Question about Native XML import

Using OJS3. We are trying to bulk ingest a new journal which has not previously been in OJS. I’ve followed the template from github and have the XML formatted as I understand it to be correct, but I cannot get my test articles to load. (Note that, using oXygen, the XSD would not validate, so was unable to test the XML against the XSD.)

When I import, I just get a blank results page. From a comment on another thread (linked below), it was suggested that this was a PHP error. (For the record, we tested exporting from an existing OJS article, and the reimport works fine.)

Here is the PHP log from one of our blank-result tests:

127.0.0.1 -  26/Mar/2019:16:58:48 -0500 "GET /index.php" 200
10.107.83.239 - - [26/Mar/2019:16:58:48 -0500] "GET
/testpath/notification/fetchNotification?_=1553637519080 HTTP/2.0" 20071 "https://libtstojsap02.lib.ku.edu/testpath/management/importexport/plugin/NativeImportExportPlugin" "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like
Gecko) Chrome/70.0.3538.77 Safari/537.36"
127.0.0.1 -  26/Mar/2019:16:58:48 -0500 "GET /index.php" 200
[26-Mar-2019 16:58:48] WARNING: [pool www] child 44 said into stderr:
"NOTICE: PHP message: ojs2: DB Error: Data truncated for column 'seq'  at row 1"
2019/03/26 16:58:48 [error] 10#10: *174 FastCGI sent in stderr: "PHP
message: ojs2: DB Error: Data truncated for column 'seq' at row 1"
while reading response header from upstream, client: 10.107.83.239,
server: _, request: "GET
/testpath/management/importexport/plugin/NativeImportExportPlugin/impor
t?temporaryFileId=67&_=1553637519079 HTTP/2.0", upstream:
"fastcgi://127.0.0.1:9000", host: "libtstojsap02.lib.ku.edu", referrer:
"https://libtstojsap02.lib.ku.edu/testpath/management/importexport/plugin/NativeImportExportPlugin"
10.107.83.239 - - [26/Mar/2019:16:58:48 -0500] "GET
/testpath/management/importexport/plugin/NativeImportExportPlugin/impor
t?temporaryFileId=67&_=1553637519079 HTTP/2.0" 200 78 "
https://libtstojsap02.lib.ku.edu/testpath/management/importexport/plugin/NativeImportExportPlugin
" "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like
Gecko) Chrome/70.0.3538.77 Safari/537.36"

Here is a sample XML. Maybe I’m missing something obvious? Perhaps something about how the article PDF is included?

<?xml version="1.0" encoding="utf-8"?>
<article date_published="2008-05-01" locale="en_US" section_ref="ART" stage="production"
    xmlns="http://pkp.sfu.ca" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
    xsi:schemaLocation="http://pkp.sfu.ca native.xsd">
    <id type="doi">10.1123/jis.1.1.98</id>
    <title locale="en_US">Issues Related to Academic Support and Performance of Division I
        Student-Athletes</title>
    <subtitle locale="en_US">A Case Study at the University of Minnesota</subtitle>
    <copyrightHolder locale="en_US">Journal of Intercollegiate Sport</copyrightHolder>
    <copyrightYear>2008</copyrightYear>
    <authors>
        <author user_group_ref="Author">
            <firstname>Mary</firstname>
            <middlename>Jo</middlename>
            <lastname>Kane</lastname>
            <affiliation locale="en_US">University of Minnesota</affiliation>
            <email>kuscholarworks@ku.edu</email>
            <biography locale="en_US">Mary Jo Kane is with the School of Kinesiology, University of
                Minnesota. Perry Leo is with the Department of Aerospace Engineering and Mechanics,
                University of Minnesota. Lynn K. Holleran is Associate to the Vice President and
                Chief of Staff, Office of the President, University of Minnesota, Minneapolis,
                MN.</biography>
        </author>
        <author user_group_ref="Author">
            <firstname>Perry</firstname>
            <middlename/>
            <lastname>Leo</lastname>
            <affiliation locale="en_US">University of Minnesota</affiliation>
            <email>kuscholarworks@ku.edu</email>
            <biography locale="en_US">Mary Jo Kane is with the School of Kinesiology, University of
                Minnesota. Perry Leo is with the Department of Aerospace Engineering and Mechanics,
                University of Minnesota. Lynn K. Holleran is Associate to the Vice President and
                Chief of Staff, Office of the President, University of Minnesota, Minneapolis,
                MN.</biography>
        </author>
        <author user_group_ref="Author">
            <firstname>Lynn</firstname>
            <middlename>K.</middlename>
            <lastname>Holleran</lastname>
            <affiliation locale="en_US">University of Minnesota</affiliation>
            <email>kuscholarworks@ku.edu</email>
            <biography locale="en_US">Mary Jo Kane is with the School of Kinesiology, University of
                Minnesota. Perry Leo is with the Department of Aerospace Engineering and Mechanics,
                University of Minnesota. Lynn K. Holleran is Associate to the Vice President and
                Chief of Staff, Office of the President, University of Minnesota, Minneapolis,
                MN.</biography>
        </author>
    </authors>
    <submission_file id="1" stage="submission">
        <revision filename="jis.1.1.98.pdf" filetype="application/pdf" genre="Article Text"
            number="1" viewable="true">
            <name locale="en_US">jis.1.1.98.pdf</name>
            <embed encoding="base64">JVBERi0xLjQNJe...jMwOTA3OQ0KJSVFT0YNCg==</embed>
        </revision>
    </submission_file>
    <issue_identification>
        <volume>1</volume>
        <number>1</number>
        <year>2008</year>
    </issue_identification>
    <pages>98-129</pages>
</article>

Hi Erin:

I recommend that you manually publish an issue in a test OJS (include the PDFs) and then download the Native XMLs. Then compare them with the template that you are using. Remember to add the list of sections, their properties, and asigne them to your articles.

You may download the XMLs both as individual articles and as an issue. You can even download a whole volume in a single XML. In your case it’s better to download the articles individually.

Your example is similar to an XML of a fully published article but it’s not quite the same than the XML that you have written. The template that you are using is for an article that has been submitted and waiting to be reviewed by peers.
The use of the <submission_file/> element is not necessary in this case and instead you should use the <article_galley/> element to incorporate the pdf file to the article.
I’m not so sure about the way that you are using and this is what might be causing the conflicts that you are getting.
The correct way might be {doi}, but the best way to know is by creating a test issue and then take a look at the XML.

I recommend that you read the following thread
https://forum.pkp.sfu.ca/t/preserve-article-ids-in-native-article-issue-import/45548

Thanks for the advice and the clarification. As someone new to OJS, I think I misunderstood some of the terminology (submission vs galley).

We had done a test publish, but probably not with enough detail. I will give that a try and see how it goes.