JavaScript Propositional Laboratory

Provides support for the JavaScript Propositional Laboratory Format (JSPL), a format developed for Ecma-/JS-Developers, to be used for discussing contentious features of proposals.

Available on the VSCode Marketplace

Based on Langium

Getting started

Laboratory Information

laboratory {
    title "The title of the generated laboratory"
    description "This laboratory is for x"
    icon "./path/to/favicon.svg or some url" 
    format MD
    author "Philipp Riemer"
    version "1.0"
}

Concerns

concern name {
    summary "a short summary"
    description "a more detailed description of the concern, possibly containing html elements."
}

Proposition

proposition name {
    expression "what Expression to display"
    default value True
    value False {
        raise someConcern when someCondition is True
    }
    value "some custom value"
    disabled {
        message "this is disabled, because ..." when someCondition is True and someOtherProposition is "undefined"
    }
}

proposition given {
    expression "this is given"
    value True
}

Conditions

condition name holds when someProposition is "value1" and someOtherCondition is False

Markdown support

Multiline strings (laboratory description, concern description) are by default interpreted as markdown. Because markdown is indentation-sensitive, it is important to always indent the descriptions to the level of the original node. This following example of a concern is correctly indented:

concern test {
    summary "CONCERN!"
    description "
    
    # This is concerning.
            
    **Very concerning!** 

    *Or is it?*

        \{
            int i = 0;
            i++;
            return i;
        \}
    "
}

The usage of markdown can be disabled in favour of direct html-interpretation on a general basis by changing the format in the Laboratory Information or on a per-description basis by prefixing a string with "HTML":

concern test {
    summary "summary"
    description HTML "
        <p>This is my concern</p>
    "
}

Optimizer (Experimental)

WARNING: This feature is still experimental and not very user-friendly.

The VSCode-Extension allows to create an experimental optimizer from the command palette. This optimizer has the original laboratory included, but comes with multiple new "tabs". In the weights tab, penalties can be assigned to the different concerns to indicate "how concerning" they are. In the optimize tab the optimize button can be clicked to automatically find an optimal selection of the values for every tweakable (proposition). This however requires a local python server to be run, that does the actual solving of the problem. Such a server can be found in the git repository in the file "scip-server". Using the optimizer might require you to disable CORS-checks, if the python server is run locally.