Upgrade StorageExport from Zotero 4.0.8 to FileMaker Pro
Hello all
I'm a historian building a relational database tracking relationships between historical entities. For various reasons, FileMaker Pro is likely to be the platform.
I want to not just populate entities in the database, but actually provide supporting documentation to verify their attributes. I would also like to be able to create a single search function that accesses both the entities in the database, and any metadata and indexed text in my related Zotero library.
Therefore, I'm looking for a robust mechanism that would:
* export metadata, notes fields and (ideally) indexed text from Zotero into a format which can be searched in FileMaker (there's no need to work in the reverse direction)
* provide a link from results FileMaker out to our Zotero group library, making sure that it points at the correct, unique record
* drag-and-drop Zotero citation information into notes fields within FileMaker.
Looking for solutions in these forums has left me with a few questions:
* is there ever a stable attribute of any Zotero item which could be considered its unique ID? Preferably, this would hold true for items even if they are moved between libraries, including both standalone and online groups. It seems that an item's rdf:about in the <UserItem> tag is unique to each online library, but changes when you move items between libraries. What about the date added, which in Zotero RDF export produced the <dcterms:dateSubmitted> tag? Does this property track with all Zotero items no matter where they are dragged, even if they are modified?
* is there a preferred export format from Zotero that would be easily read in FileMaker? TEI looked promising, but at the moment it's not an export option with online libraries. It also doesn't include an equivalent to the <dcterms:dateSubmitted> tag. I could possibly work with the xml:id attribute in the <biblStruct> tag, and just accept that if I ever change the author name or year then the links would break.
* would any of the above options allow me to build in a hyperlink from FileMaker direct to the matching entry in an online Zotero library? i.e., could I script a URL link that would incorporate the unique ID attribute?
* is there any way that the full indexing information from a Zotero library could be exported, or is that a bridge too far?
Apologies if these questions seem naive! I'm relatively new to databasing and to Zotero, although I'm fairly familiar with bibliographic software through a 20 years of EndNote. Much obliged for any replies.
I'm a historian building a relational database tracking relationships between historical entities. For various reasons, FileMaker Pro is likely to be the platform.
I want to not just populate entities in the database, but actually provide supporting documentation to verify their attributes. I would also like to be able to create a single search function that accesses both the entities in the database, and any metadata and indexed text in my related Zotero library.
Therefore, I'm looking for a robust mechanism that would:
* export metadata, notes fields and (ideally) indexed text from Zotero into a format which can be searched in FileMaker (there's no need to work in the reverse direction)
* provide a link from results FileMaker out to our Zotero group library, making sure that it points at the correct, unique record
* drag-and-drop Zotero citation information into notes fields within FileMaker.
Looking for solutions in these forums has left me with a few questions:
* is there ever a stable attribute of any Zotero item which could be considered its unique ID? Preferably, this would hold true for items even if they are moved between libraries, including both standalone and online groups. It seems that an item's rdf:about in the <UserItem> tag is unique to each online library, but changes when you move items between libraries. What about the date added, which in Zotero RDF export produced the <dcterms:dateSubmitted> tag? Does this property track with all Zotero items no matter where they are dragged, even if they are modified?
* is there a preferred export format from Zotero that would be easily read in FileMaker? TEI looked promising, but at the moment it's not an export option with online libraries. It also doesn't include an equivalent to the <dcterms:dateSubmitted> tag. I could possibly work with the xml:id attribute in the <biblStruct> tag, and just accept that if I ever change the author name or year then the links would break.
* would any of the above options allow me to build in a hyperlink from FileMaker direct to the matching entry in an online Zotero library? i.e., could I script a URL link that would incorporate the unique ID attribute?
* is there any way that the full indexing information from a Zotero library could be exported, or is that a bridge too far?
Apologies if these questions seem naive! I'm relatively new to databasing and to Zotero, although I'm fairly familiar with bibliographic software through a 20 years of EndNote. Much obliged for any replies.
See that same blog post for hyperlinks to Zotero items.
You can't export the Zotero index, no.
I'm not aware of a preferred format for FileMaker. Bibliontology RDF is probably the richest standard output format, but no idea if FileMaker can be made to read that.
One of the contingencies is that the items in the group library might get updated and we would like to be able to relookup the data in our database periodically, so that the information stays up to date. Is there a way to do this? I am thinking something along the lines of calling that url, but with a parameter so that the server returns the "Bibliontology ZDF" structured data instead of the HTML. Eg, something like:
http://zotero.org/groups/162079/items/7ECMMS88?style=bibliontology_zdf
The "/groups/162079/items/7ECMMS88" part of the URL is the closest thing we have to a unique identifier. If possible, we would like some way access structured data on the bases of this, or some other readily accessible unique identifier. I suppose we could try scraping the HTML but that is not much fun and likely to fall over if the formatting is ever changed at the server end.
Is what we are try to do possible, or can you think of another way to achieve this sort of thing?
Regards
Daniel Kaan
FileMaker Developer
I haven't been following this thread, but you can get RDF from the Zotero API with format=rdf_bibliontology, if that's what you mean. See the API docs for more info.
I'm not sure if I have understood you (or Zotero) correctly as I am fairly new to this, but I think you might be talking about accessing the Zotero library on the LOCAL machine. We have multiple researchers working in a research database and all using a common GROUP library stored on Zotero's servers and this is the one we need to reference.
Ideally, I would like to be able to issue a URL which returns the structured xml data for a particular reference in our group database, instead of returning HTML.
Initially we will import every reference in our group library into the database in the Bibliontology RDF format and parse the data out into fields. This data includes the URL to the item on the server, structured something like:
http://zotero.org/groups/<groupID>/items/<itemID>;
This url returns HTML. I was hoping there might be a url which points to a particular item in this particular group database which would return only the structured Bibliontology RDF data instead. What I would really like is a result in XML instead of HTML.
Is this possible or am I barking up the wrong tree?
https://api.zotero.org/groups/1/items/F94BD3ZJ?format=rdf_bibliontology
But you'll need an API key if the group isn't public. See the docs for details.
I am an intern and my assignment is to link Zotero with FileMaker. However, I lack knowledge in the area of APIs, and my employer is not well-versed in computer science. Thus, I am on my own to complete this task. I would like to know if it is possible for you to help and guide me in accomplishing this task and ensuring the success of my internship. Thank you.
Filemaker Pro is obviously just a general database program, not a reference manager. So the field names used in a custom FP database to store reference data would usually just be whatever the original creator of your database devised. As such there would be no 1:1 mapping of its field names to those in any of the common bibliographic file formats that Zotero can import (or the field names Zotero uses internally)...
https://www.zotero.org/support/kb/importing_standardized_formats
If your use case is indeed the one-off transfer from FP to Zotero, your best bet would probably be picking a straightforward import file format that Zotero can read (eg RIS), and writing a FP macro (or other code language, for example working on a CSV export from FP) to map your database's fields to the required RIS fields.
https://en.wikipedia.org/wiki/RIS_(file_format)
https://groups.google.com/g/zotero-dev
https://www.zotero.org/support/dev/start
But it's not clear why your employer would want such an interface (the fact that they are "not well-versed in computer science" is not encouraging). And it's obviously not a supported use of Zotero or Filemaker Pro. So you might implement it today in some way, and it could stop working tomorrow for reasons beyond your control. If you have not yet scoped out what your employer wants to be able to do with their data - so that possible better/much simpler ways of doing it are canvassed - that would be the first step.
BTW Zotero does not contain bibliographies. It contains 'items' that the user has saved to it (eg individual published research papers, books, etc), organized by the user into collections. As supplied, it is empty. But once populated with such data by the user, they can search for authors it includes, from *within* Zotero.