Mimic behavior of \cites without BibLaTeX?

Per this answer, BibLaTeX allows for the \cites command to combine multiple citations with page numbers. I was wondering if there's a solution to allow for similar (read: identical) functionality using basic BibTeX?

I'm not sure if it's needed but I'm writing LaTeX and compiling via pdflatex + bibtex.

So, to be explicit: I'd like to use something of the form

\cites[arg1]{key1}[arg2]{key2}

and end up with [val1, note1; val2, note2] where val1, val2,... and note1, note2,... are the citation values and document notes, respectively, determined by my .sty file and by arg1, arg2, ..., respectively.

• Please: how can anybody possibly answer this without a minimal example? A MWEB is definitely needed here. We don't even know what \cite[arg]{key} produces. Plus, the behaviour you describe may be somewhat similar to \cites but it is certainly not 'identical' and, since you emphasise that mere similarity is insufficient, this seems problematic. Especially given that \cites will not necessarily produce the results you state as desiderata if using Biblatex. So it can't hard code that and be 'identical' as a macro. Besides, why not use Biblatex? – cfr Jan 4 '16 at 2:21
• You can probably cobble something together with natbib and xparse if you are determined to do things the hard way. But the power and flexibility of Biblatex cannot be emulated using BibTeX and the kind of relatively simple framework provided by natbib. (Which is not a criticism of that package or BibTeX: they are just older with more limited aims and assume more constrained resources.) – cfr Jan 4 '16 at 2:25
• @cfr - Thanks for your comments! Perhaps my use of identical was a bit drastic. If your comments tell me anything, it's that (a) I need to do better about providing MWE with things (sorry about that!) and (b) perhaps I should try to find a way to make my project framework a bit more flexible so as to allow natbib. I'll work on the latter now and will attempt to adopt the prior when posting here in the future. Many thanks for your insight, and sorry again about the deficiencies in my post. – cstover Jan 4 '16 at 2:42
• I look forward to seeing the clarified version! natbib was really just a guess because a lot of people using BibTeX use it. But what you are using affects what would work as a solution, which is why a minimal example is so important. And it would be helpful to know why Biblatex isn't an option since that is the obvious solution here. – cfr Jan 4 '16 at 3:18
• @cfr - Thanks again! I posted a solution as an answer below rather than re-editing the question itself. In the answer, I provide some details about hesitation to use bibLaTeX, etc. :) – cstover Jan 4 '16 at 3:38

After talking with @cfr above, I realized that finding an alternative to bibLaTeX would likely be more of a headache than figuring out bibLaTeX itself.

Turns out, many of my woes were the result of my being completely ignorant with respect to the package. After some exploration and some tinkering, I came up with a MWE that does what I want:

\documentclass{article}
\usepackage[backend=bibtex,style=alphabetic, citestyle=alphabetic]{biblatex}
\usepackage{stoversymb}

\begin{document}
Some words \cite{Fenley1}.

Some more words \cites()()[pg~1]{Fenley1}{Calegari1}[par~123, thm.~4.1.2.7(a)]{C&D}
\printbibliography
\end{document}


The output:

I'm still not 100% sure that this solution will work with my overall project. Long story short: I'm writing a dissertation using a University-built (and -enforced) style file that is, at best, poorly written and clumsy (to the point that it takes away a lot of foundational functionality of packages like subcaption and subfigure, etc.). Even so, having this knowledge will make my attempts more streamlined and focused.

Thanks again for taking the time to put up with my poorly-presented question; I truly do appreciate your insight!