smarg19

Well, I've already learned something. Your Gnotero-based script can cache even with Zotero running because the libzotero script creates a clone .sqlite database first, and then reads from that. You can clone a .sqlite database, even if it is locked, while you cannot read/write a locked database. I'm about to update my code accordingly. Would you be interested in collaborating to create a single workflow? I tried hard to comment out much of my code, to make it fairly explanatory, so that should help somewhat in reaching a modicum of equal knowledge. It will take me a while to digest all of your code (I'm still fairly new to Python, plus, like most decent workflows, there are alot of moving parts and scripts). But I think having one kick-ass workflow for Zotero would be awesome. And I know splitting the coding would save some time (after the initial catch up on each ind. project). It will undoubtedly take time to figure out which parts to merge and how best to merge them, so if that sounds to time-consuming, we can go our separate ways. Tho I would still like to chat. Let me know what you think, stephen

Alright. Thanks. I can see the code. Could you also push the actual .workflow file? Its really hard to get a sense of the structure without the big-picture as well. I already see that we both worked off of libzotero. I ended up highly modifying that stuff (creating my own hybrid versions of sqlite queries) but its at the foundation of my workflow as well. With your auto-cache updating, did you have any problems when Zotero itself was open. I've found that when the app is open, the sqlite database is locked, so I can't read any data. It's primarily for this reason that I made the cache updating manual (so the user can ensure that the app is closed). Anyway, I'm going to explore some, and I'll get back to you. I'd love to hear your thoughts tho on the functionality you were envisioning. It seems like the goal was to have a simpler search mechanism. So the actions you could take on the item once you chose it was either to open its attachment, or open the Zotero app at that item? I'm interested in formatting the item for inserting in a document, as well as your functions. But like I said, I haven't gotten to that stuff yet.

The first collection of code is now up on GitHub. The project isn't complete, particularly on the backend. I haven't finished my code for formatting the data once you choose an item. I've spent a vast majority of my time on the true heart of the workflow, which is the script that reads the user's Zotero sqlite database and creates a JSON cache of the citation information. I have also written filters to search that cache generally, title-specific, author-specific, tag-specific, and collection-specific. This workflow is written in Python using the alp module. All of the code can be found here: https://github.com/smargh/ZotQuery Feel free to fork anything and help me along.

I have a beta version of a Zotero workflow that I've just gotten off the ground. No actions yet, I've just been fiddling with the script filter. I think I've figured out how to access the local zotero database, regardless of the user (assuming it is consistently located nested within the ~/Library/Application Support/Zotero folder). It is also pretty darn fast. I would love to have a really robust Zotero + Alfred workflow out there, so I'm willing to merge/work together. Hopefully I will get this beta code up on GitHub soon. If you're interested tho, just shoot me a message. stephen

I'd love to see a beta of this workflow, and I'd also be willing to work on keeping it running. Is it up on GitHub yet?