alecigne / deezer-datasync

Java application that backups data from Deezer.

deezer-datasync's Issues

Run a build in PRs

Group dependabot updates

Just apply the groups option (see docs).

Creation time in playlists is actually local time

All creation times are UTC:

• time_add in albums ✅
• time_add in artists ✅
• time_add in the tracks of a playlist ✅

...except in the info section of playlists;

• creation_date is a local time

• time_add is a local time (last time a song was added), despite being a UNIX timestamp

• time_mod is a local time (last time the playlist was modified, either by adding a song or by changing something like the description -- I don't really understand the behaviour)

• When backuping playlist info, I should either reflect the fact that this is a local time or convert it to UTC using the zone in the configuration file. However, I have to be sure that the local time that Deezer uses is always the same for a given user (from the "preferences" in the user profile). If they take the local time of the place from which the playlists is created instead, I can't convert to UTC since I can't guess the zone.

• Maybe I should also add time_add and time_mod in the backup.

Improve logs

• INFO at most
• Improve log quality

Automate version management

For every new release I need to:

• Create a release version in pom.xml.
• Create an entry in the CHANGELOG.
• Tag the commit.
• Push the tag and then the commit.
• Create a commit to prepare the next development version.
• Create a release with a JAR inside.

This needs to be automated.

Implement caching in workflows

Create Gitpod environment

Do not commit if commit is empty

Comparing the SHA of the trees should be enough (I need to test this):

if (newTree.getSha().equals(baseTreeSha)) {
    log.debug("The commit is empty. No changes to push.");
    return; // Skip the commit creation and update
}

Create more options for backup of individual playlists

At this time a list of playlist IDs has to be provided in the config file. It would be nice to also give the possibility to backup all playlists by using the IDs gathered from the "playlist info" data.

Improve playlist filenames

The playlists directory contains files whose filename is in the [id].json format. That makes identification of a given playlist difficult. The goal of this dev would be to include the playlist title in the filename, but in downcase and cleaned from any undesired characters (whitespace, dashes, special characters, etc.).

• Hip-Hop -> [id]_hip_hop.json -- downcase, dash transformed into underscore
• Funk and Soul -> [id]_funk_and_soul.json -- downcase, whitespace transformed into underscore
• Funk & Soul -> [id]_funk_soul.json -- downcase, whitespace and "&" transformed into underscore, more than 1 underscore limited to 1 underscore
• Downtempo / Nu Jazz -> [id]_downtempo_nu_jazz -- downcase, whitespace and "/" transformed into underscore, more than 1 underscore limited to 1 underscore
• Drone Zone [Study / Relax / Meditate] -> `[id]_drone_zone_study_relax_meditate.json -- downcase, whitespace, "[", "]" and "/" transformed into underscore, more than 1 underscore limited to 1 underscore

Use jqwik to write a property-based test for this logic.

Create a code coverage badge