xflights
A reuse package serving master-data entities (Airlines, Airports and Flights) as a service that other CAP applications compose into their own data model.
Two service interfaces split maintenance from consumption: an admin service to maintain the data from a UI or a remote system, and a data service that denormalises Flights together with its FlightConnections for consuming apps. cds export turns the data service into a standalone client package, @capire/xflights-data, published to GitHub Packages so other applications, such as xtravels, import and mix its entities into their own schema instead of redefining them. An npm workspace setup lets the same package be linked locally during development, ahead of any publish.
Facts
Where it sits
Who is behind it
In the atlas: maintains 20, wrote 1@capiremaintainerJohannes VogtContributes to this
In the atlas: maintains 1, contributes to 28, wrote 10@AkatuoroBob den OsContributes to this
In the atlas: contributes to 13@BobdenOsChristian GeorgiContributes to this
In the atlas: maintains 1, contributes to 53@chgeoDaniel HutzelContributes to this
In the atlas: contributes to 32, wrote 3@danjoaDJ AdamsContributes to this
Developer Advocate
In the atlas: contributes to 20, wrote 22@qmacroNick JosipovicContributes to this
In the atlas: contributes to 4, wrote 1@nkaputnikRené JeglinskyContributes to this
In the atlas: contributes to 20@renejeglinskysjvansContributes to this
In the atlas: contributes to 45@sjvansSteffen WeinstockContributes to this
In the atlas: contributes to 14@stewskSteffen WaldmannContributes to this
In the atlas: contributes to 25@swaldmann10 contributors
What it declares
Every artifact
- Type
- Sample
- Origin
- By SAP
- Use as
- A reference
As a sample
- Form
- A reference application
Connections
Defining entities, their relations and the views services consume
Using a remote service as if it were one of your own
Download
Learn more
- CAP-level Data FederationCAP-level Data FederationA Devtoberfest 2026 session (CAP-31) on letting one CAP application read another's data as if it were local. The modelling half is the same for both answers, which is why the two technical edges hold: the provider defines an API service and exports it with CDS export as an npm package carrying the model, mock data and translated texts, the consumer imports it and puts consumption views on top, and only then is it wired either as service-level replication that copies the rows into local tables each time the server starts, or over the SAP HANA synonyms that cds-federation-synonyms generates, with HDI grants on the provider side and a service replacement that redirects each synonym from its mock table to the provider's container. It is worked through on xflights and xtravels, including the local run where CDS watch mocks the imported service from the CSV files inside the package.SAP Developers, published 2026-10-02SAP Developers, YouTube — SAP Developers playlists
README
quoted from capire/xflights under Apache-2.0, read@capire/xflights
This is a reuse package to manage and serve master data like Airlines, Airports, and Flights. It publishes a pre-built client package, that is used in the xtravels application.
Table of Contents
Domain Model
The domain model is defined in db/schema.cds. It centers around normalized FlightConnections, which connect two Airports operated by an Airline, while entity Flights represents scheduled flights on specific dates with a specific aircraft and price.
Service Interfaces
Two service interfaces are defined in srv/admin-service.cds, and srv/data-service.cds, to serve different use cases as shown below:
- an admin service to maintain the master data from UIs or remote systems
- a data service to consume it from remote applications
[!tip]
Serving denormalized views
The data service exposes a denormalized view of `Flights` and associated `FlightConnections` data, essentially declared like that:entity Flights as projection on my.Flights { *, // all elements from Flights flight.*, // all elements from FlightConnections }With that consumers aren't bothered with normalized data but can just consume flat data, looking like that:
Exporting APIs
Given the respective service definition, we create a pre-built client package for the data API, which can be used from consuming apps in a plug-and-play fashion.
We use cds export to create the API package, based on the Data Service definition:
cds export srv/data-service.cds
This generates a separate CAP reuse package within subfolder apis/data-service that contains only the effective service API definitions, accompanied by automatically derived test data and i18n bundles.
Initially, cds export also adds a package.json, which we can modify as appropriate, and did so by changing the package name to @capire/xflights-data:
{
- "name": "@capire/xflights-data-service",
+ "name": "@capire/xflights-data",
...
}
Publishing APIs
We can finally share this package with consuming applications using standard ways, like npm publish:
cd apis/data-service
npm publish
[!tip]
Using GitHub Packages
Within the capire org, we're publishing to GitHub Packages, which requires you to npm login once like that:
npm login --scope=@capire --registry=https://npm.pkg.github.comAs password you're using a Personal Access Token (classic) with
read:packagesscope (for retrieving and installing a package). Read more about that in Authenticating to GitHub Packages.
Consuming APIs
Use the published client package in your consuming application by installing it via npm:
npm add @capire/xflights-data
With that, we can use the imported models as usual, and as if they were local in mashups with our own entities like so:
using { sap.capire.flights.FlightsService as imported } from '@capire/xflights-data';
entity TravelBookings { //...
flight : Association to imported.Flights;
}
▷ Learn more about consuming APIs and CAP-level data integration in the xtravels application.
Using Workspaces
Instead of exercising a workflow like that again and again:
- ( develop → export → publish ) → npmjs.com → ( update → consume )
... we can use npm workspaces technique to work locally and speed up things as follows:
mkdir -p cap/works; cd cap/works
git clone https://github.com/capire/xflights
git clone https://github.com/capire/xtravels
echo '{"workspaces":["xflights","xtravels"]}' > package.json
Add a link to the local @capire/xflights-data API package, enclosed with the cloned xflights sources:
npm add ./xflights/apis/data-service
Check the installation using npm ls, which would yield output as below, showing that @capire/xtravel's dependency to @capire/xflights-data is nicely fulfilled by a local link to ./xflights/apis/data-service:
npm ls @capire/xflights-data
works@ ~/cap/works
├── @capire/xflights-data@0.1.11 -> ./xflights/apis/data-service
└─┬ @capire/xtravels@1.0.0 -> ./xtravels
└── @capire/xflights-data@0.1.11 deduped -> ./xflights/apis/data-service
Start the xtravels application → and note the sources loaded from ./xflights/apis/data-service, and the information further below about the FlightsService service mocked automatically:
cds watch xtravels
[cds] - loaded model from 20 file(s):
xtravels/srv/travel-service.cds
xtravels/db/schema.cds
xtravels/db/xflights.cds
xflights/apis/data-service/index.cds
xflights/apis/data-service/services.csn
...
[cds] - mocking sap.capire.flights.FlightsService {
at: [ '/odata/v4/data', '/rest/data', '/hcql/data' ],
decl: 'xflights/apis/data-service/services.csn:3',
}
Using Proxy Packages
The usage of npm workspaces technique as described above streamlined our workflows as follows:
- Before: ( develop → export → publish ) → npmjs.com → ( update → consume )
- After: ( develop → export ) → ( consume )
We can even more streamline that by eliminating the export step as follows...
Create a new subfolder xflights-api-shortcut in which we add two files as follows:
mkdir xflights-api-shortcut
Add a package.json file in there with that content:
{
"name": "@capire/xflights-data",
"dependencies": {
"@capire/xflights": "*"
}
}
And an index.cds file with that content:
using from '@capire/xflights/srv/data-service';
Using the shell's "here document" technique
You can also create those two files from the command line as follows:
cat > xflights-api-shortcut/package.json << EOF
{
"name": "@capire/xflights-data",
"dependencies": {
"@capire/xflights": "*"
}
}
EOF
Take the same approach for the index.cds file:
cat > xflights-api-shortcut/index.cds << EOF
using from '@capire/xflights/srv/data-service';
EOF
npm add ./xflights-api-shortcut
Check the effect of that → note how @capire/xflights-data dependencies now link to ./xflights-api-shortcut:
npm ls @capire/xflights-data
works@ ~/cap/works
├── @capire/xflights-data@ -> ./xflights-api-shortcut
└─┬ @capire/xtravels@1.0.0 -> ./xtravels
└── @capire/xflights-data@ deduped -> ./xflights-api-shortcut≤
Start the xtravels application → and note the sources loaded from ./xflights-api-shortcut, and the information further below about the FlightsService service now being served, not mocked anymore:
cds watch xtravels
[cds] - loaded model from 20 file(s):
xtravels/srv/travel-service.cds
xtravels/db/schema.cds
xtravels/db/xflights.cds
xflights-api-shortcut/index.cds
xflights/srv/data-service.cds
xflights/db/schema.cds
...
[cds] - serving sap.capire.flights.FlightsService {
at: [ '/odata/v4/data', '/rest/data', '/hcql/data' ],
decl: 'xflights/apis/data-service/services.csn:3',
}
Which means we've streamlined our workflows as follows:
- Before: ( change → export → publish ) → npmjs.com → ( update → consume )
- Step 1: ( change → export ) → ( consume )
- Step 2: ( change ) → ( consume )
License
Copyright (c) 2026 SAP SE or an SAP affiliate company. All rights reserved. This file is licensed under the Apache Software License, version 2.0 except as noted otherwise in the LICENSE file.
Details
Sources
Sources are graph entities: the same page may back several entries, and an unreachable source flags every row that rests on it.
Compatibility and provenance
Contributors
History in the atlas
- First seen.
For your agent
- Ref
artifact:capire-xflights- JSON
- /artifacts/capire-xflights.json
- MCP
- Connecting an agent