Just wanted to save this explanation. Not sure of where in the tutorial to insert it:
The fields in GrimoireLab come to be through different paths:
- Most are just retrieved from the data sources. We call this data raw data.
- The enrichment process that runs after the data load (usually called by Morderd for similar) adds more standard-calculated fields. We call this data enriched data.
- Then, the user can inject fields via configuration (instead of retrieving them directly or indirectly from the sources). We can call these configuration-injected fields "tags".
Currently, all tags are defined via the project.json file).
Some tags have standard field names. The user only injects their arbitrary values.
Some other tags are user-defined fields. The user injects both their values and the name of these fields.
Grimoirelab can inject user-defined data through at least 3 paths:
- The
project field. It holds the name of the projects. Their values (project names) and their applicabilities (which data source belongs to each project) are defined in the projects.json file.
- The
meta member of each project (set in the projects.json file).
- It generates the
cm_* fields we find in every item retrieved from sources (commits, issues, PRs, repos, areas_of_code, ...) attributed to that project.
- All these project-level custom fields are marked with a
cm_ appendix. So, if you ask for a my_tag field in the meta member of a project you get a cm_my_tag field.
- A standard field called
labels.
- It is a list of values.
- We can inject a user-defined list of values per repository.
- This, again, propagates to many items retrieved from sources (commits, issues, PRs, repos, areas_of_code, ...) attributed to that repository.
Just wanted to save this explanation. Not sure of where in the tutorial to insert it: