Skip to content

anaconda-graveyard/anaconda-build

Repository files navigation

Anaconda Build Client

This is a command line client that provides an interface to the Anaconda Cloud build system.

Quickstart:

Create an account:

First create an account on Anaconda Cloud

Install the build client:

With Conda:

    $ conda install anaconda-build

Login

$ anaconda login

Test your login with the whoami command:

$ anaconda whoami

For a complete tutorial on building and uploading Conda packages to Anaconda Cloud read the Getting Started Guide.

For detailed information on registering Workers with Anaconda Build Queues read the Build Reference.

Installing from source

run

git clone https://github.com/Anaconda-Server/anaconda-build
cd anaconda-build
python setup.py [develop|install]

Here is an example of informally testing anaconda-build that can be run after installing prerequisite packages, such as Visual Studio, Apple Developer Tools, and git.

Make sure build-related Python packages are up-to-date

conda update -n root conda-build conda
conda create -n build_env python=2.7 anaconda-build chalmers anaconda-client ndg-httpsclient jinja2 mock
source activate build_env

Login and come up with a queue name (user/queue)

# login to anaconda
anaconda login

# Make an env variable for your anaconda.org username, for me that is
export anaconda_user=psteinberg
export queue=${anaconda_user}/abc

create a build queue if you don't have one in your account yet

anaconda build queue --create ${queue}

Register a worker, adjusting the worker name, distribution and platform as necessary

export WORKER_ID=worker_1
anaconda worker register ${queue} --dist darwin --platform osx-64 --name ${WORKER_ID}

Run a worker list command to check registered workers

anaconda worker list

Create an anaconda.org auth token so worker can be run while you are not logged in

anaconda auth --create -n "anaconda-build-OSX-64" --scopes "api:build-worker" --out ~/.anaconda.token

Logout

anaconda logout

Add a chalmers managed program for the worker

chalmers add --name worker -c "anaconda --show-traceback -t ${HOME}/.anaconda.token worker run ${WORKER_ID} --status-file" ${HOME}/.worker.status

Start chalmers

# Start chalmers with streaming output (--all can be replaced by "worker" in this case)

chalmers start --all --stream -w

Leave that running, open a separate tab and submit a build to your queue

# First make an env var for your queue, for me that is
export queue=psteinberg/abc # username/queuename
# Submit a local package directory with a .binstar.yml,
# using the `--platform` filter to only include the build/test entries
# relevant to this platform.

anaconda build submit ./path/to/package --queue ${queue} --platform osx-64

# Then watch the output for the builds you can tail, with something like
anaconda build tail -f username/packagename 1.0 

# Alternatively trigger a build for a github repo
# that has CI enabled.  Find CI options for your package at a url
# like: https://anaconda.org/psteinberg/astropy/settings/ci ,
# replacing psteinberg and astropy with your username and package name.
# Triggering uses your anaconda.org username/packagename with arguments,
# again limiting the builds from the .binstar.yml in the github repo
# to those on this platform
anaconda build trigger psteinberg/myorg.package1 --queue ${queue} --platform osx-64

# optionally tail the results or view them at anaconda.org