[pdftex] SVG Graphics

[Prof Brian Ripley]

On Sat, 11 Jun 2011, Robin Fairbairns wrote:

> John Culleton <john at wexfordpress.com> wrote:
>
>> Getting back to the original topic, we can already insert graphics in a
>> variety of formats. But two of the most useful formats, svg (vector) and
>> TIFF (bitmap) are not included. Is it more difficult to handle these than
>> to handle eps and jpg?
>
> pdftex handles pdf inclusions, and png and jpg.  jpg is obviously a
> start with the myriad formats that tiff encapsulates, but (last i

Sure: one of my cameras (a Canon 1Ds) has 'tiff' as its raw format. 
As no TIFF program can handle it (apart from Canon's) if certainly 
shows up mistaken assumptions.

> looked) png isn't.  basically, tiff _could_ include any bitmap format,
> so allowing it (unrestricted) is inevitably a heavy load.  (note that
> gif was a common format when the pdftex project first started, but it
> was rejected on the grounds that (a) it's an inferior format, and (b) at
> the time a licence was required before you could generate gif.)
>
> original (pre-)pdftex claimed it was able to read tiff graphics.  the
> ability was withdrawn after complaints about its many failures.  i'm
> involved with the project, but i assumed the complaints were about tiffs
> with weird formats embedded.
>
> i don't know anything about svg, beyond the fact that it's a vector
> graphics format.  i presume there's a library to convert it to pdf, as
> would be needed if pdftex is going to include it.  so apart from the
> fact that pdftex is frozen...

There are several of varying qualities.  Use a macro package to 
convert for you ....

>> Color and hyperlinks have been available in pdftex for many years. Their
>> inclusion is not IMO a weakness but a necessary modernization.
>
> colour use is a big deal, but as you rightly say it has been there since
> the beginning.
>
> hyperlinks are a comparably big deal, and covering all possible cases
> has led to a frighteningly huge bunch of macros.
>
> there _is_ a school of thought that claims that anything beyond what
> knuth provided is anathema.  some of your remarks lead one to imagine
> that you're in that camp, so perhaps that suggestion was in the light of
> a mistaken perception.
>
> rovin
>

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595