search

WingLongitude / lontra-harvester

Lontra is a tool used as a Harvester to ingest biodiversity data
MIT License
1 stars 3 forks source link

Export/import publisher/resource metadata #38

Open tigreped opened 8 years ago

tigreped commented 8 years ago

Currently, harvesting data is input manually. There is no way to export or import this data to/from SQL. Implement in order to provide a proper method of avoiding rework.

cgendreau commented 8 years ago

Why would you export/import publisher?

tigreped commented 8 years ago

As a backup measure. So far, we need to enter publisher/resource information manually. If there is any problem with the database, currently, we already would have to reenter 91 publishers and 160 resources.

Once this process is automated in the indexer to fetch this info from the IPT url, for instance, this will still be useful to export/import the list of IPTs to be managed.

cgendreau commented 8 years ago

Why not simply using a pg_dump -t ?

tigreped commented 8 years ago

I'm always keeping the angle that there should be user interfaces that are friendly enough so that non IT people can manage harvesting. The way I see it, that eliminates the dump possibility. What do you think?

cgendreau commented 8 years ago

The lontra-harvester should be considered an expert tool that requires documentation and an IT person to at least install and updating it. I don't think a such feature should be implemented unless all the other issues are solved. I would rather document the procedure to backup and restore the tables.

tigreped commented 8 years ago

I agree it should require an IT person to install and update it (as the versions improve, right?). I also agree we can document the procedure to backup and restore these basic tables. I also agree we should solve other issues first, but I don't see how that removes the importance of this issue. If the issue makes sense, despite of other priorities, I suppose we should keep it open, right?