woodpecker/docs/docs/10-intro.md

88 lines
2.1 KiB
Markdown
Raw Normal View History

2019-11-13 14:47:14 +00:00
# Welcome to Woodpecker
2019-11-13 11:26:38 +00:00
Woodpecker is a simple CI engine with great extensibility. It runs your pipelines inside [containers](https://opencontainers.org/), so if you are already using them in your daily workflow, you'll love Woodpecker for sure.
2019-11-13 11:26:38 +00:00
![woodpecker](woodpecker.png)
2019-11-15 10:13:16 +00:00
## .woodpecker.yml
2019-11-13 11:26:38 +00:00
- Place your pipeline in a file named `.woodpecker.yml` in your repository
2019-11-13 14:47:14 +00:00
- Pipeline steps can be named as you like
- Run any command in the commands section
2019-11-13 11:26:38 +00:00
2019-11-13 14:47:14 +00:00
```yaml
# .woodpecker.yml
2019-11-13 14:47:14 +00:00
pipeline:
build:
image: debian
commands:
- echo "This is the build step"
a-test-step:
image: debian
commands:
- echo "Testing.."
```
### Pipeline steps are containers
2019-11-13 14:47:14 +00:00
- Define any container image as context
- either use your own and install the needed tools in custom image or
- search for available images that are already tailored for your needs on container registries like [Docker Hub](https://hub.docker.com/search?type=image)
- List the commands that should be executed in your container, in order to build or test your application
2019-11-13 14:47:14 +00:00
```diff
pipeline:
build:
- image: debian
2019-11-13 14:47:14 +00:00
+ image: mycompany/image-with-awscli
commands:
- aws help
```
2019-11-13 18:50:54 +00:00
### File changes are incremental
2019-11-13 14:47:14 +00:00
- Woodpecker clones the source code in the beginning pipeline
- Changes to files are persisted through steps as the same volume is mounted to all steps
```yaml
# .woodpecker.yml
2019-11-13 14:47:14 +00:00
pipeline:
build:
image: debian
commands:
- touch myfile
a-test-step:
image: debian
commands:
- cat myfile
```
## Plugins are straightforward
2019-11-13 14:47:14 +00:00
- If you copy the same shell script from project to project
- Pack it into a plugin instead
- And make the yaml declarative
- Plugins are Docker images with your script as an entrypoint
```Dockerfile
# Dockerfile
FROM laszlocloud/kubectl
COPY deploy /usr/local/deploy
ENTRYPOINT ["/usr/local/deploy"]
```
```bash
# deploy
kubectl apply -f $PLUGIN_TEMPLATE
```
```yaml
# .woodpecker.yml
2019-11-13 14:47:14 +00:00
pipeline:
deploy-to-k8s:
image: laszlocloud/my-k8s-plugin
template: config/k8s/service.yml
```
2019-11-13 11:26:38 +00:00
2022-08-31 23:52:52 +00:00
See [plugin docs](./20-usage/51-plugins/10-plugins.md).