Editing Rulesets

A ruleset consists of darl code and various ancillary items, such as input/output formatting, the text to display for data elicitation and ruleset metadata. The edit page is broken up into several tabbed areas that can be opened to reveal parts of the ruleset model.

ruleset editor

Rules

This DARL editor displays the full code associated with the ruleset. Details on using the DARL editor Various elements in the DARL code, for instance the inputs and outputs need to be synchronized between the different tabs. This occurs whenever the ruleset is saved. The DARL code is always the canonical source. ruleset edit rules

IO Format

This section permits you to edit how a ruleset input, and thus a potential question for the user, is formatted. ruleset edit io

Default questions

This determines the number of questions asked at a time. For bots this is 1. In other applications some other number might be chosen.

inputs

This displays the inputs listed in the DARL code. The names and types are read-only, but you can set other parameters that determine how the inputs are displayed and validation performed on them.

Edit increment

for numeric inputs this determines the incremental value used in a control that supports increments. The value '0' turns this off.

Max length

For textual inputs the maximum text length. The value '0' turns this off.

Min value

Minimum range value for a numeric input. The initial value of this is determined by the lowest bounds of any fuzzy sets specified for this input in the DARL code.

Max value

Maximum range value for a numeric input. The initial value of this is determined by the highest bounds of any fuzzy sets specified for this input in the DARL code.

Regular expression

Specifies a regular expression used to check the format of any text entered, for instance an email address or a phone number.

Show sets

For numeric inputs, where sets are defined, rather than asking for a numeric value, the user will be asked for one of the set names if this is true.

Outputs

This displays and edits formatting data for the outputs of a rule set. io outputs

Hide result

Outputs marked true are not reported back to the user.

Display Type

choices are

Text

The output value is reported as text

ScoreBar

The numeric output value is reported as a score bar, where supported.

The output is formatted as a link.

Redirect

The output is interpreted as the name of another ruleset to call. This is superseded by 'Call' stores.

Bar Color

An RGB value specifying the score bar color, where supported.

Bar max val

The upper bound of the score bar, where supported

Bar min val

The lower bound of the score bar, where supported.

Uncertainty

Since DARL is a fuzzy logic expert system, all inferences come with certainties and tolerances. Usually only the central or dominant value is reported. Selecting True will add this data where supported.

Value format

For numeric values this determines the numeric display format, for instance the number of decimal places, following the microsoft format.

Language

This displays and edits the text items associated with the inputs and outputs. language text The text field contains the text that will be displayed. The text items are determined by the ruleset and also contain a set of fixed items that determine the text that appears before questions are asked, and before results are supplied. these are:

Name Purpose
Format.preamble Text provided before the questions are presented
Format.questionHeader Text provided immediately before the first question
Format.resultHeader Text provided once the inference process is complete and before the results are displayed.

Categorical io

Where categorical inputs are specified, the displayed text for each is specified by the item with the name

<io name>.<category name in the code>

Variants

Where supported alternate texts in different languages can be provided.

Language

The ISO standard language

Text

The text in that language

Default language

The ISO standard language specifier that is used as default. If empty "en_US will be employed.

Preloaded Data

As a ruleset is run, the inference engine attempts to satisfy the outputs of the rule set by acquiring the value of the various inputs. You can preload the data that is supplied to one or more inputs.

Actions at completion

Test data

Metadata

Each ruleset can be annotated with a range of metadata that can be used to classify and potentially market rulesets.

Name

The name of the ruleset.

Description

A textual description of the rulesets purchase and features.

Version

Version of this copy of the rule set.

Author

The name of the author of the ruleset.

Copyright information for this ruleset.

License

License information, either the text of the license or reference to a standard license type like "MIT".

Image Url

An image hosted externally used in association with this ruleset. In future this may be used for a directory of rulesets.

Price

If supported on the channel used, it may be possible to charge for the use of this ruleset. Specify the numeric price you expect here. (numbers and decimal point only.)

Currency

The currency for the above price.

Save Rulest

Saves all changes to the current loaded ruleset model. The changes will only become permanent once the whole model is saved.

Run the Form

This has the same affect as selecting Run on the side bar.