tabde: tabde: Define and Check The Structure of Data Frames and...

In s-fleck/tabde: Define and Check The Structure of Data Frames and Database Tables

Description

Define the structure (column names and types) of tables and save them as csv files. Tools are provided to use such table design to create database tables, read fwf and csv files, and check data.frames for validity.

Table designs are data.frames that contain meta-info on other data.frames such as column names and types. Such table_designs can be used for validating the structure of data.frames, generating SQL code, reading files, etc...

Usage

tabde(
  col_name,
  col_type = rep(NA_character_, length(col_name)),
  col_domain = NULL,
  ...
)

tabde_fwf(
  col_name,
  col_type = rep(NA_character_, length(col_name)),
  fwf_start,
  fwf_end,
  col_domain = NULL,
  ...
)

tabde_sql(
  col_name,
  col_type = rep(NA_character_, length(col_name)),
  sql_type = rep(NA_character_, length(col_name)),
  sql_opts = rep(NA_character_, length(col_name)),
  col_domain = NULL,
  ...,
  .sql_header = NULL
)

Arguments

col_name	character vector. column names
col_type	character vector. Valid data.frame column types. NAs and the string "#skip" have special meanings. NAs will be converted to readr::col_guess() by as_col_spec() and matches_tabde() will not check the classes of NA columns. In a similar fashion #skip columns will be converted to readr::col_skip() and matches_tabde will not check if they are present in dat or not.
col_domain	character vector. A valid domain for col_name. Domains can be used by matches_tabde() and as_sql() to check for valid (discrete) values if a values Object is passed to these functions.
...	passed on to methods
fwf_start	integer vector. fwf start positions
fwf_end	integer vector. fwf end positions
sql_type	character vector. SQL Data Types as supported by target DBMS System. Columns with col_type NA will be skipped when creating SQL statements.
sql_opts	character vector. SQL Options to be used by as_sql() (for example NOT NULL)
.sql_header	a sql_header() data.frame

Value

a data.frame of class table_design

Database constraints

A table design can contain a sql_header() object that represents database constraints (such as primary key) as the attribute 'sql_header'. If such an attribute is present, as_sql() will use it when transforming a table design to an SQL CREATE TABLE statement.

Storing table_designs

table_desings are designed to be stored as .csv files. This has some disadvantages over storing them in a binary format such as .rds or .rda, but makes it easy to edit them and track changes in a VCS such as git.

Author(s)

Maintainer: Stefan Fleck [e-mail]

See Also

• validate data.frames with matches_tabde()

• read csv files with td_fread()

• generate SQL with as_sql

• shed: An experimental csv editor implemented in shiny

Examples

tabde_sql(
   col_name = c("first_name", "last_name", "age"),
   col_type = c("character", "character", "integer"),
   sql_type = c("varchar(255)", "varchar(255)", "integer"),
    .sql_header = sql_header(
      "person_pk",
      "primary key",
      list(c("first_name", "last_name"))
    )
 )

s-fleck/tabde documentation built on April 4, 2021, 10:11 p.m.