I'm currently not actively using Couchbase nor this generator for any project.
If you are starting a new project from scratch, check the alternatives at the FastAPI docs: Project Generation.
For example, the project generator Full Stack FastAPI PostgreSQL might be a better alternative, as it is actively maintained and used. And it includes all the new features and improvements.
You are still free to use this project if you want to, it should probably still work fine, and if you already have a project generated with it that's fine as well (and you probably already updated it to suit your needs).
But maintaining it in sync with Full Stack FastAPI PostgreSQL including all the new features and development done there is currently not being feasible. So, I won't be able to update it much.
Here are some extra thoughts you might want to consider if you decide to go with Couchbase and/or this project generator.
Couchbase has a great set of features that is not easily or commonly found in alternatives.
It's a distributed database, so, if you have a cluster of several nodes with the data replicated, you don't have a single point of failure.
It provides very high performance.
It has built-in full-text search integrated (using Bleve). It's possible to make all the records be automatically indexed in the search engine without additional code or extra components.
It can be connected/integrated with the Couchbase Sync Gateway, that can be used to synchronize a subset of documents (records) with mobile devices. And with some effort, it can be connected with PouchBD for frontend synchronization (although not querying and/or searching from the frontend, as would be possible with CouchDB).
The documentation is very scarce. This is sadly a big problem. Many features are not well documented, or not documented at all.
Many configurations are not documented and have to be inferred from comments in forums, reading source code, or checking the configurations for the C client and trying different query parameters in URLs, that becomes quite error prone.
Couchbase didn't seem to be designed to be integrated into CI systems as in this project (or alternatives), at least until the last time I used it. It expects all the set up to be done once via the web UI, not automated. The official Docker image can't be configured. So, this project does all the configuration by sending the HTTP requests from the backend code to the container replicating the HTTP requests done in the web UI. But those steps are not documented, there's no "official" way to configure and start it without using the web UI, so the integration with CI could be error prone.
The Couchbase Sync Gateway official Docker image is not designed to be configured much either, so, the Dockerfile
included in this project adds a good amount of custom logic to support that, but that's not really official.
As it uses N1QL, a flavor of SQL, and there's no easy integration with Python, you have to do all the operations in N1QL strings and integrate them with your own code. Without editor support, completion, nor syntax checks for N1QL.
Generate a backend and frontend stack using Python, including interactive API documentation.
- Full Docker integration (Docker based).
- Docker Swarm Mode deployment.
- Docker Compose integration and optimization for local development.
- Production ready Python web server using Uvicorn and Gunicorn.
- Python FastAPI backend:
- Fast: Very high performance, on par with NodeJS and Go (thanks to Starlette and Pydantic).
- Intuitive: Great editor support. Completion everywhere. Less time debugging.
- Easy: Designed to be easy to use and learn. Less time reading docs.
- Short: Minimize code duplication. Multiple features from each parameter declaration.
- Robust: Get production-ready code. With automatic interactive documentation.
- Standards-based: Based on (and fully compatible with) the open standards for APIs: OpenAPI and JSON Schema.
- Many other features including automatic validation, serialization, interactive documentation, authentication with OAuth2 JWT tokens, etc.
- Secure password hashing by default.
- JWT token authentication.
- CORS (Cross Origin Resource Sharing).
- Celery worker that can import and use code from the rest of the backend selectively (you don't have to install the complete app in each worker).
- NoSQL Couchbase database that supports direct synchronization via Couchbase Sync Gateway for offline-first applications.
- Full Text Search integrated, using Couchbase.
- REST backend tests based on Pytest, integrated with Docker, so you can test the full API interaction, independent on the database. As it runs in Docker, it can build a new data store from scratch each time (so you can use ElasticSearch, MongoDB, or whatever you want, and just test that the API works).
- Easy Python integration with Jupyter Kernels for remote or in-Docker development with extensions like Atom Hydrogen or Visual Studio Code Jupyter.
- Email notifications for account creation and password recovery, compatible with:
- Mailgun
- SparkPost
- SendGrid
- ...any other provider that can generate standard SMTP credentials.
- Vue frontend:
- Generated with Vue CLI.
- JWT Authentication handling.
- Login view.
- After login, main dashboard view.
- Main dashboard with user creation and edition.
- Self user edition.
- Vuex.
- Vue-router.
- Vuetify for beautiful material design components.
- TypeScript.
- Docker server based on Nginx (configured to play nicely with Vue-router).
- Docker multi-stage building, so you don't need to save or commit compiled code.
- Frontend tests ran at build time (can be disabled too).
- Made as modular as possible, so it works out of the box, but you can re-generate with Vue CLI or create it as you need, and re-use what you want.
- Flower for Celery jobs monitoring.
- Load balancing between frontend and backend with Traefik, so you can have both under the same domain, separated by path, but served by different containers.
- Traefik integration, including Let's Encrypt HTTPS certificates automatic generation.
- GitLab CI (continuous integration), including frontend and backend testing.
Go to the directory where you want to create your project and run:
pip install cookiecutter
cookiecutter https://github.com/tiangolo/full-stack-fastapi-couchbase
You will be asked to provide passwords and secret keys for several components. Open another terminal and run:
openssl rand -hex 32
# Outputs something like: 99d3b1f01aa639e4a76f4fc281fc834747a543720ba4c8a8648ba755aef9be7f
Copy the contents and use that as password / secret key. And run that again to generate another secure key.
The generator (cookiecutter) will ask you for some data, you might want to have at hand before generating the project.
The input variables, with their default values (some auto generated) are:
-
project_name
: The name of the project -
project_slug
: The development friendly name of the project. By default, based on the project name -
domain_main
: The domain in where to deploy the project for production (from the branchproduction
), used by the load balancer, backend, etc. By default, based on the project slug. -
domain_staging
: The domain in where to deploy while staging (before production) (from the branchmaster
). By default, based on the main domain. -
docker_swarm_stack_name_main
: The name of the stack while deploying to Docker in Swarm mode for production. By default, based on the domain. -
docker_swarm_stack_name_staging
: The name of the stack while deploying to Docker in Swarm mode for staging. By default, based on the domain. -
secret_key
: Backend server secret key. Use the method above to generate it. -
first_superuser
: The first superuser generated, with it you will be able to create more users, etc. By default, based on the domain. -
first_superuser_password
: First superuser password. Use the method above to generate it. -
backend_cors_origins
: Origins (domains, more or less) that are enabled for CORS (Cross Origin Resource Sharing). This allows a frontend in one domain (e.g.https://dashboard.example.com
) to communicate with this backend, that could be living in another domain (e.g.https://api.example.com
). It can also be used to allow your local frontend (with a customhosts
domain mapping, as described in the project'sREADME.md
) that could be living inhttp://dev.example.com:8080
to communicate with the backend athttps://stag.example.com
. Notice thehttp
vshttps
and thedev.
prefix for local development vs the "staging"stag.
prefix. By default, it includes origins for production, staging and development, with ports commonly used during local development by several popular frontend frameworks (Vue with:8080
, React, Angular). -
smtp_port
: Port to use to send emails via SMTP. By default587
. -
smtp_host
: Host to use to send emails, it would be given by your email provider, like Mailgun, Sparkpost, etc. -
smtp_user
: The user to use in the SMTP connection. The value will be given by your email provider. -
smtp_password
: The password to be used in the SMTP connection. The value will be given by the email provider. -
smtp_emails_from_email
: The email account to use as the sender in the notification emails, it would be something likeinfo@your-custom-domain.com
. -
couchbase_user
: Couchbase main user to be used by the application (code). By defaultadmin
. -
couchbase_password
: Password of the main user, for the backend code. Generate it with the method above. -
couchbase_sync_gateway_cors
: List of CORS origins that the Sync Gateway should allow to talk to directly. Similar tobackend_cors_origins
. -
couchbase_sync_gateway_user
: User to be created for the Couchbase Sync Gateway. This is what allows synchronization using the CouchDB protocol, with Couchbase Lite in mobile apps and PouchDB in the web and hybrid mobile apps. -
couchbase_sync_gateway_password
: Couchbase Sync Gateway password. Generate it with the method above. -
traefik_constraint_tag
: The tag to be used by the internal Traefik load balancer (for example, to divide requests between backend and frontend) for production. Used to separate this stack from any other stack you might have. This should identify each stack in each environment (production, staging, etc). -
traefik_constraint_tag_staging
: The Traefik tag to be used while on staging. -
traefik_public_network
: This assumes you have another separate publicly facing Traefik at the server / cluster level. This is the network that main Traefik lives in. -
traefik_public_constraint_tag
: The tag that should be used by stack services that should communicate with the public. -
flower_auth
: Basic HTTP authentication for flower, in the formuser:password
. By default: "root:changethis
". -
sentry_dsn
: Key URL (DSN) of Sentry, for live error reporting. If you are not using it yet, you should, is open source. E.g.:https://1234abcd:5678ef@sentry.example.com/30
. -
docker_image_prefix
: Prefix to use for Docker image names. If you are using GitLab Docker registry it would be based on your code repository. E.g.:git.example.com/development-team/my-awesome-project/
. -
docker_image_backend
: Docker image name for the backend. By default, it will be based on your Docker image prefix, e.g.:git.example.com/development-team/my-awesome-project/backend
. And depending on your environment, a different tag will be appended (prod
,stag
,branch
). So, the final image names used will be like:git.example.com/development-team/my-awesome-project/backend:prod
. -
docker_image_celeryworker
: Docker image for the celery worker. By default, based on your Docker image prefix. -
docker_image_frontend
: Docker image for the frontend. By default, based on your Docker image prefix. -
docker_image_sync_gateway
: Docker image for the Sync Gateway. By default, based on your Docker image prefix.
This stack can be adjusted and used with several deployment options that are compatible with Docker Compose, but it is designed to be used in a cluster controlled with pure Docker in Swarm Mode with a Traefik main load balancer proxy handling automatic HTTPS certificates, using the ideas from DockerSwarm.rocks.
Please refer to DockerSwarm.rocks to see how to deploy such a cluster in 20 minutes.
After using this generator, your new project (the directory created) will contain an extensive README.md
with instructions for development, deployment, etc. You can pre-read the project README.md
template here too.
- Based on PostgreSQL: https://github.com/tiangolo/full-stack-fastapi-postgresql.
- Fix Windows line endings for shell scripts after project generation with Cookiecutter hooks. PR #28.
- Remove
package-lock.json
to let everyone lock their own versions (depending on OS, etc). - Simplify Traefik labels for services. PR #27.
- Fix Flower Docker configuration. PR #24.
- Update testing scripts and typo.
- Add normal user Pytest fixture. PR #23.
- Update Dockerfiles to use Couchbase from Debian image. PR #20 by @Gjacquenot.
- Use new Pydantic types. PR #21 by @tehtbl.
-
Fix security on resetting a password. Receive
password
andtoken
as body, not query. PR #16. -
Forward arguments from script to
pytest
inside container. -
Update Jupyter Lab installation and util script/environment variable for local development.
- PR #15:
- Update CRUD utils to use types better.
- Simplify Pydantic model names, from
UserInCreate
toUserCreate
, etc. - Upgrade packages.
- Add new generic "Items" models, crud utils, endpoints, and tests. To facilitate re-using them to create new functionality. As they are simple and generic (not like Users), it's easier to copy-paste and adapt them to each use case.
- Update endpoints/path operations to simplify code and use new utilities, prefix and tags in
include_router
. - Update testing utils.
- Update linting rules, relax vulture to reduce false positives.
- Add full text search for items.
- Update project README.md with tips about how to start with backend.
-
Fix frontend hijacking /docs in development. Using latest https://github.com/tiangolo/node-frontend with custom Nginx configs in frontend. PR #14.
-
Update generated README. Minor typos.
-
Update Couchbase installation, to include Couchbase command line tools.
-
Set
/start-reload.sh
as a command override for development by default.
- Simplify and update backend
Dockerfile
s. - Refactor and simplify backend code, improve naming, imports, modules and "namespaces".
- Improve and simplify Vuex integration with TypeScript accessors.
- Standardize frontend components layout, buttons order, etc.
- Add local development scripts (to develop this project generator itself).
- Add logs to startup modules to detect errors early.
- Improve FastAPI dependency utilities, to simplify and reduce code (to require a superuser).
- Fix/update logic to update users.
This project is licensed under the terms of the MIT license.