# Docker instructions
Navigate to the 'docker' folder that contains the Dockerfile and run the following command to build the image with palisade-development repository latest commit:

**Building docker image:**

```docker build -t <imagename> . ```

example:
```docker build -t palisade-development . ```

no_threads is an argument (with a default value of 1 if it is not passed as an argument) for running make with the -j flag while installing palisade in the docker build. A different value for the number of threads can be passed as an argument to the docker build with the following command:

```docker build -t palisade-development . --build-arg no_threads=4```

CC_param and CXX_param are used to set the compiler for C++, the default compiler is g++-10 if the arguments are not set.

example:
```docker build -t palisade-development . --build-arg no_threads=1 --build-arg CC_param=/usr/bin/clang-10 --build-arg=/usr/bin/clang++-10```

The palisade repository and the release tag are also arguments that can be passed to the docker build to build the corresponding image (the default repository is palisade-development, master branch and the latest commit).

For example, to build palisade-release v1.10.6, run the following command:
```docker build -t palisade-release . --build-arg repository=palisade-release --build-arg tag=tags/v1.10.6```

The docker build also works for compiling specific commits of a specific branch of a repository:
```docker build -t palisade-release . --build-arg repository=palisade-release --build-arg branch=<branch_name> --build-arg tag=<commit_hash>```

Note: To checkout a specific release/tag, pass the argument as "tags/<tag_id>" and not just the <tag_id>.


Summary of the arguments to docker build:

- repository: refers to the repository to be built.
- branch: specific branch of the repository.
- tag: either the release tag (example tags/v1.10.6-dev) or specific commit hash.
- CC_param: only passed as argument if the compiler is clang-10.
- CXX_param: only passed as argument if the compiler is clang++-10.
- no_threads: argument to run make with as many threads.

**Running docker container with the docker image:**

Once the image is built, run the container by using:

```docker run -it --name palisade <imagename>```

example:
```docker run -it --name palisade palisade-development```

- the -it flag is used to run the container in interactive mode with a bash shell
- --name flag creates the docker container with name as palisade.
- -p can be used to expose the port of the container to the host machine, example: ```docker run -it -p 80:80 --name palisade palisade-development```

**Running additional examples with docker palisade installation:**

The palisade installation in the docker container can be used for running palisade examples by mounting the examples repository into the container with the -v flag. The following command runs a docker container and mounts the palisade-serial-examples folder into the docker container:

```docker run -it -v ${PWD}/palisade-serial-examples:/palisade-serial-examples palisade```

Multiple terminals of the same container (that may be needed for client/server examples) can be run using

```docker exec -it palisade /bin/bash```

**Clean up:**

Type 'exit' to leave the container shell.

To clean all containers once palisade-development is done running

```docker kill palisade```

```docker container prune```

To remove all the docker images and containers in the system, run ```docker system prune```

**Instructions for registry**:

To build and push docker build to the registry, run the upload-tagged-docker.sh as
```./upload-tagged-docker.sh -r <repository> -j <no_threads> -t <tag_id>```

Follow instructions in [Container_Registry.md](https://gitlab.com/palisade/palisade-development/-/blob/master/docker/Container_Registry.md) to run docker container with an image from the registry.