STRAVID - Building software collectively.

Facetbox

The filter used to browse the Zeichenfabrik program is built with PostgreSQL. During the implementation many things were learned. Facetbox is a research project to determine if alternative solutions are better suited for the problem space.

Possible API to build the index:

index = Facetbox.build({
  1: {
    term: '2019-summer',
    weekdays: [
      'monday',
      'tuesday',
    ],
  },
  2: {
    term: '2019-summer',
    weekdays: [
      'tuesday',
    ],
  },
  3: {
    term: '2019-winter',
    weekdays: [
      'tuesday',
      'friday',
    ],
  }
})

Possible API to query index:

index.find({
  term: [
    '2019-summer',
  ],
  weekdays: [
    'monday',
  ],
})

Possible result of a query:

{
  ids: [1],
  facets: {
    term: [
      { value: '2019-summer', count: 1 },
      { value: '2019-winter', count: 0 },
    ],
    weekdays: [
      { value: 'monday', count: 1 },
      { value: 'tuesday', count: 1 },
      { value: 'friday', count: 0 },
    ],
  },
}

Update 28.06.2019

The very basic prototype in Ruby works. While it's a apples to oranges comparison it's a lot faster than the PostgreSQL solution.

Right now it's unclear how to proceed. Invest more time into the prototype and figure out how to do time based filtering and the best way to integrate it into the application. On the other hand ressources could be put into the PostgreSQL solution to make it better.

You have questions, ideas or feedback? I like to hear from you and I'm looking forward to exhange thoughts. Please send an email to david@strauss.io and say “Hello”.

Random Page
Sitemap
Imprint
Privacy Policy