Upgrade StorageImporting pdf as LINK to hd located file -- how to?
Hi everybody!
I wasn't sure if this is really a request for a new feature, so I decided to post it here in the Import/Export category, where it fits for sure.
In fact the topic has already been discussed:
https://forums.zotero.org/discussion/2142/possible-to-import-pdfs-as-links-not-copies/
https://forums.zotero.org/discussion/8584/drag-links-to-files/
https://forums.zotero.org/discussion/18052/ris-import-link-not-file/
Reason for a new topic is just that these are quite old and maybe more confusing than helpful, don't now.
Let ME try to explain in an easy way:
The following import code (as bib format / for rdf see below):
@book{mayer_2014,
title = {Mayer (2014)},
file = {mayer2014.pdf:C\:\\papers\\mayer2014.pdf:application/pdf}
}
...results always in a copy of mayer2014.pdf somewhere in a firefox-profile/zotero-storage folder.
Consider the difference in Zotero's GUI of the add functions 'Store Copy of File...'(=result) and 'Link to file...'(=wanted!)
Has anybody any clue how to get Links instead of PDF(copie)s?
Would be otherwise an awesome and really necessary feature...
Best regards,
Clemens
(import as zotero-rdf format)
<z:Attachment rdf:about="#item_37">
<z:itemType>attachment</z:itemType>
<rdf:resource rdf:resource="C:/papers/mayer2014.pdf"/>
<dc:title>Mayer (2014)</dc:title>
<link:type>application/pdf</link:type>
</z:Attachment>
I wasn't sure if this is really a request for a new feature, so I decided to post it here in the Import/Export category, where it fits for sure.
In fact the topic has already been discussed:
https://forums.zotero.org/discussion/2142/possible-to-import-pdfs-as-links-not-copies/
https://forums.zotero.org/discussion/8584/drag-links-to-files/
https://forums.zotero.org/discussion/18052/ris-import-link-not-file/
Reason for a new topic is just that these are quite old and maybe more confusing than helpful, don't now.
Let ME try to explain in an easy way:
The following import code (as bib format / for rdf see below):
@book{mayer_2014,
title = {Mayer (2014)},
file = {mayer2014.pdf:C\:\\papers\\mayer2014.pdf:application/pdf}
}
...results always in a copy of mayer2014.pdf somewhere in a firefox-profile/zotero-storage folder.
Consider the difference in Zotero's GUI of the add functions 'Store Copy of File...'(=result) and 'Link to file...'(=wanted!)
Has anybody any clue how to get Links instead of PDF(copie)s?
Would be otherwise an awesome and really necessary feature...
Best regards,
Clemens
(import as zotero-rdf format)
<z:Attachment rdf:about="#item_37">
<z:itemType>attachment</z:itemType>
<rdf:resource rdf:resource="C:/papers/mayer2014.pdf"/>
<dc:title>Mayer (2014)</dc:title>
<link:type>application/pdf</link:type>
</z:Attachment>
This is an old discussion that has not been active in a long time. Before commenting here, you should strongly consider starting a new discussion instead. If you think the content of this discussion is still relevant, you can link to it from your new discussion.
If there wouldn't exist a manual function for that in the GUI of Zotero, I really would not claim that, but shouldn't be that relativ easy to implement? I'm just wondering... really not blaming...
This function is really important for massive imports. And that's somehow the requirement for using this amazing software.
Does anybody know a way around the importing way? (API or something)
https://github.com/zotero/zotero/issues/329
I think the biggest issue about implementation is actually the UX part:
how/when does a user decide on the behavior. This is hard to get right--as a general rule, we'd want people who don't know exactly what they're doing to store the PDFs in Zotero (since that's the only way to get some of the functionality, like file syncing). But on the other hand, since this option will principally benefit first-time users, we also don't want to hide it too much. Turning this into a hidden preference seems rather unattractive: I don't really want people's first interaction with Zotero to be to fiddle with about:config
(There is also some code that would need to expose the ability to import as links to import translators, so this isn't completely trivial from that perspective, either, but that's the easier part here).
Well, and in fact I didn't know from this (second) newest discussion:
https://forums.zotero.org/discussion/29033/bibtex-and-other-bibliographic-import-adding-pdfs-as-links-instead-of-attachments/
So as far as I understand you, the technical side should be not the main issue. I try to consider the UX argument:
You want to have as default the PDF-store-function, so you're not sure how to implement additionally the link-function without hidding it too much (for the first-time users).
Okay, can follow you so far, I hope...
but wouldn't all of these criteria be met, if there is just an element in Zotero's rdf format like <z:itemType>link</z:itemType>
- or is this breaking too much the (at all not really guaranteable) compatibility within the import/export formats?
Really just try to understand... (also because I'm interested in general about the working background of decisions and programming from such (open source) projects)
Thanks for your both continual
invested time in that topic ;)
Clemens
That doesn't, unfortunately, solve this issue for most users, though, since no one else exports Zotero RDF. So, this would still require us to find a way to do this for RIS and BibTeX, where a) we can't just decide to change something about the standard and b) there wouldn't be an easy way to distinguish types of links anyway.
Further more, as far as I see, there is no working solution otherwise existing!? I'm not familiar with the other formats, but if they don't distinguish between links vs. stored files you will never be able to have a fully featured import/export function for all formats, which allows you to have both, "linked" and "stored" entries. Will you??
(What is your use case, btw.? I see few scenarios where you'd import Zotero RDF with local links).
So for really big archives, like I have, this could become tremendous time-consuming -
in fact in my case not, because I don't want to store any (copied) files, only have links - for that reason that kind of implementation would be quite fine for me, too (although less appropriate in my eyes).
In fact I'm not coming from another reference manager, so I just planning to create one big import (plain text) file, based on a hd search result. Do you see a solution for that? Could you explain these 'scenarios' more detailed?
And my last thought; to the formats intercompatibility claim:
The formats seem all quite similar to me, so shouldn't it be relative easy to convert between them, wouldn't be surprised if there are even already converters existing!?
Based on that, a move from one reference manager to Zotero shouldn't be a big barrier anymore...
But you'll have to trust me that this won't help any of the other people who have requested this in the past. There is no RIS to Zotero RDF converter and I doubt there should be.
would you think, a new issue for that Zotero-RDF-implementation would make any sense? or adding a comment in the already existing (329) one?
btw, your 'few scenarios' sentence I've read quite wrong (sry). Here my comment after right understanding:
In fact I would thing it's not that 'few', because it would relate to all new customers (for example students or even later academics who realize how valuable a reference management is, and come by this to Zotero), who already have a huge archive of papers and ebooks
-- edited: but instead of drag and drop prefer an automatically batch-like import.
(see for example https://forums.zotero.org/discussion/23748/automating-massimport-from-pdfs/ )
@aurimas: good point ;)