I am having a problem with the m-bib module, in that it does not seem to be handling crossrefs properly. For example, in my bibtex database I have the following entries:

@inproceedings{ bos92sigs, author = "Jurjen N. E. Bos and David Chaum", title = "Provably Unforgeable Signatures", crossref={CRYPTO92}, pages = "1-14"}

@proceedings{CRYPTO92, editor = {Ernest F. Brickell}, title = {12th Annual International Cryptology Conference, Santa Barbara, California, USA, August 16-20, 1992, Proceedings}, booktitle = {Advances in Cryptology - CRYPTO '92}, publisher = {Springer}, series = {Lecture Notes in Computer Science}, volume = {740}, year = {1992}, isbn = {3-540-57340-2}, bibsource = {DBLP, http://dblp.uni-trier.de%7D }

if I cite bos92sigs in a LaTeX document, the following reference is produced:

[3] Jurjen N. E. Bos and David Chaum. Provably unforgeable signatures. In Ernest F. Brickell, editor, Advances in Cryptology - CRYPTO 92, volume 740 of Lecture Notes in Computer Science, pages 1-14. Springer, 1992.

But ConTeXt + m-bib produce: [4] Jurjen N. E. Bos, and David Chaum, (1992). Provably unforgeable signatures. In [??], pages 1-14.

What this looks like is that m-bib does not handle bibtex crossrefs properly; it tries to create another entry in the publication list and put a reference to that new entry in the crossrefrencing entries (although it did not add an entry for the CRYPTO 92 proceedings in my publication list...). What it should be doing (according to the bibtexing document) is supplying the missing entries (such as booktitle, publisher, etc.) to the crossreferencing entry, and then creating a new entry in the publication list only if the same item is crossref'ed by min-crossrefs (default=2) normal entries.

Is this indeed the case? If so, is there a simple fix (e.g. some option that is not in the m-bib documentation)? Or is there some other bibliography package that handles bibtex crossrefs properly?

Thanks, William D. Neumann

---

"Well I could be a genius, if I just put my mind to it. And I...I could do anything, if only I could get 'round to it. Oh we were brought up on the space-race, now they expect you to clean toilets. When you've seen how big the world is, how can you make do with this? If you want me, I'll be sleeping in - sleeping in throughout these glory days."

-- Jarvis Cocker

Hello Thomas

[...]

as run texexec on the file I'm editing). A while ago, Patrick mentioned that new versions of auctex will be context-aware, but the "stable" version I dlded knew nothing about ConTeXt.

It is in the cvs-version of AUCTeX.

Has anybody succeeded in installing a new version that's working? Any how-tos about it?

Do you know how to use cvs? If yes, look at

http://savannah.nongnu.org/cvs/?group=auctex

for instructions on cvs access.

Patrick

Hello Thomas,

sorry, I'm too dumb for CVS. I succeeded in finding your file context.el and dlding it, but how do I make it work in auctex? Is it sufficient to just copy it into site-lisp/auctex/ ?

Oh, no. context.el only works with the cvs version of the other files that come with AUCTeX.

I guess I'll have to modify tex-site.el, right? Sorry for those stooooopid questions, but these are my first steps with emacs--it took me ages just to find out how to make it use mac-roman in the right way so I could use files in it and TeXShop.

If you are interested, I can send you the current cvs version of AUCTeX off-list. You need to follow the install-instructions that are mentioned in file README.CVS. But: you need autoconf installed on your system for the cvs version.

Patrick

On Fri, 28 Nov 2003 13:28:51 -0700 (MST), William wrote:

But ConTeXt + m-bib produce: [4] Jurjen N. E. Bos, and David Chaum, (1992). Provably unforgeable signatures. In [??], pages 1-14.

If you run ConTeXt one more time, you will get:

Jurjen N. E. Bos, and David Chaum, (1992). Provably unforgeable signatures. In (Brickell, 1992), pages 1-14.

The module inserts a 'default' \cite command for the referenced item. This is exactly the behavior I intended, even if it is not compatible with LaTeX practise. Personally I like 'real' cross-referencing better than filling in fields in order to minimize the number of keystrokes in the BIB file.

Unfortunately, there is an error in the code that prevents the referenced entry from automatically appearing in the publication list.

A quick hack: replace '\completepublications' with

\setbox0\vbox{\placepublications} \completepublications

This will essentially typeset the list twice, the internal \cite will be processed in the first pass, and will therefore be known in the second pass. I have written down this bug for a future release, implementing a clean solution for this bug is too hard for me to allow me to write a fix right this instant.

Is this indeed the case? If so, is there a simple fix (e.g. some option that is not in the m-bib documentation)? Or is there some other bibliography package that handles bibtex crossrefs properly?

I don't think there is any other package. If there was, I wouldnt have written m-bib in the first place.

A little side note: "-min-crossrefs" is not applicable since the entire database *always* appears in the bbl file. This is a problem that is caused indirectly by the fact that Bibtex is not a generic bibliographic manager, but instead is a LaTeX-centric data preprocessor. :-/

Greetings,

Taco

On Mon, 1 Dec 2003 12:00:08 -0700 (MST), William wrote:

...

Personally I like 'real' cross-referencing better than filling in fields in order to minimize the number of keystrokes in the BIB file.

I suppose that's a valid sentiment, although it saves many, many keystrokes when your bib file is full of proceedings and journal articles. Plus it looks fairly silly to reference the proceedings when you reference only one or two of the articles in it (the same goes for books, journals, etc.). Oh well... until I'm willing to write my own package I suppose I have to play along or go back to LaTeX.

If there are more people that agree with you I might change my mind ...

However, the entry for the proceedings themselves have the wrong number, they get reset and start over from 1 e.g.

Do You know of any way around this, or am I just going to have to manually edit the bbl file?

You could always \cite the proceedings manually. If you put the reference in a \setbox command it wont show up in the running text, like so:

\setbox0\hbox{\cite[XX]}

Wrt to your other questions: could you send me your input file(s) or the exact settings you used? It will be (a lot) easier to re-enact your observations if I do not have to guess and invent stuff myself.