Currently, the scripts are painful and slow to use to download and convert the gnd data.

The script downloads and processed 3 different files, in sequential order. If the third download fails or I have to cancel because I want to stop it and continue later (because it takes so long), then the only choice is to restart the entire script (e.g. re-download and process the first and second thing).

It would be nice if it I could have it not re-download stuff and be able to continue with the third step, without repeating 1 and 2.

It would also be nice if the script could be made to run faster.

Finally, in my last attempt just now, i encounted a DownloadError timeout for the third dump file, and then the script dies. (so i have to restart the whole process and somehow increase the timeout)

PS - also nice if these were not maintained on github (https://github.com/WikidataQuality/DumpConverter)

I consider this a blocker for deployment, as I struggling a bit to be able to produce any csvs, which we would need for Wikidata.