[cairo] Unit testing? (PDF)

[Ian Britten]

Chris Wilson wrote:

> A true logging surface is another wishlist item. This use-case is
> slightly different than the one I envisaged for the CairoScript backend,
> but it may be close enough to be useful. [The goal of that backend is to
> create a sequence of cairo user calls that recreates the input.]

Sounds ideal!  Consider this one vote for it!  :)

>> Or, is that what cairo-trace is about?  Are there any details around
>> about it?  I'm not familiar with it... (And, is it in 1.8.x?)
> 
> cairo-trace is an LD_PRELOAD interception library that captures all
> calls to cairo and generates a replayed CairoScript file. It is only
> available in the development tree. It's format is a condense,
> human-readable variant of PostScript. 

Hmm... Does that mean it's Linux-only?  Unfortunately, I need an
approach that'll work on Window too...
Or, is LD_PRELOAD just the typical usage, and it's actually just a
library that I could choose to explicitly link in to my unit tests?

> For your purposes, I think an even
> more verbose log of cairo calls would be more useful. libcairowrap
> [http://people.freedesktop.org/~jrmuizel/libcairowrap-0.1.tar.gz] is a
> little old (it may be missing a few calls now) but it may be a suitable

Interesting.  I'm not familiar with either of these Cairo tools, so
I'll have to go off and research them a bit...

Are they both destined to be a maintained part of the mainstream Cairo
product and its releases, or are will they be (unsupported?) add-ons?

Thanks for all the info!
Ian