63  64  65  66  67  68
##  [69]  69  70  71  72  73  74  75  76  77  78  79  80  81  82  83  84  85
##  [86]  86  87  88  89  90  91  92  93  94  95  96  97  98  99 100

sessionInfo()

## R version 3.1.1 (2014-07-10)
## Platform: x86_64-pc-linux-gnu (64-bit)
## 
## locale:
##  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
##  [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
##  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
##  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 
##  [9] LC_ADDRESS=C               LC_TELEPHONE=C            
## [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       
## 
## attached base packages:
## [1] stats     graphics  grDevices utils     datasets  methods   base     
## 
## other attached packages:
## [1] qdapTools_1.0.2         magrittr_1.0.1          ProjectTemplate_0.5-1.3
## 
## loaded via a namespace (and not attached):
##  [1] chron_2.3-45     data.table_1.9.2 evaluate_0.5.6   formatR_1.0     
##  [5] knitr_1.6.18     packrat_0.4.1-1  plyr_1.8.1       Rcpp_0.11.2     
##  [9] RCurl_1.95-4.3   reshape2_1.4     stringr_0.6.2    tools_3.1.1

[krlmlr]

Reversing also helps: subkey <- key[c("x", "y")] %>% rev

[trinker]

Hmm this can be problematic. I asked on SO. I hope there's an easy way to address this: http://stackoverflow.com/questions/26046143/cut-scoping-link-between-key-and-data-frame-for-data-table

[trinker]

I think this has been fixed. If you feel it hasn't please re-open this issue. Thanks for the report.

[krlmlr]

Thanks for the quick fix!