Upgrade StorageIs Import the Inverse of Export?
Hi,
I am new to Zotero - just found it on the firefox add-on page and started trying things out. First, I'd like to say thank you to the developers - it really looks promising and I am already thinking of switching to zotero.
Just to get a feeling for how import/export works, I wonder whether there is a format for which 'Import' is the inverse operation to 'Export', i.e.
a) when I export a reference from zotero and then immediately import it, I get a precise duplicate of the metadata in zotero
b) when I have some external file and first import it and then immediately export it, I get a precise duplicate of the external file (perhaps up to ordering/formatting of XML or whatever)
Preferably all data including snapshots, attached files, tags, related items and collection structure.
Here are my questions:
1) Is any of the import/export formats supposed to behave like this? I played around with a few and I am not quite satisfied (details see below)
2) If not, do you think this would be a good idea, preferably some XML format on which you can then operate with external tools
3) Is someone already working on this?
4) If there is no support for this idea, where do I have to look in order to find the source code for import/export of "Zotero RDF"? What should I read in order to understand the format "Zotero RDF"? Is there a DTD or Schema for it? I already had a look at the internal SQLite data base file and I have got an idea of where the data are stored, and I think I could learn some Javascript if needed.
Why am I dissatisfied with the existing import/export formats? I tried the following.
I have Zotero 1.0.1. I go to MathSciNet (http://www.ams.org/mathscinet/), go to an entry for a journal article (I took MR1154897), load the data into zotero. The result looks good to me. Zotero got title, authors, journal ("Publication" field), volume, issue, pages, year ("Date"), ISSN.
When I export this one-item library as "Zotero RDF" and immediately import it again, I do not get a duplicate of the original reference, but the following fields are missing: publication, volume, issue, journal abbrev.
By the way, the firefox cursor keeps showing the busy icon until I quit firefox (some process still running? for minutes?)
Am I supposed to use "MODS" instead? This does fine with the MathSciNet references, but it loses the "Repository" field of data from http://www.arxiv.org. How about snapshots, attached files in MODS?
Finally, I have some unrelated remarks which I can report in other forums if that helps.
1) Looking at userdata.sql, it seems there are some configurable item types (TABLE userItemTypes). What is the status of these? Can I configure them anywhere in Zotero?
2) Zotero extracts most fields from MathSciNet (http://www.ams.org/mathscinet/), but not everything that is possible. For example, on the main article page, there is the (official?) journal abbreviation, but Zotero takes the full journal name as journal abbrev. Also, the "Article" link in MathSciNet is usually just the official DOI link, and so one could also extract the DOI from these pages.
3) For journal articles, I eventually want to store the DOI, the arxiv identifier (http://www.arxiv.org/), the MathSciNet number, my own bibtex label, and perhaps the classmark/callno of our library. Which fields can I use? I'd either like to have more than one "Extra" field or to be able to add my own fields. How would you do that?
4) Is it possible to export only some "collection" rather than the entire library?
5) In each collection or subcollection, the folders are sorted alphabetically. Is it possible to get an arbitrary ordering? Just as bookmarks in firefox?
Any hints, comments, corrections would be appreciated.
G.
I am new to Zotero - just found it on the firefox add-on page and started trying things out. First, I'd like to say thank you to the developers - it really looks promising and I am already thinking of switching to zotero.
Just to get a feeling for how import/export works, I wonder whether there is a format for which 'Import' is the inverse operation to 'Export', i.e.
a) when I export a reference from zotero and then immediately import it, I get a precise duplicate of the metadata in zotero
b) when I have some external file and first import it and then immediately export it, I get a precise duplicate of the external file (perhaps up to ordering/formatting of XML or whatever)
Preferably all data including snapshots, attached files, tags, related items and collection structure.
Here are my questions:
1) Is any of the import/export formats supposed to behave like this? I played around with a few and I am not quite satisfied (details see below)
2) If not, do you think this would be a good idea, preferably some XML format on which you can then operate with external tools
3) Is someone already working on this?
4) If there is no support for this idea, where do I have to look in order to find the source code for import/export of "Zotero RDF"? What should I read in order to understand the format "Zotero RDF"? Is there a DTD or Schema for it? I already had a look at the internal SQLite data base file and I have got an idea of where the data are stored, and I think I could learn some Javascript if needed.
Why am I dissatisfied with the existing import/export formats? I tried the following.
I have Zotero 1.0.1. I go to MathSciNet (http://www.ams.org/mathscinet/), go to an entry for a journal article (I took MR1154897), load the data into zotero. The result looks good to me. Zotero got title, authors, journal ("Publication" field), volume, issue, pages, year ("Date"), ISSN.
When I export this one-item library as "Zotero RDF" and immediately import it again, I do not get a duplicate of the original reference, but the following fields are missing: publication, volume, issue, journal abbrev.
By the way, the firefox cursor keeps showing the busy icon until I quit firefox (some process still running? for minutes?)
Am I supposed to use "MODS" instead? This does fine with the MathSciNet references, but it loses the "Repository" field of data from http://www.arxiv.org. How about snapshots, attached files in MODS?
Finally, I have some unrelated remarks which I can report in other forums if that helps.
1) Looking at userdata.sql, it seems there are some configurable item types (TABLE userItemTypes). What is the status of these? Can I configure them anywhere in Zotero?
2) Zotero extracts most fields from MathSciNet (http://www.ams.org/mathscinet/), but not everything that is possible. For example, on the main article page, there is the (official?) journal abbreviation, but Zotero takes the full journal name as journal abbrev. Also, the "Article" link in MathSciNet is usually just the official DOI link, and so one could also extract the DOI from these pages.
3) For journal articles, I eventually want to store the DOI, the arxiv identifier (http://www.arxiv.org/), the MathSciNet number, my own bibtex label, and perhaps the classmark/callno of our library. Which fields can I use? I'd either like to have more than one "Extra" field or to be able to add my own fields. How would you do that?
4) Is it possible to export only some "collection" rather than the entire library?
5) In each collection or subcollection, the folders are sorted alphabetically. Is it possible to get an arbitrary ordering? Just as bookmarks in firefox?
Any hints, comments, corrections would be appreciated.
G.
This is an old discussion that has not been active in a long time. Instead of commenting here, you should start a new discussion. If you think the content of this discussion is still relevant, you can link to it from your new discussion.