Can I use a local bibtex file as feed?

Hello, I am new to Zotero and I am particularly interested in its tagging features and searches in tags.

I am bibtex based, and I do not want to drop bibtex, even if Zotero can export to bibtex I do not want to have anything else that bibtex, just because I can give bibtex to any colleauge and they know what to do. Any other format, even free or OSS, would not work for collaboration as well as bibtex does.

That said, I would like to exploit the tags and search tag in Zotero. For this reason I wanted to let the bibtex data into Zotero as if they were a feed. Then I would use Zotero to track these papers from the bibtex and will add tags to them to be saved in Zotero.

The next time I add anything in the bibtex file Zotero will find the same stuff and will help me find stuff through the keywords, tag and related search features. The newly added bibtex entries, of course, will be new stuff that Zotero needs to digest.

Do you think this workflow can be attained using Zotero?
  • I don't think so, no.
    Look at JabRef.

    (Though I'll say that I think you underestimate Zotero bibtex export. Functionally you can store the data in Zotero and then have a .bib file update automatically as you add items, so I'm not sure why storing the db in bibtex matters)
  • @roberto.franceschini: To be clear, you'd want to make sure you have the Better BibTeX plugin if using a BibTeX-based workflow with Zotero.
  • Yes, maybe I underestimate the import/export Zotero features.
    bibtex+BibDesk has been serving well for over 10y, you see why I do not want to drop it. At the same time the taggin feature of BibDesk are really becoming an issue because it is all too static and rigid.

    For a reasonable on-par exchange I'd need:
    - capability to push citations into LyX (not MS Word!)
    - ability to add references to Zotero via command-line interface from arxiv string or DOI or other forms of ID. a sub-optimal substitute could be adding from any string I highlight in the GUI. In any case this should include retrieval of PDF and renaming of the PDF with a template based on author, title and ID and archival in a folder of choice
    - ability to script the addition of keywords possibly via bash, or python, or applescript, or something else (not to esoteric please(!)

    Maybe Zotero already has all this? I have seen the feature to import from arXiv, but could not find a way to let Zotero also download the PDF and store it somewhere I can access ... any suggestion?
  • edited 7 days ago

    capability to push citations into LyX (not MS Word!)

    http://www.klubko.net/en/lyz/

    ability to add references to Zotero via command-line interface from arxiv string or DOI or other forms of ID. a sub-optimal substitute could be adding from any string I highlight in the GUI.

    This would be feasible to add through something like debug bridge.

    In any case this should include retrieval of PDF and renaming of the PDF with a template based on author, title and ID and archival in a folder of choice

    If we'd trigger Add Item by Identifier from debug-bridge it'd trigger PDF download as usual; http://zotfile.com/ can do renames. "folder of choice" is ambiguous; if with "folder" you mean "Zotero collection", doable, if you mean "folder on disk", you'd be fighting Zotero's workings. I would not recommend it.

    ability to script the addition of keywords possibly via bash, or python, or applescript, or something else (not to esoteric please(!)

    Add by citation key I suppose? debug bridge could facilitate that too.

    Maybe Zotero already has all this? I have seen the feature to import from arXiv, but could not find a way to let Zotero also download the PDF and store it somewhere I can access ... any suggestion?

    I think this is the standard behavior?

    All in all, doable to get close even if some scripting would need to be written, but if you really want a bibtex-centered workflow, you should indeed also look at JabRef. I hardly ever use Word, or Lyx even, I do most of my work in LaTeX, and I prefer Zotero over JabRef. Not because I think JabRef is inferior, but I preferred the UI of Zotero in the early days, and I definitely like the infrastructure that makes it easy to build plugins that can tweak almost everything about Zotero.
  • Thanks for the many suggestions.
    https://github.com/wshanks/lyz works, thanks for pointing it out. I am exploring the rest of the plugins ...
  • The debug-bridge is only a conduit into the Zotero desktop client. It can be used to write scripts that implement the desired functionality, but on it's own, it does nothing.
  • With zotfile I can do a decent autofilename job.

    { %N | %e | %o | %j | %r } - {%a} - {%y} - {%T}

    where the custom wildcards are defined as follows seems to do a decent job

    {
    "N": {
    "field": "url",
    "operations": [
    {
    "function": "exec",
    "regex": "\/(?:.(?!\/))+$"
    }
    ]
    },

    "o": {
    "field": "DOI",
    "operations": [
    {
    "function": "replace",
    "regex": "\/",
    "replacement": "-"

    }
    ]
    },

    "r": {
    "field": "reportNumber",
    "operations": [
    {
    "function": "exec",
    "regex": ".*"
    }
    ]
    },

    "e": {
    "field": "extra",
    "operations": [
    {
    "function": "exec",
    "regex": ":.*"
    }
    ]
    },


    "T": {
    "field": "title",
    "operations": [
    {
    "function": "replace",
    "regex": "\\s*[.:;?!|].*",
    "replacement": ""
    },
    {
    "function": "replace",
    "regex": "[\\\/()~!@#$%^&*{«»„““”‘’|….,;`^<>'}+:?!®©]*",
    "replacement": "",
    "flags": "g"
    },
    {
    "function": "trim"
    }
    ]
    }

    }
Sign In or Register to comment.