[texhax] create float based on longtable?

Christoph Haug christoph.haug at gmail.com
Thu Jan 28 18:38:22 CET 2010


Thank you, Susan, for your quick response.

2010/1/28 Susan Dittmar <Susan.Dittmar at gmx.de>:
> Dear Christoph,
>
> it is impossible to have *floating* elements longer than a page.

As far as I understood, that's the general rule, but then I thought
longtable behaves like a float, but maybe I'm confusing the
possibility to refer to longtables as "table 4.1" etc (using \cref and
\caption).

> But as far as I understand, you do not need those examples to float.

well, it would be better if they float, since many of them are small
enough to float around an hop into wherever they fit. An example
should not be split if it is smaller than one page.

> For boxing you
> could use the framed or the boites packages. Both have restrictions for
> use, and those restrictions differ. (In case you need translation of the
> french comments in the boites package, I can send them to you.)

I don't think the framed package will do the job, at least not if I
want all examples to float. And if I use non-floating examples, I
might aswell use longtable or supertabular (the latter even gives me
the possibility to have a standardline at the bottom of a page
"continued on next page" or so.)

But let's assume for a moment that I decide manually, which examples
should float (the small ones) and which dont (the long ones). How do I
manage that both my floating and my non-floating examples use the same
caption numbering and are listed in the same "list of examples"?
Before I can even think about that question, I need to know how I can
tell longtable or supertabular that it should use the "Example" as a
caption rather than "Table" and use the corresponding list. Or
rather---instead of changing the longtable environment---how do I
create a "duplicate" of longtable (or supertabular) which behaves like
an example (except that it doesn't float?

Do you understand what I mean?

> For the caption and naming/referencing part I would have a look at the AMS
> theorem stuff. I never used it myself, but it sounds like solving a similar
> problem.

I never used it myself but looked at the amsthm package now. I
couldn't see in what way it might be usefull. Could you explain?

Best,
Christoph

>
> Hope this helps,
>
>        Susan
>



More information about the texhax mailing list