# Display API information without storing it
**URL:** https://discuss.tryton.org/t/display-api-information-without-storing-it/9282
**Category:** Developer
**Created:** [July 9, 2026, 2:30pm UTC](https://discuss.tryton.org/t/display-api-information-without-storing-it/9282 "2026-07-09T14:30:35Z")
**Posts on this page:** 6
**Page:** 1
### Author:  [@German](https://discuss.tryton.org/u/German)
#### Post date: [July 9, 2026, 2:30pm UTC](https://discuss.tryton.org/t/display-api-information-without-storing-it/9282/1 "2026-07-09T14:30:35Z")
Hello,
I am looking for some architectural guidance on a Tryton implementation.
I have two external API endpoints:
1. The first endpoint returns a list of car models for a specific brand.
2. The second endpoint returns detailed information for a specific model selected from that list.
My goal is to display this information inside Tryton without persisting any data to the database. Ideally, the workflow would look like this:
- The user types a brand (e.g., “BMW”).
- Tryton displays a tree view with the corresponding rows (e.g., X1, M3, Z4).
- When the user selects a row, Tryton fetches and displays the detailed view for that specific model.
Which components would best suit this requirement? Is it possible to use a `ModelView` tree without `ModelSQL` for this, or should I approach it differently?
Thanks for your help!
---
### Author:  [@pokoli](https://discuss.tryton.org/u/pokoli)
#### Post date: [July 9, 2026, 2:34pm UTC](https://discuss.tryton.org/t/display-api-information-without-storing-it/9282/2 "2026-07-09T14:34:16Z")
> [@German](#):
>
> Which components would best suit this requirement? Is it possible to use a `ModelView` tree without `ModelSQL` for this, or should I approach it differently?
Yes, for that reason we have ModelStorage to define a generic storage.
You can create any subclass that extends ModelStorage and implements the needed operations.
> [@German](#):
>
> My goal is to display this information inside Tryton without persisting any data to the database. Ideally, the workflow would look like this:
>
> - The user types a brand (e.g., “BMW”).
>
> - Tryton displays a tree view with the corresponding rows (e.g., X1, M3, Z4).
>
> - When the user selects a row, Tryton fetches and displays the detailed view for that specific model.
With such workflow in mind I will consider having a wizard that guides the users througth each step of your workflow. Proably it will be simpler that implementing a custom storage not based on ModelSQL.
---
### Author:  [@ced](https://discuss.tryton.org/u/ced)
#### Post date: [July 9, 2026, 2:44pm UTC](https://discuss.tryton.org/t/display-api-information-without-storing-it/9282/3 "2026-07-09T14:44:48Z")
Another option would be to use a [foreign data](https://www.postgresql.org/docs/current/ddl-foreign-data.html).
---
### Author:  [@German](https://discuss.tryton.org/u/German)
#### Post date: [July 9, 2026, 2:48pm UTC](https://discuss.tryton.org/t/display-api-information-without-storing-it/9282/4 "2026-07-09T14:48:49Z")
Thanks both for your answers, I’ll take into account each one, so inside a wizard I can create kind of a tree view to list tabular data? I was thinking like a Function field as well for these purpose.
---
### Author:  [@ced](https://discuss.tryton.org/u/ced)
#### Post date: [July 9, 2026, 2:55pm UTC](https://discuss.tryton.org/t/display-api-information-without-storing-it/9282/5 "2026-07-09T14:55:00Z")
> [@German](#):
>
> so inside a wizard I can create kind of a tree view to list tabular data?
No they must implement all the `ModelStorage` API.
---
### Author:  [@pokoli](https://discuss.tryton.org/u/pokoli)
#### Post date: [July 10, 2026, 9:01am UTC](https://discuss.tryton.org/t/display-api-information-without-storing-it/9282/6 "2026-07-10T09:01:55Z")
> [@German](#):
>
> so inside a wizard I can create kind of a tree view to list tabular data?
Yes, you have an example in the [Open Bom Tree wizard of the production module](https://foss.heptapod.net/tryton/tryton/-/blob/955b08fb104b0716e2f306f0b4271e3dc3e30a74/modules/production/bom.py#L453)