<div dir="ltr">yes, in source code, there is a project sub dir with LibreOffice spreadsheet with DB schema.<div><br></div><div>For obscure reason, the dir where document are hosted cannot be accessed through git web interface.</div><div><br></div><div>But DB do not use exif info has schema. Look in Exiv2 shared library document to see where face tags are stored in XMP. Exif is not dedicated for that.</div><div><br></div><div>Also we follow MWG to store face tags region : <a href="http://www.metadataworkinggroup.org/pdf/mwg_guidance.pdf">http://www.metadataworkinggroup.org/pdf/mwg_guidance.pdf</a></div><div><br></div><div>Gilles Caulier</div></div><div class="gmail_extra"><br><div class="gmail_quote">2016-06-01 19:33 GMT+02:00 Mateusz Konieczny <span dir="ltr"><<a href="mailto:matkoniecz@gmail.com" target="_blank">matkoniecz@gmail.com</a>></span>:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="HOEnZb"><div class="h5">On Tue, 31 May 2016 11:27:47 +0100<br>
Richard Mortimer <<a href="mailto:richm%2Bdigikam@oldelvet.org.uk">richm+digikam@oldelvet.org.uk</a>> wrote:<br>
<br>
> On 31/05/2016 10:41, Mateusz Konieczny wrote:<br>
> > Is there an existing tool/setting to export/import data created<br>
> > during face tagging?<br>
> I'm not aware of any existing software that will do this. But...<br>
><br>
> ><br>
> > I am also willing to create software to do this do documentation<br>
> > where and how this data is stored also would be useful (though<br>
> > obviously I prefer to discover that needed software already exists).<br>
> ><br>
> > I want to create backup of this data. I really would like to<br>
> > tag faces on my photos and Digikam seems to be currently the best<br>
> > available tool but I want to avoid lock-in (especially as face<br>
> > tagging will consume hundreds of hours for my photo collection).<br>
> ><br>
> > Note that I am not looking for folder with sqlite database file - I<br>
> > want to export this data to some format that will be easily<br>
> > processed.<br>
> I'm not sure what you mean by easily processed. But why do you<br>
> discount the SQLite database? That is a public domain database format<br>
> that is widely supported on just about every computing platform. You<br>
> can use SQL queries to access the data. If you prefer to keep the<br>
> data outside of a database you can dump the data using the sqlite3<br>
> commandline and the .dump command. That will give you a SQL copy of<br>
> the data that you can process/manipulate in your favourite scripting<br>
> language.<br>
><br>
> I would not discount the use of a sqlite database. Your data is<br>
> certainly not locked in and the digikam schema/data layout is public<br>
> so if you do ever need to transfer to a different application it<br>
> should be easy to do.<br>
<br>
</div></div>Maybe I was to quick to consider database as dead end. Unfortunately<br>
"Database schema" at <a href="https://www.digikam.org/docs" rel="noreferrer" target="_blank">https://www.digikam.org/docs</a> links nowhere (to<br>
<a href="https://quickgit.kde.org/?p=digikam.git" rel="noreferrer" target="_blank">https://quickgit.kde.org/?p=digikam.git</a>) but at least it means that<br>
database schema is supposed to be documented somewhere...<br>
<div class="HOEnZb"><div class="h5">_______________________________________________<br>
Digikam-users mailing list<br>
<a href="mailto:Digikam-users@kde.org">Digikam-users@kde.org</a><br>
<a href="https://mail.kde.org/mailman/listinfo/digikam-users" rel="noreferrer" target="_blank">https://mail.kde.org/mailman/listinfo/digikam-users</a><br>
</div></div></blockquote></div><br></div>