2016-03-22 -

[andkov]

Notes during meeting

• ds - used outside functions
• d - used within the scope of a function

[wibeasley]

@andkov Here's an example of stacking/combining a list of data.frames. If I were doing it again, I'd probably use the .id parameter in rbind_all(), instead of this line: ds_list[[i]]$source_table <- block_names[i].

regex_table_name       <- "^(\\w{3})(\\w{3})Blocks(\\d{1,2})$"

# ---- load-data ---------------------------------------------------------------
... 
table_names <- RODBC::sqlTables(channel_access)$TABLE_NAME
block_names <- grep(regex_table_name, table_names, perl=T, value=T)

ds_list <- list()
for( i in seq_along(block_names) ) {
  ds_list[[i]] <- RODBC::sqlFetch(channel_access, sqtable=block_names[i], stringsAsFactors=F)
  ds_list[[i]]$source_table <- block_names[i]
}

rm(i, table_names, block_names)

# ---- tweak-data --------------------------------------------------------------
ds <- ds_list %>% 
  dplyr::rbind_all() %>% 
  dplyr::select_(
    "id"          = "ID",
    "row_id"      = "rowID",
    "block_value" = "BlockValue",
    "used"        = "Used",
    "source_table"
  ) %>% 
  dplyr::mutate(
    record_id     = seq_len(n()),
    id            = as.integer(id),
    used          = as.logical(used),
    value_1       = sub(regex_table_name, "\\1", source_table),
    value_2       = sub(regex_table_name, "\\2", source_table),
    value_3       = as.integer(sub(regex_table_name, "\\3", source_table))
  )