Uploaded image for project: 'ListenBrainz'
  1. ListenBrainz
  2. LB-171

alpha import for user marc2k3 yields tons of 400 errors on insert to influx.

      This crazy parse error happens:

      https://gist.github.com/mayhem/737f5c9202c2be21219061f030ea1bc7

      We need to ensure that everything works as expected.

          [LB-171] alpha import for user marc2k3 yields tons of 400 errors on insert to influx.

          Param Singh added a comment -

          Each of the listens rejected contains a field called `releasetype` which is a list. These lists cause parsing problems in influx. We were flattening dictionaries inside additional_info before putting them into influx but doing nothing for lists. What would be the preferable way to store lists in influx? For lists of mbids and tags, we personally convert them into comma-seperated strings and we could do this for additional_info fields too, but how will we know that the field is a list when we get it out of influx.

          Param Singh added a comment - Each of the listens rejected contains a field called `releasetype` which is a list. These lists cause parsing problems in influx. We were flattening dictionaries inside additional_info before putting them into influx but doing nothing for lists. What would be the preferable way to store lists in influx? For lists of mbids and tags, we personally convert them into comma-seperated strings and we could do this for additional_info fields too, but how will we know that the field is a list when we get it out of influx.

            iliekcomputers Param Singh
            rob Robert Kaye
            Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

              Created:
              Updated:
              Resolved:

                Version Package
                beta release