EPrints Technical Mailing List Archive

Message: #09344

< Previous (by date) | Next (by date) > | < Previous (in thread) | Next (in thread) > | Messages - Most Recent First | Threads - Most Recent First

Re: [EP-tech] bulk import terminates with <rev_number>0</rev_number>

Hi Thomas,

Using the text file you attached with the EPrinst XML I have tested this on my dev server running EPrint 3.4.5 and I could import the two records using he "EPrints XML" importer from a web browser.  I did get a few warnings messages about unrecognised entities (all three appeared twice):

Invalid XML element: repository
Invalid XML element: folder
Invalid XML element: FKZ

I was similar able to import these on a 3.4.4 repository although this has local modification, which may have fixed the issue.  However, the issue you are reporting does not sound familiar to me, so I don't think it is something purposefully fixed for 3.4.5.  Two places worth checking are the known issues section for 3.4.4:

https://wiki.eprints.org/w/EPrints_3.4.4#Known_Issues

and all the GitHub issues for 3.4.5:

https://github.com/eprints/eprints3.4/issues?q=is%3Aissue+milestone%3A3.4.5+

I cannot see anything related to your issue the former, so it may be worth trawling through the latter.  GitHub issues tend to have relatively descriptive subjects.

Looking closer at your error message it suggests that there may be something in your configuration that is causing the issue.  I would probably look for lines like:

$values = &$f( $self, $values, $fieldname );

In particular the '&$' part, which match up with the eval line number.  One of the places you might find these are if you are setting something that is not a function to a field configuration attribute that must be a function.  One such attribute is "fromform".  I think the <rev_number>0</rev_number> is due to the fact that EPrints will create an initial empty record (dataobj) and then try to add the initial metadata provided to create revision number 1.  So it would see the issue with the code reference occurs whilst the metadata is be evaluated and assigned to the dataobj, this will be before it is saved to create revision number 1.

I think that if you remove the fields you are trying to import one by one you should be able to isolate the field causing the issue and then review its field configuration to figure out what about it is causing the problem.

Regards

David Newman



On 20/06/2023 10:09, th.lauke--- via Eprints-tech wrote:
CAUTION: This e-mail originated outside the University of Southampton.
 
Dear all,

currently I have no idea, why following phenomenon occurs:
The bulk import of the attached file (2 nearly identical entities, but second without creators field) terminates (after first entity is imported, but <rev_number>0</rev_number> set) with
'Not a CODE reference at (eval 107) line 42.'

Import of second entity alone works as usual.

What is going wrong in 3.4.4?

Many thanks for any hint in advance
Thomas

*** Options: http://mailman.ecs.soton.ac.uk/mailman/listinfo/eprints-tech
*** Archive: http://www.eprints.org/tech.php/
*** EPrints community wiki: http://wiki.eprints.org/