View Source Informative Row
The second row can be an informative row, to indicate the type and/or description of the variables. This is useful when the input and output are getting too long.
For each stub, either an input or output, the format is the same, as:
(type[, description])where type shall be one of the supported variable types, and description can be string,
or not specified (-).
examples
Examples
iex> table = Tablex.new("""
...> F Symptoms "Test Results" "Medical History" || Treatment Medication Follow-up
...> (string) (string, either NORMAL or ABNORMAL) (string) || (string) (string) (integer, in weeks)
...> 1 fever,cough NORMAL "No Allergies" || Rest OTC 2
...> 2 "Chest Pain" ABNORMAL ECG,"Family Hx of Heart Disease" || Hospital Prescription 1
...> 3 Headache NORMAL "Migraine Hx" || Rest Prescription 3
...> """)
...>
...> for %{name: name, type: type, desc: desc} <- table.inputs, do: {name, type, desc}
[
{:symptoms, :string, nil},
{:test_results, :string, "either NORMAL or ABNORMAL"},
{:medical_history, :string, nil}
]Descriptions are optional.
skipping-stubs
Skipping Stubs
In case you want to skip some field(s), "-" can be used, as:
iex> table = Tablex.new("""
...> F "Customer Segment" "Purchase Frequency" || "Campaign Type" Discount "Email Frequency"
...> - (integer, times/year) || - - -
...> 1 Loyal >8 || Personalized "20%" Monthly
...> 2 Occasional 2..8 || Seasonal "10%" Quarterly
...> 3 Inactive <2 || Re-engagement "15%" Biannually
...> """)
...>
...> for %{name: name, type: type, desc: desc} <- table.inputs, do: {name, type, desc}
[
{:customer_segment, :undefined, nil},
{:purchase_frequency, :integer, "times/year"}
]
...> Tablex.decide(table, customer_segment: "Inactive", purchase_frequency: 1)
%{campaign_type: "Re-engagement", discount: "15%", email_frequency: "Biannually"}