search

elifesciences / elife-pubmed-feed

code to support uploading feeds to pubmed for POA articles and VOR articles
1 stars 4 forks source link

Individual members of group authors not getting delivered to PubMed #28

Closed Melissa37 closed 9 years ago

Melissa37 commented 9 years ago

Hi Graham

We've found for 02935 that the group authors listed in the final PoA xml have not been converted and sent to PubMed.

This can be found in commit: https://github.com/elifesciences/elife-pubmed-feed/commit/4871b6d60a5d4d10ec2e27f5a5d5f99876c32fad

The eLife xml is: https://github.com/elifesciences/JATs-xml/commit/63595d71d9bab4ba342172643cd4e8aea9173c2e

Can we add this in ASAP?

Sorry, I did not realise this was missed off. M

gnott commented 9 years ago

My understanding is these are contributors linked to a group author by a contrib-id? E.g.

<contrib-id contrib-id-type="group-author-key">group-author-id1</contrib-id>

02935 as an example, I don't think the data of the actual group author's contributors are included. For this I may need to punt it back to James for more query development.

Melissa37 commented 9 years ago

This is a VoR stage item. EJP does not store this data, so we cannot do it at PoA, but the final xml TNQ produces contains it where present at VoR stage.

gnott commented 9 years ago

I understand now, it's the VoR XML to Pubmed issue. I will take this on!

Melissa37 commented 9 years ago

Fab thanks! Lets chat about timelines on Thurs

gnott commented 9 years ago

In parsing the XML file, only these were extracted as contributors

<contrib contrib-type="author">

Individual contributors as part of a group are specified as

<contrib contrib-type="author non-byline">

These non-byline authors will be added to the end of the contributor list.

Crossref XML generation already only includes contrib-type="author" contributors.

Pubmed XML generation will be set to include both types, contrib-type="author" and contrib-type="author non-byline".

Melissa37 commented 9 years ago

Hi there This is fixed now isn't it? Closing the ticket. M

gnott commented 3 years ago

Sitting down today, I resolve some errors with existing test cases and trying to confirm the strategy was sound I was following. The code change is in PR https://github.com/elifesciences/elife-pubmed-xml-generation/pull/44.

It is an improvement, I think, to remove the group_author_contrib_types configuration file value, because it was not working and was going to cause more trouble than instead relying on contributor group_author_key attributes that we are already setting when parsing the article XML.

I feel confident with the test coverage and the impact of the change, it's ready to merge and deploy, which I will move to do today. We can adjust or rollback the week of January 24th if we observe any errors.