Skip to main content

CLI Application - administration documentation

Here you will find the information needed for running the CLI version of Validator.

You have two options:

Don't forget to check out the user documentation of CLI App!

Docker​

To use this app you need to have docker installed on your local machine and have the docker daemon running!

  1. Clone the repository.
  2. Relocate to the directory CSV_Validator.
  3. Build the docker image. Run this command (you can replace the validate-image with your desired image name):
docker build -t validate-image -f ValidateCLI/Dockerfile .

Now you can run the application using command:

docker run validate-image

This will run the application. However the application will expect some command line arguments to follow the command.

Example usage (docker)​

We expect that you have completed steps 1, 2 and 3 from the docker part. Please be aware of the name of your docker container.

Remote (online) files​

You can run the image you have created in the step 3. like this:

docker run validate-image --metadataIRI https://w3c.github.io/csvw/tests/test035/csv-metadata.json

This will start remote validation on a file that is part of the integration tests for the validator. You can replace the URL with arbitrary URL of metadata file you want to validate. In this case the metadata starts with metadata file so no other metadata is located, this type of validation is called Metadata validation

If you want to start tabular validation you can do so by invoking:

docker run validate-image --tabularIRI https://w3c.github.io/csvw/tests/test006.csv

If you provide both parameters:

docker run validate-image --tabularIRI https://w3c.github.io/csvw/tests/test012/tree-ops.csv --metadataIRI https://w3c.github.io/csvw/tests/test012/csv-metadata.json

The validation will be conducted as overriding.

Local files​

When you want to validate local files you need to somehow map your local directory into the container. Lets look at an example:

docker run -v /path/to/your/directory:/data validate-image --tabularIRI /data/my.csv

So if you have your local files in /path/to/your/directory it will map them into the directory /data in the docker container. This also means that if you used absolute paths inside metadata file, the validation won't succeed as the paths will be invalid.

Another example is that for example you want to start validation from metadata file lets say it's path is:

/opt/myFiles/metadata.json

and this file references tabular data file with path:

/opt/myFiles/tabular.csv

for this we will use command:

docker run -v /opt/myFiles:/data validate-image --metadataIRI /data/metadata.json

If you provide both paths:

docker run -v /opt/myFiles:/data validate-image --metadataIRI /data/metadata.json --tabularIRI /data/tabular.csv

the validation will be done as overriding. That means that the reference from the metadata file metadata.json to another file will be overriden to the path you provided, in this case /data/tabular.csv. This is useful if you want to use the same schema for different tabular files.

Please checkout the arguments documentation for the command line for more info about the usage.

Mapping of local files into the Docker container​

If you want to use local files in the docker container you firstly need to map directory containing such files into the Docker container. This is done using the parameter -v:

-v /my/local/path:/my/docker/container/path

this will map the directory /my/local/path of your local machine into the directory /my/docker/container/path in the docker container. Since this mapping we will be using only the mapped paths as the Validator running inside the Docker container cannot access your local files.

Now if we want to start for example tabular validation we can do so by invoking:

docker run -v /my/local/path:/my/docker/container/path validate-image --tabularIRI /my/docker/container/path/my.csv --rdfPaths /my/docker/container/path/result.ttl

In this example we have a local file /my/local/path/my.csv and the validator sees it as /my/docker/container/path/my.csv. Also notice that if we want to create for example a RDF result we also need to use mapped directories not the local ones!

We recommend to map your directories into the directory /data in the docker container.

Build with dotnet​

This has been tested on .NET 7.0 and .NET 8.0.

Note that if you use relative paths with the local files, they will be relative to the location of the ValidateCLI.dll.

If you prefer to build the CLI app on your own from the source files, we have step-by-step guide for you:

  1. Clone the repository.
  2. Relocate to the directory ValidateCLI.
  3. To restore the project run:
dotnet restore
  1. To start the validation run:
dotnet run --configuration Release [CLI arguments]

Examples​

Start Tabular validation:

dotnet run --tabularIRI https://w3c.github.io/csvw/tests/test005.csv

Start Metadata validation:

dotnet run --metadataIRI https://w3c.github.io/csvw/tests/test034/csv-metadata.json

Start Overriding validation:

dotnet run --metadataIRI https://w3c.github.io/csvw/tests/test011/tree-ops.csv-metadata.json --tabularIRI https://w3c.github.io/csvw/tests/test011/tree-ops.csv

You can replace the URLs with your local file paths to validate local files!

Please checkout the arguments documentation for the command line for more info about the usage.

Alternative​

Alternatively you can follow these steps:

  1. Clone the repository.
  2. Relocate to the directory ValidateCLI.
  3. To restore the project run:
dotnet restore
  1. Build the application:
dotnet build --configuration Release --output /output/directory
  1. Run the application
dotnet /output/directory/ValidateCLI.dll [CLI arguments]