Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/vsm/vsm-dictionary-local

Local (in-memory), full implementation of a VSM-dictionary
https://github.com/vsm/vsm-dictionary-local

dictionary identifiers in-memory ontology-search terms vsm

Last synced: 20 days ago
JSON representation

Local (in-memory), full implementation of a VSM-dictionary

Awesome Lists containing this project

README 

        
# vsm-dictionary-local







## Summary



`vsm-dictionary-local` is a full, local (=in-memory) implementation

of the 'VsmDictionary' parent-class/interface (from the package

[`vsm-dictionary`](https://github.com/vsm/vsm-dictionary)).







## Background



- [VSM-sentences](http://scicura.org/vsm/vsm.html)

  are built from terms that are linked to identifiers.

- The '[`vsm-dictionary`](https://github.com/vsm/vsm-dictionary)'

  package defines a standardized interface,

  for VSM-related tools to communicate with services

  that provide terms+IDs (e.g. a webserver API).

- That packages also includes the 'VsmDictionary' parent class that provides

  some shared functionality for concrete subclasses (like this package).







## A local implementation of a VsmDictionary



This is VsmDictionaryLocal:



- It extends the VsmDictionary parent class, and provides

  a concrete implementation that fully implements the vsm-dictionary

  [specification](https://github.com/vsm/vsm-dictionary/blob/master/Dictionary.spec.md).

  - i.e. it has the complete Create, Read, Update, Delete, search,

    sort, filter, etc. functionality for terms, subdictionary-info objects, etc.

- It does not use an online server for data storage or lookup.

  Instead, it stores all data in-memory, as long as the application is running.







## When to use VsmDictionaryLocal



Because of the above:



- During the development of new tools that depend on a VsmDictionary,

  this module can be used as a fully functional placeholder

  that does not need an online server.

- Or, it could provide mock terms+ids while running

  standalone demos of VSM-sentence building tools.

- The many automated tests in VsmDictionaryLocal can give inspiration

  for testing future, webserver-connecting implementations.







## Specification



Apart from following the parent class

[specification](https://github.com/vsm/vsm-dictionary/blob/master/Dictionary.spec.md),

VsmDictionaryLocal follows the additional the spec described in

[DictionaryLocal.spec.md](DictionaryLocal.spec.md).  

• 

(Note: we simply use the name 'DictionaryLocal' for VsmDictionaryLocal,

in the spec & source code).  







## Installation



### Node.js



```

npm install vsm-dictionary-local

```



```

const VsmDictionaryLocal = require('vsm-dictionary-local');

```



### Browsers



```



```

after which it is accessible as the global variable `VsmDictionaryLocal`.







## Example use in Node.js



Example that (only):  

 • adds one subdictionary (by adding a dictionary-info object),  

 • adds entries to it (=concepts/IDs + terms),  

 • string-searches for matching terms:



```javascript

const VsmDictionaryLocal = require('vsm-dictionary-local');



var dict = new VsmDictionaryLocal();

var dictInfos = [ { id: 'DictID_12', name: 'Example subdictionary' } ];

var entries = [

  { id: 'URI:001', dictID: 'DictID_12', terms: [{str: 'aaa'}, {str: 'synonym'}] },

  { id: 'URI:002', dictID: 'DictID_12', terms: [{str: 'aab'}] },

  { id: 'URI:003', dictID: 'DictID_12', terms: [{str: 'abc'}], descr: 'description' }

];



dict.addDictInfos(dictInfos, (err) => {  // Add 1 subdictionary-info object.

  dict.addEntries(entries, (err) => {    // Add 3 entries.

    dict.getMatchesForString('ab', {}, (err, res) => {  // Query for string 'ab'.

      console.dir(res.items, { depth: 3 });

    });

  });

});

```



This gives the output:



```javascript

[ { id: 'URI:003',         // Concept-ID.

    dictID: 'DictID_12',   // Dictionary-ID.

    descr: 'description',  // Description of the meaning of concept `URI:003`.

    terms: [ { str: 'abc' } ],  // Term-objects: one term as an unstyled string.

    str: 'abc',            // The term-string that this match pertains to.

    type: 'S' },           // Match type. Prefix(S)-matches come before infix(T).

  { id: 'URI:002',

    dictID: 'DictID_12',

    terms: [ { str: 'aab' } ],

    str: 'aab',

    type: 'T' } ]

```






Or we can load all data in one synchronous call, by giving it to

VsmDictionaryLocal's constructor:



```javascript

const VsmDictionaryLocal = require('vsm-dictionary-local');



// Create.

var dict = new VsmDictionaryLocal({

  dictData: [

    { id: 'DictID_12',

      name: 'Example subdictionary',

      entries: [

        { id: 'URI:001', terms: [{str: 'aaa'}, {str: 'synonym'}] },

        { id: 'URI:002', terms: [{str: 'aab'}] },

        { id: 'URI:003', terms: [{str: 'abc'}], descr: 'description' }

      ]

    },

  ],

});



// Query.

dict.getMatchesForString('ab', {}, (err, res) => {

  console.dir(res.items, {depth: 3});

});

```



which gives the same output.







## Tests



Run `npm test`, which runs tests with Mocha.  

Run `npm run testw`, which automatically reruns tests when any file changes.







## Demo in Node.js



More examples are included in [demoInNode.js](demo/demoInNode.js)

(based on [demoData.js](demo/demoData.js)).  

Run it with: `node demo/demoInNode.js`.







## Interactive demo in the browser



Run `npm run demo` to start an interactive demo of (only)

the string-search functionality, based on example data.  

This opens a browser page with an input-field to search

on [demoData.js](demo/demoData.js).



The demo works by making a Webpack dev-server bundle all source code 

(VsmDictionaryLocal and its dependencies) and serve it to the browser.  

This is useful during development, to see immediate effects of changes

to the source code (excl. the demo code).



(For normal use in browsers, just include the browser-build via a

<script>-tag, see above).







## License



This project is licensed under the AGPL license - see [LICENSE.md](LICENSE.md).