www type bibtex entries - generating bibtex for webpages + prior theme.

Peter Flynn peter at silmaril.ie
Mon Sep 16 00:06:05 CEST 2019


On 15/09/2019 22:43, Mike Marchywka wrote:
> On Sun, Sep 15, 2019 at 11:14:59AM +0100, Peter Flynn wrote:
>>
>>
>> Use biblatex for formatting, not BiBTeX, because the older formats tend not
> 
> I tried to install it but I'm using revtex4 and probably ran into this problem
> ( it is loading natbib so I assume same thing but don't have time to track down now),

I've never used revtex so perhaps it is built solely on old BiBTeX 
(loading natbib would be a symptom of that). Pity.

> I guess one thing that may be nice is if sites had a meta field for a link to the bibtex for that page.

a) it certainly would

b) they already have: that's what Dublin Core, COinS, and other metadata
    formats are for: embedded in the headers: another things on the cards

c) no-one out there has ever heard of bibtex, so asking for a special
    field for it is a non-starter.

The metadata at [b] can be used by citation-grabbers like Zotero and 
Mendeley to get the data and save it as a .bib entry.

> [...] Note also that I added the source url to the bibtex although the code
> has become such a mess that is not always the case but in theory I can
> clean up any downloaded stuff, run it in a dummy file to make sure bibtex likes
> it, and then finally copy to my main bib files.   Once I made a "plainurl.bst"
> I wanted to make sure all my entries had a url field ...

Yep. But if I add the DC and COinS metadata, all that can be done for 
you by clicking on the icon that Zotero will pop up automatically, and 
saving the entry.

> And the reason I gave up on bash for c++ is all the tentative junk on sites
> that don't support bibtex. In the bash scripts, I was struggling to manage the
> tentative results and back up - I ended up with temp files etc.
> The c++ has data structures for all these

In my case, I'd use one of the automated metadata grabbers and save the 
whole lot as a non-committal XML file. Then convert to BiBTeX or 
whatever format du jour people want (JSON, EndNote, etc).

But the digging for unannotated sites will always be a pain.

Peter


More information about the texhax mailing list