Bouquet new release available
Bouquet is a open source analytics framework that allows you to expose, explore and export data.
We are looking for beta-testers. If you want to test: head to openbouquet.io and register to install your local docker container!
If you already have Bouquet installed using Docker, please check the instruction to upgrade while preserving your existing projects…
The RELEASE/ 12 Oct 2015 / Build 59
This new release introduces a new “Bouquet” theme for the UI. Expect more vibrant and yellowish colors, and a direct link to the support site. Note that the application URL got themed also: http://openbouquet.io/apps/release/bouquet/index.html?apiUrl=…
You can browse the support site directly from the Help widget, so you can easily follow the instruction while performing a task:
If you cannot find an answer don’t hesitate to leave us a message. We are working hard to provide you with rich content, and we will be happy to help you.
The Project creation wizard has been improved to give you better feedback while setting up your database connection.
Once the connection is defined and valid, Bouquet will automatically populate the data model with dynamic content for you so you can start exploring your data instantly. A dynamic object is prefixed with the symbol ~ (tilde). Note that only the project owner or the super user are able to see dynamic content. They will chose what to expose to the general users.
So here you can quickly have a look at the Actor domain and configure what you want to expose:
For example here I want to create a Full Name dimension that combines the actor first and last name, and make it availables for faceted search:
Here I quickly create a dimension that concatenate first name and last name, and make it searchable from the filter panel (by selecting indexed).
This new release provides also a revamped expression editor that give you more information regarding the type of the expression, and make creating complexe formula easier.
Since it is indexed the new Full Name dimension appears immediately in the Filters panel. You can use it to search the actors naturally. We fixed a bug in the search box that prevented you to perform a partial match. You can also use multiple tokens to filter.
Note that if the Actor is a very big table with millions of values, indexing will take some time. Hopefully Bouquet leverages Elastic capabilities and thus support indexing your data in the background, so you can start using the facet right away. The Filters panel will give you an indication of the index progress in that case.
Now that you exposed some of the Actor properties, the domain is no more dynamic. Hence it will be visible to any user (unless you set-up a specific access control rule for it). Also the Actor domain dropped the tilde (non-dynamic object will be always listed first):
Now let’s select the Film domain and configure the dimensions. I want to expose some of the available dimensions: Title, Rating and Description.
Also I want to be able to list the Actors participating in a given film. Because it is a Many-to-Many relation, Bouquet does not dynamically create it, but I can easily define a new dimension to list the film’s actors:
The formula to define the dimension traverses the film_actor domain to list the actors taking part to the film. This dimension actually creates a link between the Film and Actor domains, and will automatically expose all the Actor dimensions (and filters) in the Film domain.
Now let’s check which of “Alec Wayne” Film is the best selling one. Just head to the Payment domain and select the available properties: Film Title and Sum Amount.
Note that Bouquet automatically made the Film domain available from Payment because the DVDRental database defined a set of foreign-keys: Payment to Rental, Rental to Inventory and Inventory to Film. Also because we defined a way to filter Film by Actors, it is automatically available too.
Note that Bouquet also detected automatically that it could use a period dimension to help you filter the Payments. You can still change that by configuring a dimension.
That’s all folks!