Closed val99erie closed 6 years ago
I just ran through these instructions on http://epigaea.curationexperts.com and got the expected behavior. see: http://epigaea.curationexperts.com/xml_imports/50
I can't reproduce this either. @val99erie Are you still seeing this?
@bess Sorry, there was some discussion on slack that didn't get added to this ticket. Tom realized that the order that the files are uploaded in the UI affects whether or not the error happens. I think he's working on a solution.
@no-reply I'm re-assigning this to you, since it sounds like you're working on it with more context than I have
Steps to reproduce the problem:
Edit
docs/all_inputs.xml
and delete the<tufts:id>
entry from the file.Create the sample files that are listed in the
filename
element indocs/all_inputs.xml
:Visit
http://localhost:3000/sidekiq/queues
and note the number of jobs that have already been processed so you can figure out if any new jobs get queued during this batch import.In the browser, select "Import Objects" from the nav menu
When the form loads, upload the
docs/all_inputs.xml
file, and click 'NextClick "add additional files to this batch"
Click "Add Files" and upload the 2 sample files you created in the steps above.
Click "Add Files to Batch"
Expected Behavior:
If you wait a few seconds (long enough for background jobs to run), the batch status will change to 'Complete' and you will see a link for the newly-created record. If you click the link, you'll see the new record with your 2 files correctly attached to it.
Actual Behavior:
If you wait and reload the page, the batch status is still 'Unavailable', and if you visit
http://localhost:3000/sidekiq/queues
, notice that no new jobs have been queued or processed.Note:
If you remove one of the two
filename
entries fromdoc/all_inputs.xml
, then the import will work as expected.