Skip to main content

container database (cdb) metadata generation tool.

Project description

Container Database (cdb)

This is the Python support tool for containerdb to support generation of data containers. Python is more friendly to generating arbitrary data structures, and is popular among the data science community, so I chose it for metadata generation instead of using GoLang.

PyPI version

Have your data and use it too!


For documentation and full examples see These examples are also available in the examples folder.

Getting Started

What is a Data Container?

A data container is generally an operating-system-less container that is optimized to provide data, either for query/search, or binding for analysis. The qualities of the data container should be:

  1. It can be mounted to containers with operating systems to run analysis
  2. It can be interacted with on it's own to search metadata about the data
  3. It should not have an operating system.

How do we generate one?

The generation is fairly simple! It comes down to a three step multistage build:

  1. Step 1 We install cdb to generate a GoLang template for an in-memory database for our data)
  2. Step 2 We compile the binary into an entrypoint
  3. Step 3 We add the data and the binary entrypoint to a scratch container (no operating system).

And then we interact with it! This tutorial will show you the basic steps to perform the multistage-build using a simple Dockerfile along with the data folder. The Dockerfile in the base of the repository also is a good example.


Docker Usage

The intended usage is via Docker, so you don't need to worry about installation of Python, GoLang, and multistage builds to basically:

  1. Generate a db.go template
  2. Compile it
  3. Add to scratch with data as data container entrypoint.

Thus, to run the dummy example here using the Dockerfile:

$ docker build -t data-container .

We then have a simple way to do the following:


If we just run the container, we get a listing of all metadata alongside the key.

$ docker run entrypoint 
/data/avocado.txt {"size": 9, "sha256": "327bf8231c9572ecdfdc53473319699e7b8e6a98adf0f383ff6be5b46094aba4"}
/data/tomato.txt {"size": 8, "sha256": "3b7721618a86990a3a90f9fa5744d15812954fba6bb21ebf5b5b66ad78cf5816"}


We can also just list data files with -ls

$ docker run entrypoint -ls


Or we can list ordered by one of the metadata items:

$ docker run entrypoint -metric size
Order by size
/data/tomato.txt: {"size": 8, "sha256": "3b7721618a86990a3a90f9fa5744d15812954fba6bb21ebf5b5b66ad78cf5816"}
/data/avocado.txt: {"size": 9, "sha256": "327bf8231c9572ecdfdc53473319699e7b8e6a98adf0f383ff6be5b46094aba4"}


Or search for a specific metric based on value.

$ docker run entrypoint -metric size -search 8
/data/tomato.txt 8

$ docker run entrypoint -metric sha256 -search 8
/data/avocado.txt 327bf8231c9572ecdfdc53473319699e7b8e6a98adf0f383ff6be5b46094aba4
/data/tomato.txt 3b7721618a86990a3a90f9fa5744d15812954fba6bb21ebf5b5b66ad78cf5816


Or we can get a particular file metadata by it's name:

$ docker run entrypoint -get /data/avocado.txt
/data/avocado.txt {"size": 9, "sha256": "327bf8231c9572ecdfdc53473319699e7b8e6a98adf0f383ff6be5b46094aba4"}

or a partial match:

$ docker run entrypoint -get /data/
/data/avocado.txt {"size": 9, "sha256": "327bf8231c9572ecdfdc53473319699e7b8e6a98adf0f383ff6be5b46094aba4"}
/data/tomato.txt {"size": 8, "sha256": "3b7721618a86990a3a90f9fa5744d15812954fba6bb21ebf5b5b66ad78cf5816"}


The start command is intended to keep the container running, if we are using it with an orchestrator.

$ docker run data-container -start


It's more likely that you'll want to interact with files in the container via some analysis, or more generally, another container. Let's put together a quick docker-compose.yml to do exactly that.

version: "3"
    restart: always
    image: busybox
    entrypoint: ["tail", "-f", "/dev/null"]
      - data-volume:/data

    restart: always
    image: data-container
    command: ["-start"]
      - data-volume:/data


Notice that the command for the data-container to start is -start, which is important to keep it running. After building our data-container, we can then bring these containers up:

$ docker-compose up -d
Starting docker-simple_base_1   ... done
Recreating docker-simple_data_1 ... done
$ docker-compose ps
        Name                Command         State   Ports
docker-simple_base_1   tail -f /dev/null    Up           
docker-simple_data_1   /entrypoint -start   Up           

We can then shell inside and see our data!

$ docker exec -it docker-simple_base_1 sh
/ # ls /data/
avocado.txt  tomato.txt

The metadata is still available for query by interacting with the data-container entrypoint:

$ docker exec docker-simple_data_1 /entrypoint -ls

Depending on your use case, you could easily make this available inside the other container. This is very simple usage, but the idea is powerful! We can interact with the dataset and search it without needing an operating system. It follows that we can develop customized data-containers based on the format / organization of the data inside (e.g., a data-container that knows how to expose inputs, outputs, etc.)

Python Usage

The above doesn't require you to install the Container Database (cdb) metadata generator, however if you want to (to develop or otherwise interact) you can do the following. First, install cdb from pypi or a local repository:

$ pip install cdb


git clone
cd cdb
pip install -e .

Command Line

The next step is to generate the goLang file to compile. You'll next want to change directory to somewhere you have a dataset folder. For example, in tests we have a dummy "data" folder.

cd tests/

We might then run cdb generate to create a binary for our container, targeting the tests/data folder:

$ cdb generate data --out db.go

The db.go file is then in the present working directory. You can either build it during a multistage build as is done in the Dockerfile, or do it locally with your own GoLang install and then add to the container. For example, to compile:

go get && \
GOOS=linux GOARCH=amd64 go build -ldflags="-w -s" -o /db -i /db.go

And then a very basic Dockerfile would need to add the data at the path specified, and the compiled entrypoint.

FROM scratch
COPY data/ .
COPY db /db
CMD ["/db"]

A more useful entrypoint will be developed soon! This is just a very basic start to the library.


You can run the same generation functions interactively with Python.

from cdb.main import ContainerDatabase
db = ContainerDatabase(path="data")
# <cdb.main.ContainerDatabase at 0x7fcaa9cb8950>

View that there is a files generator at db.files

<generator object recursive_find at 0x7fcaaa4ae950>

And then generate! If you don't provide an output file, a string will be returned. Otherwise, the output file name is returned.

output = db.generate(output="db.go", force=True)

Currently, functions for parsing metadata are named in cdb/, however you can also define a custom import path. This has not yet been tested and will be soon. We will also be added more real world examples soon.


  • Free software: MPL 2.0 License

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

cdb-0.0.1.tar.gz (23.6 kB view hashes)

Uploaded source

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Huawei Huawei PSF Sponsor Microsoft Microsoft PSF Sponsor NVIDIA NVIDIA PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page