Configuration

Before you run the importer, you may want to create a simple configuration file. By default, the importer looks for this config file in

~/.local/etc/searcch-importer.ini

You’ll want a config file that looks like this:

[DEFAULT]
user_email = johnsond@flux.utah.edu
user_name = johnsond

[github]
token = 1a6c4be55514fe544b7712309eab910d3dd7c8ba

[zenodo]
token = ysPeKI7yZ3A2qxSTQ5aqQSSTGbJhga7DhVs7KKzBM3vYtVuYK1bAWzO3hl3N

You should obtain API tokens from Github (https://github.com/settings/tokens , https://docs.github.com/en/github/authenticating-to-github/creating-a-personal-access-token) and Zenodo (https://zenodo.org/account/settings/applications/tokens/new/). Your Github and Zenodo tokens should not include any scopes; you only need read-only access to public repositories and records, and to avoid possible low API rate limits enforced when not using an access token.

By default, the importer will store its imports in a sqlite database (e.g., ~/.local/var/searcch-importer.db on UNIX-like systems). You can change this location (or change the database type and credentials entirely) like this:

[db]
url = sqlite:////tmp/searcch-foo.db

(The value for the url key is a sqlalchemy database URL, as described at https://docs.sqlalchemy.org/en/13/core/engines.html#database-urls .)

Finally, by default, the importer will automatically upgrade your database schema each time you run a command that would access the database, unless you set:

[db]
auto_upgrade = false

You should probably leave auto_upgrade set to true unless you’re hacking on the importer tool and adding your own migrations.