Upgrade StorageWhy is unAPI import of format=MODS different from importing raw MODS file?
I am interested in figuring out how to spec suggested improvements to Zotero import from the DLF Aquifer site: http://www.dlfaquifer.org/
That site has a unAPI interface that exposes records in format=MODS. The test record I'm working with is:
http://www.dlfaquifer.org/search/item/Libertys-national-toast/oai%253Alcoa1%252Eloc%252Egov%253Aloc%252Emusic%252Fsm1880%252E10201
If I select the icon from my Firefox toolbar, it pulls from:
http://www.dlfaquifer.org/unapi?id=86841&format=mods
So far so good. But if I save the XML from that page and then import that same MODS file, there are some differences in mapping. Importing from the raw MODS file omits the "Accessed" and "Repository" fields which makes sense to me, but it also omits the Tags field and the "Short Title" field, which I don't expect.
My questions:
1. Where is the actual code that is doing the field mapping in both cases? Which translators should I ask my developers to look at? I would have thought that both cases (coming from unAPI where format=MODS and doing an import of a MODS file both use the doImport on the MODS translator. But the fact there are discrepancies makes me thing there must be something else at play.
2. What's the best way that us non-techies can import test MODS records without having to load these into http://www.dlfaquifer.org/? We metadata folks want to futz with a bunch of fictitious MODS records to see how well they import into Zotero so we can spec the changes we'd like to see. Is it possible for us to do this without asking our developers to load records into our web interface? We want the import/mapping behavior to be as close to the "real" thing as possible without having to load these test case records into the live site.
Thanks in advance!
That site has a unAPI interface that exposes records in format=MODS. The test record I'm working with is:
http://www.dlfaquifer.org/search/item/Libertys-national-toast/oai%253Alcoa1%252Eloc%252Egov%253Aloc%252Emusic%252Fsm1880%252E10201
If I select the icon from my Firefox toolbar, it pulls from:
http://www.dlfaquifer.org/unapi?id=86841&format=mods
So far so good. But if I save the XML from that page and then import that same MODS file, there are some differences in mapping. Importing from the raw MODS file omits the "Accessed" and "Repository" fields which makes sense to me, but it also omits the Tags field and the "Short Title" field, which I don't expect.
My questions:
1. Where is the actual code that is doing the field mapping in both cases? Which translators should I ask my developers to look at? I would have thought that both cases (coming from unAPI where format=MODS and doing an import of a MODS file both use the doImport on the MODS translator. But the fact there are discrepancies makes me thing there must be something else at play.
2. What's the best way that us non-techies can import test MODS records without having to load these into http://www.dlfaquifer.org/? We metadata folks want to futz with a bunch of fictitious MODS records to see how well they import into Zotero so we can spec the changes we'd like to see. Is it possible for us to do this without asking our developers to load records into our web interface? We want the import/mapping behavior to be as close to the "real" thing as possible without having to load these test case records into the live site.
Thanks in advance!