On 4/11/2018 10:51 PM, Alan Braslau wrote:

On Wed, 11 Apr 2018 22:41:41 +0200 Hans Hagen wrote:

...

...

Question2: What is going on with \setupfloatcaption ? some heuristics to determine the best width ... so that you still get a reasonable caption with very narrow figures and not excessive captions with too wide figures and when possible a one liner ... it's no problem adding some variants (there are some dimensions that can be set already) but the current mechanism is not going to change

Hans, I did not ask about funny behavior with captions in general, just about the handling of key=value for align and with location=. (And no MWE needed, as I am asking a question about the interface.) sure but i want to see why something you expect to work doesn's

tolerance in text/par alignment is just that: adding tolerance and it is bound to align and it only kicks in in par building and there is not always a par involved location has to do with the placement and float as a whole and there tolerance has to deal with snapping so ... unrelated tolerances Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl -----------------------------------------------------------------

On 4/12/2018 5:54 PM, Alan Braslau wrote:

On Thu, 12 Apr 2018 10:09:58 +0200 Hans Hagen wrote:

...

tolerance in text/par alignment is just that: adding tolerance and it is bound to align and it only kicks in in par building and there is not always a par involved

Why does a float caption not build a paragraph? Does a \framed sometimes not build a paragraph?

it's not framed ... framed is advanced in the sense that it can analyze the build paragraph for the natural width float captions look at the size oif the float plus some overshoot ... unrelated mechanisms

...

location has to do with the placement and float as a whole and there tolerance has to deal with snapping

Not float location but \setupfloatcaption location (i.e. top, bottom, ...): relative location of the float caption.

those are locations of the caption relative to the float

...

so ... unrelated tolerances

Please explain.

Thanks

Alan

-- ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl -----------------------------------------------------------------

On 4/12/2018 6:56 PM, Alan Braslau wrote:

On Thu, 12 Apr 2018 18:11:08 +0200 Hans Hagen wrote:

...

On 4/12/2018 5:54 PM, Alan Braslau wrote:

...

On Thu, 12 Apr 2018 10:09:58 +0200 Hans Hagen wrote:

...

tolerance in text/par alignment is just that: adding tolerance and it is bound to align and it only kicks in in par building and there is not always a par involved

Why does a float caption not build a paragraph? Does a \framed sometimes not build a paragraph?

it's not framed ... framed is advanced in the sense that it can analyze the build paragraph for the natural width

float captions look at the size oif the float plus some overshoot ... unrelated mechanisms

Maybe, a float caption *should* be a frame?

it's not wise to change that now

And if it isn't, I am asking why would a caption not be treated as a paragraph?

because tex then tends to make it the width of the whole text so now (by default we measure and decide to go hbox or vbox Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl -----------------------------------------------------------------