| dm_enum_fk_candidates | R Documentation |
Determine which columns would be good candidates to be used as foreign keys of a table,
to reference the primary key column of another table of the dm object.
dm_enum_fk_candidates(dm, table, ref_table, ...)
enum_fk_candidates(dm_zoomed, ref_table, ...)
dm |
A |
table |
The table whose columns should be tested for suitability as foreign keys. |
ref_table |
A table with a primary key. |
... |
These dots are for future extensions and must be empty. |
dm_zoomed |
A |
dm_enum_fk_candidates() first checks if ref_table has a primary key set,
if not, an error is thrown.
If ref_table does have a primary key, then a join operation will be tried using
that key as the by argument of join() to match it to each column of table.
Attempting to join incompatible columns triggers an error.
The outcome of the join operation determines the value of the why column in the result:
an empty value for a column of table that is a suitable foreign key candidate
the count and percentage of missing matches for a column that is not suitable
the error message triggered for unsuitable candidates that may include the types of mismatched columns
enum_fk_candidates() works like dm_enum_fk_candidates() with the zoomed table as table.
A tibble with the following columns:
columnscolumns of table,
candidateboolean: are these columns a candidate for a foreign key,
whyif not a candidate for a foreign key, explanation for for this.
These functions are marked "experimental" because we are not yet sure about
the interface, in particular if we need both dm_enum...() and enum...()
variants.
Changing the interface later seems harmless because these functions are
most likely used interactively.
Other foreign key functions:
dm_add_fk(),
dm_get_all_fks(),
dm_rm_fk()
dm_nycflights13() %>%
dm_enum_fk_candidates(flights, airports)
dm_nycflights13() %>%
dm_zoom_to(flights) %>%
enum_fk_candidates(airports)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.