How to download artifacts using an URL in a gitlab job?

Question

In the gitlab documentation some URL's are described for the purpose of downloading artifacts from pipelines HERE. They seem to have forgotten to describe HOW to download artifacts given these URLs.

Can it be done in a simple way? Or do I have to install e.g. wget, create a token, define a token, use a token?

If someone could give an example that would be great.

Answer 1

The documentation referenced in the question is supposed to be an REST API call, using the Job Artifact API:

GET /projects/:id/jobs/artifacts/:ref_name/download?job=name

To use this in a script definition inside .gitlab-ci.yml, you can use either:

• The JOB-TOKEN header with the GitLab-provided CI_JOB_TOKEN variable.
  For example, the following job downloads the artifacts of the test job of the main branch.
  The command is wrapped in single quotes because it contains a colon (:):

artifact_download:
stage: test
script:
  - 'curl --location --output artifacts.zip --header "JOB-TOKEN: $CI_JOB_TOKEN" "https://gitlab.example.com/api/v4/projects/$CI_PROJECT_ID/jobs/artifacts/main/download?job=test"'

• Or the job_token attribute with the GitLab-provided CI_JOB_TOKEN variable.
  For example, the following job downloads the artifacts of the test job of the main branch:

artifact_download:
 stage: test
 script:
   - 'curl --location --output artifacts.zip >"https://gitlab.example.com/api/v4/projects/$CI_PROJECT_ID/jobs/artifacts/main/download?job=test&job_token=$CI_JOB_TOKEN"'

---

But the artifact: directive is meant to store data in the job workspace, for a new iteration of the job to get back, in the same folder.

No "download" involved, as illustrated in the article "GitLab CI: Cache and Artifacts explained by example" by Anton Yakutovich.
As such, no curl/wget/TOKEN should be needed to access an artifact stored by a previous job execution.