42 ≈ [ lexem( lexicon, name, arguments ), ... ]
Really small, simple, incredibly powerful and super fast Descriptive Internal DSLs tools.
"A Babelut(t)e is a sort of long toffee flavoured with honey or vergeoise (demerara sugar) from [...] Flanders, Belgium".
Etymology : "[A french word that] is likely to come from the Flemish "babbelen", speaking a lot, and "uit", finished because when you eat the toffee, you cannot speak anymore (either because you are enjoying it or because you cannot open the mouth).[...]" - src : wikipedia
Babelute.js core library (this lib) provides helpers to :
- define and manage Internal DSLs for modeling any kind of problems
- with a simple (meta)grammar - based on Method Chaining - for writing structured unambiguous "sentences" (called babelutes)
- and to provide ways :
- to manage dictionaries of related lexems (called lexicons) that form DSL semantic fields and their Internal Denotation
- to translate DSL sentences to other DSL sentences (by example through dedicated bridge-DSLs with their own lexicon(s))
- to interpret sentences in many context with specific fine grained dedicated implementations (called pragmatics)
The idea is really simple : lets write structured sentence with some DSL to encapsulate information (aka modeling). Then use different "interpretation" engines to do something with this information (aka implementations).
- babelute (this lib) : provides Lexicon and Pragmatics base for writting DSLs.
- babelute-uus : Universal Unambiguous Sentences proposal : Welcome in Sharing Era. For storing and sharing sentences.
- babelute-ldl : Babelute Lexicon Definition DSL (Sentences Schema) and its generators.
Low Level DSLs (Developement related domains) :
- aright-* : Objects and types validation DSL (ultra-fast, ultra-modular)
- htsl-* : HTML5 DSL and its render engines. (modern, one of the world's fastest, one-way-binding templating (React philosophy))
High Level DSLs (Human related domains) :
- babelute-cooking : High Level Cooking DSL demo and its bunch of transformations and DSLs targets. (realease in summer 2017)
Please read Designing a DSL for more infos.
Theorical considerations are exposed here (work in progress).
npm i babelute --save
The aim is to define Descriptive Internal DSLs :
- Descriptive : Write what you want and not how to achieve it
- Internal : made and handleable with pure host-language objects and syntax (v.s. an External DSL that has it's own particular syntax)
- DSL : a bunch of related words (and so concepts) from a particular domain
By defining Words as Functions that receive arguments :
.aWords(arg1, arg2, ...)And to write structured sentences with them through Method Chaining :
h.this().is().a(h.sentence(true)) // this is an example : there is no such DSL... ;)How to get that :
// my-dsl-lexicon.js
import babelute from 'babelute';
/* A lexicon is where to store your words */
const lexicon = babelute.createLexicon('my-dsl');
/*
Atoms are words (of your DSL) that are not expressed with other words from the same lexicon (aka Domain).
In other words, they are words that should be expressed precisely
with some other domains concepts for particular contexts.
In other words, they need translation or implementation that are outside the scope of the current Domain.
(see designing-a-dsl in babelute's doc for more infos)
*/
lexicon.addAtoms([
'foo', // .foo(...args)
'bar', // .bar(...args)
'zoo', // .zoo(...args)
'doo' // .doo(...args)
]);
// don't worry, there is strict formal ways of defining allowed atoms arguments
/* Compounds words are words (of your DSL) that are expressed with other words from the same lexicon */
lexicon.addCompounds((h) => {
// h is the lexicon's initializer (see below)
return {
goo(arg1, arg2) {
// this is called the "Internal Denotation" of the compound lexem (here 'goo')
return this.foo(arg1).bar(arg2);
},
boo(...some){
return this.zoo(h.doo(some).foo('lollipop'));
}
};
});
export default lexicon;import lexicon from 'my-dsl-lexicon';
/* initializer are just a helper to start sentence with your lexicon */
const h = lexicon.initializer();
/* then write sentences with your DSL to describe what you want */
const mySentence = h.goo('hello', 'world').boo('one', 'two', 'three').foo(true);Compounds words (here goo and boo) are made of atoms (or of other compounds words that are themselves made of atoms by recursivity), and so sentences below are exactly equivalent (they hold the same lexems list and structure):
const mySentence = h.goo('hello', 'world').boo('one', 'two', 'three').foo(true);
// is deeply equal to
const mySentence2 =
h.foo('hello')
.bar('world')
.zoo(
h.doo(['one', 'two', 'three'])
.foo('lollipop')
)
.foo(true);So sentences are finally always composed of atoms (those words that need implementation - see below).
Of course, you are totally free to create any Internal DSL you want. Apart they should be "Description Oriented" (it is babelute's purpose), you could imagine whatever you want... Remember just that the aim is to define descriptive DSL to catch informations.
So, until here, we have just described things (so we've stored information and knowledge in sentences - ok, imagine that we have ;) and we haven't defined yet any mean to interpret them and to make them useful.
So we need to define an "interpretation" engine which is called here a pragmatics engine.
The cool fact is that, basically, as sentences are always made of atoms, we just need to provides implementation for our atoms, whatever the number of compounds words we have.
Let's define a simple engine that use sentences made with my-dsl to decorate an object (called subject below) :
// my-dsl-to-object-pragmatics.js
const myPragmas = {
foo(subject, args /* foo's args received in sentence */){
// do something on subject with args
// ...
},
bar(subject, args /* bar's args received in sentence */){
// do something on subject with args
// ...
},
zoo(subject, args /* zoo's args received in sentence */){
subject.zoo = {};
if(args[0].__babelute__)
this.$output(args[0], subject.zoo);
else
...
},
doo(subject, args /* doo's args received in sentence */){
// do something on subject with args
// ...
},
// by convention, the method's name used for interpretation start with a '$'
$output(babelute, subject = {}){
babelute._lexems.forEach((lexem) =>
lexem.lexicon === 'my-dsl'
&& this[lexem.name]
&& this[lexem.name](subject, lexem.args)
// simple mapping between lexem's name and own methods
);
return subject;
}
};
export default myPragmas;import lexicon from 'my-dsl-lexicon';
import myDSLToObject from 'my-dsl-to-object-pragmatics';
const h = lexicon.initializer(),
subject = {},
sentence = h.goo('hello', 'world').boo(['one', 'two', 'three']);
myDSLToObject.$output(sentence, subject);Straight forward...
Again, of course you are totally free to interpret sentences exactly as you want. There is no constraints. And you could imagine many differents output kind and usage (from simple Facade for manipulating objects states, to full Code Generation Tools). Imagination is the limit.
Finally, lets define a dialect of my-dsl :
// my-dsl-dialect-lexicon.js
import myDSLLexicon from 'my-dsl-lexicon';
const myDialectLexicon = myDSLLexicon.createDialect('my-dsl-dialect');
myDialectLexicon.addCompounds((h) => ({
myNewWord(...){
return this.foo('...').goo(...);
},
myOtherNewWord(...){
return this.boo('...').bar(...).doo();
},
...
}));
export default myDialectLexicon;import lexicon from 'my-dsl-dialect-lexicon';
const h = lexicon.initializer();
const mySentence = h.myOtherNewWord('hello', 'world')...;
...
// of course always usable with pragmatics that work with my-dsl
import myDSLToObject from 'my-dsl-to-object-pragmatics';
const decoratedObject = myDSLToObject.$output(mySentence);Please read Designing a DSL for more infos.
The MIT License
Copyright 2016-2017 (c) Gilles Coomans gilles.coomans@gmail.com
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the 'Software'), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED 'AS IS', WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.