date: “last edit: 9/30/2019”

Load packages into R session. It will automatically load the package of dplyr and dbplyr.

library(SQLDataFrame)
library(DBI)

1 SQLDataFrame slots

dbfile <- system.file("extdata/test.db", package = "SQLDataFrame")
conn <- DBI::dbConnect(dbDriver("SQLite"), dbname = dbfile)
obj <- SQLDataFrame(
    conn = conn, dbtable = "state", dbkey = "state")
obj
#> SQLDataFrame with 50 rows and 4 columns
#>          state |           division        region population        size
#>    <character> |        <character>   <character>  <numeric> <character>
#>        Alabama | East South Central         South       3615      medium
#>         Alaska |            Pacific          West        365       small
#>        Arizona |           Mountain          West       2280      medium
#>       Arkansas | West South Central         South       2110      medium
#>     California |            Pacific          West      21198       large
#>            ... .                ...           ...        ...         ...
#>       Virginia |     South Atlantic         South       4981      medium
#>     Washington |            Pacific          West       3559      medium
#>  West Virginia |     South Atlantic         South       1799      medium
#>      Wisconsin | East North Central North Central       4589      medium
#>        Wyoming |           Mountain          West        376       small

1.1 Slot accessors

To make the SQLDataFrame object as light and compact as possible, there are only 5 slots contained in the object: tblData, dbkey, dbnrows, dbconcatKey, indexes. Metadata information could be returned through these 5 slots using slot accessors or other utility functions.

slotNames(obj)
#> [1] "dbkey"       "dbnrows"     "tblData"     "indexes"     "dbconcatKey"
dbtable(obj)
#> [1] "state"
dbkey(obj)
#> [1] "state"

1.2 tblData slot

The tblData slot saves the dbplyr::tbl_dbi version of the database table, which is a light-weight representation of the database table in R. Of note is that this lazy tbl only contains unique rows. It could also be sorted by the dbkey(obj) if the SQLDataFrame object was generated from union or rbind. So when the saveSQLDataFrame() function was called, a database table will be written into a physical disk space and have the unique records.

Accessor function is made avaible for this slot:

tblData(obj)
#> # Source:   table<state> [?? x 5]
#> # Database: sqlite 3.29.0
#> #   [/tmp/RtmpiO7GLs/Rinst665241634dc0/SQLDataFrame/extdata/test.db]
#>    division           region    state       population size  
#>    <chr>              <chr>     <chr>            <dbl> <chr> 
#>  1 East South Central South     Alabama           3615 medium
#>  2 Pacific            West      Alaska             365 small 
#>  3 Mountain           West      Arizona           2280 medium
#>  4 West South Central South     Arkansas          2110 medium
#>  5 Pacific            West      California       21198 large 
#>  6 Mountain           West      Colorado          2541 medium
#>  7 New England        Northeast Connecticut       3100 medium
#>  8 South Atlantic     South     Delaware           579 small 
#>  9 South Atlantic     South     Florida           8277 large 
#> 10 South Atlantic     South     Georgia           4931 medium
#> # … with more rows

1.3 dbnrows and dbconcatKey

The dbnrows slot saves the number of rows corresponding to the tblData, and dbconcatKey saves the realized (concatenated if multiple) key columns corresponding to the tblData. Accessor functions are also available for these 2 slots:

dbnrows(obj)
#> [1] 50
dbconcatKey(obj)
#>  [1] "Alabama"        "Alaska"         "Arizona"        "Arkansas"      
#>  [5] "California"     "Colorado"       "Connecticut"    "Delaware"      
#>  [9] "Florida"        "Georgia"        "Hawaii"         "Idaho"         
#> [13] "Illinois"       "Indiana"        "Iowa"           "Kansas"        
#> [17] "Kentucky"       "Louisiana"      "Maine"          "Maryland"      
#> [21] "Massachusetts"  "Michigan"       "Minnesota"      "Mississippi"   
#> [25] "Missouri"       "Montana"        "Nebraska"       "Nevada"        
#> [29] "New Hampshire"  "New Jersey"     "New Mexico"     "New York"      
#> [33] "North Carolina" "North Dakota"   "Ohio"           "Oklahoma"      
#> [37] "Oregon"         "Pennsylvania"   "Rhode Island"   "South Carolina"
#> [41] "South Dakota"   "Tennessee"      "Texas"          "Utah"          
#> [45] "Vermont"        "Virginia"       "Washington"     "West Virginia" 
#> [49] "Wisconsin"      "Wyoming"

1.4 indexes slot

The indexes slots is an unnamed list saving the row and column indexes respectively corresponding to the tblData slot, so that the SQLDataFrame could possibly have duplicate rows or only a subset of data records from the tblData, while the tblData slot doesn’t need to be changed. To be consistent, the slots of dbnrows and dbconcatKey will also remain unchanged.

obj@indexes
#> [[1]]
#> NULL
#> 
#> [[2]]
#> NULL
obj_sub <- obj[sample(5, 3, replace = TRUE), 2:3]
obj_sub
#> SQLDataFrame with 3 rows and 2 columns
#>        state |      region population
#>  <character> | <character>  <numeric>
#>      Arizona |        West       2280
#>      Arizona |        West       2280
#>      Arizona |        West       2280
obj_sub@indexes
#> [[1]]
#> [1] 3 3 3
#> 
#> [[2]]
#> [1] 2 3
identical(tblData(obj), tblData(obj_sub))
#> [1] TRUE

With a filter or select function (which is similar to [i, ] subsetting), only the indexes slot will be updated for the row or column index pointing to the tblData.

obj_filter <- obj %>% filter(division == "South Atlantic" & size == "medium")
obj_filter@indexes
#> [[1]]
#> [1] 10 20 40 46 48
#> 
#> [[2]]
#> NULL
identical(tblData(obj), tblData(obj_filter))
#> [1] TRUE

obj_select <- obj %>% select(division, size)
obj_select@indexes
#> [[1]]
#> NULL
#> 
#> [[2]]
#> [1] 1 4
identical(tblData(obj), tblData(obj_select))
#> [1] TRUE

2 SQLDataFrame methods

2.1 ROWNAMES

The ROWNAMES,SQLDataFrame method was defined to return the (concatenated if multiple) key column(s) value, so that the row subsetting with character vector works for the SQLDataFrame objects.

rnms <- ROWNAMES(obj)
obj[sample(rnms, 3), ]
#> SQLDataFrame with 3 rows and 4 columns
#>         state |           division      region population        size
#>   <character> |        <character> <character>  <numeric> <character>
#>     Louisiana | West South Central       South       3806      medium
#>  Rhode Island |        New England   Northeast        931       small
#>       Arizona |           Mountain        West       2280      medium

For SQLDataFrame object with composite keys:

obj1 <- SQLDataFrame(conn = conn, dbtable = "state",
                     dbkey = c("region", "population"))
ROWNAMES(obj1[1:10,])
#>  [1] "South:3615.0"     "West:365.0"       "West:2280.0"     
#>  [4] "South:2110.0"     "West:21198.0"     "West:2541.0"     
#>  [7] "Northeast:3100.0" "South:579.0"      "South:8277.0"    
#> [10] "South:4931.0"
obj1[c("South:3615.0", "West:365.0"), ]
#> SQLDataFrame with 2 rows and 3 columns
#>       region population |           division       state        size
#>  <character>  <numeric> |        <character> <character> <character>
#>        South       3615 | East South Central     Alabama      medium
#>         West        365 |            Pacific      Alaska       small

3 SessionInfo()

sessionInfo()
#> R version 3.6.1 (2019-07-05)
#> Platform: x86_64-pc-linux-gnu (64-bit)
#> Running under: Ubuntu 18.04.3 LTS
#> 
#> Matrix products: default
#> BLAS:   /home/biocbuild/bbs-3.10-bioc/R/lib/libRblas.so
#> LAPACK: /home/biocbuild/bbs-3.10-bioc/R/lib/libRlapack.so
#> 
#> locale:
#>  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
#>  [3] LC_TIME=en_US.UTF-8        LC_COLLATE=C              
#>  [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] parallel  stats4    stats     graphics  grDevices utils     datasets 
#> [8] methods   base     
#> 
#> other attached packages:
#> [1] DBI_1.0.0            SQLDataFrame_0.99.27 S4Vectors_0.23.25   
#> [4] BiocGenerics_0.31.6  dbplyr_1.4.2         dplyr_0.8.3         
#> [7] BiocStyle_2.13.2    
#> 
#> loaded via a namespace (and not attached):
#>  [1] Rcpp_1.0.2         pillar_1.4.2       compiler_3.6.1    
#>  [4] BiocManager_1.30.7 tools_3.6.1        zeallot_0.1.0     
#>  [7] digest_0.6.21      bit_1.1-14         memoise_1.1.0     
#> [10] evaluate_0.14      RSQLite_2.1.2      tibble_2.1.3      
#> [13] pkgconfig_2.0.3    rlang_0.4.0        cli_1.1.0         
#> [16] yaml_2.2.0         xfun_0.10          stringr_1.4.0     
#> [19] knitr_1.25         vctrs_0.2.0        bit64_0.9-7       
#> [22] tidyselect_0.2.5   glue_1.3.1         R6_2.4.0          
#> [25] fansi_0.4.0        rmarkdown_1.16     bookdown_0.14     
#> [28] purrr_0.3.2        blob_1.2.0         magrittr_1.5      
#> [31] backports_1.1.5    htmltools_0.4.0    assertthat_0.2.1  
#> [34] utf8_1.1.4         stringi_1.4.3      lazyeval_0.2.2    
#> [37] crayon_1.3.4