Paste your program in the box in the left and type a Prolog goal below. When you press ENTER on the goal textbox, the interpreter will read the goal and try to find a computed answer, showing the result here. If you press ENTER again (to keep looking for answers), the interpreter will continue looking from the last choice point.>
Look at built-in predicates and modules supported by Tau Prolog.
Tau Prolog is not just an implementation of the logic programming resolution mechanism, but a complete Prolog interpreter. Furthermore, the implementation of this interpreter has been directed by the ISO Prolog Standard.
Tau Prolog is released under the BSD 3-Clause License. This means that anyone is free to use, download, modify and share any version of this project. Tau Prolog has been developed by a team of students from UCLM (University of Castilla-La Mancha, Spain) as an open source project. See Collaborate.
Using the Tau Prolog library is simple. Just include a script tag in your html file, like this.
That's it! When the page loads, the window object is set with a pl object, which contains the logic related to Prolog.
In order to really use the library, you must create a Session object, which will contain the methods for parsing programs, looking for answers, etc. This way, we'll use Session.consult() to read a program, Session.query() to determine the goal we want to satisfy and Session.answer() to check if the goal is satisfied within the present database, and what values make it true if there were any variables on the goal. For example:
var session = pl.create(); session.consult( " likes(sam, salad). likes(dean, pie). likes(sam, apples). likes(dean, whiskey). " );
By calling the pl.create() method, we create a Session object, which contais a consult() method. This method receives the Prolog program. To set the goal we want to check, we need to call the query() method.
session.query( "likes(sam, X)" );
Now, when we call the answer() method on the session variable, the interpreter tries to compute an answer for the goal, looking for facts or rules which unify with the goal. If the search succeeds, a Substitution object is passed to the callback provided to answer() as an argument. This Substitution object contains the variables of the goal and their values. Using a pile of states, the interpreter remembers the last choice point, so it can continue looking for facts from that point in a future search. If there is not any computed answer, the callback invoked on answer() will return false.
If answer() is called again after having unified the goal previously, the interpreter will continue looking for answers starting from the last choice point.
var callback = console.log; session.answer(callback); // X = salad ; session.answer(callback); // X = apples ; session.answer(callback); // false.
For further information, check the Manual. It contains information about every function in the library, as well as code snippets to try them.
On downloading the library, you can customize the download so you only get the funcionality you need. You can download the whole library, or maybe you just the core and the list module, or maybe you already had the library and only need a specific module, etc.
The different predicates and modules available in this interpreter are documented on the Documentation section. Even though most of the elements included in this interpreter were implemented following the ISO Prolog Standard, some of them have been modified according to the team judgement. Those differences between the way something was implemented and the ISO Prolog Standard are documented as well.