Hi all,
I'm trying to include images in a PDF document being generated using
XSL:FO and XEP 3.7.5. The difficulty I'm having is because I don't know
the image format or size before doing the render, so it has to be done
automatically in the XSL-FO pass. Images can be anywhere from 100x100px
to 1000x1000px.
XEP handles the format fine (for the most part), but scaling is
unpredictable, even (or particularly) using width="auto" height="auto"
content-width="auto" content-height="auto". What I am trying to do is
scale down large images to fit within the width of my page, and keep
small images their original size (or a reasonable size) - strangely I
seem to be having more problems with small images scaling too large
rather than vice-versa.
This seems to be a case for using the scale-to-fit value
(http://xep.xattic.com/xep/testsuite/usecases/image-scale.pdf), but my
difficulty is that while I know the absolute width the area I want the
image to appear (19cm), I know neither the height (since this should be
dependent on the scaled image), and I'm having problems with the image
overflowing outside of the block.
The pertinent part of the FO for my example is:
(full XSL-FO available, it's just very large)
<fo:block space-after.optimum="6pt" space-before.optimum="6pt" span="all">
<fo:table width="100%" table-layout="fixed">
<fo:table-column/>
<fo:table-body>
<fo:table-row keep-with-next="always">
<fo:table-cell>
<fo:block id="figure1" span="all" font-size="8.5pt">
<fo:inline font-weight="bold">Figure 1: </fo:inline>Simple EAV schema
for a clinical database.</fo:block>
</fo:table-cell>
</fo:table-row>
<fo:table-row keep-with-previous="always">
<fo:table-cell>
<fo:block text-align="center" span="all">
<fo:external-graphic
src="url(file:y:/jmir/jmir-cocoon/content/v5/v5e27/jmir_v5e27_fig1.jpg)"
content-width="auto" content-height="auto" width="auto" height="auto"/>
</fo:block>
</fo:table-cell>
</fo:table-row>
<fo:table-row keep-with-previous="always">
<fo:table-cell/>
</fo:table-row>
</fo:table-body>
</fo:table>
</fo:block>
I have the image and caption in a blind table to keep them together
across page-breaks (yes, I know this is ugly but it was the best way I
could ensure consistent layout). Should I be using a block-container
instead/as well?
Full XML/PDF sources at:
http://ehealthresearch.ca/cocoon/jmir/v5e27.xml
http://ehealthresearch.ca/cocoon/jmir/v5e27.pdf
Any help is gratefully appreciated!
MJ
References:
http://xep.xattic.com/xep/testsuite/usecasestest.html
http://xep.xattic.com/xep/doc/tutorial.html#Graphics
http://xep.xattic.com/xep/doc/spec.html#Graphic_Formats