gnash-dev
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re[2]: [Gnash-dev] Compatible libagg now in Fedora Core 6 updates


From: Udo Giacomozzi
Subject: Re[2]: [Gnash-dev] Compatible libagg now in Fedora Core 6 updates
Date: Tue, 13 Mar 2007 17:20:48 +0100

Hello John,

Monday, March 12, 2007, 11:38:06 PM, you wrote:
JG> CVS Gnash does not build with the agg-2.4-2.1 that's in Fedora Core 6.
JG> Here's the error message:

JG> render_handler_agg.cpp:168:30: error: agg_gradient_lut.h: No such file or 
directory

Unfortunately AGG 2.5 is more the final 2.4 version. As mentioned
before on this list, various 2.4 versions exist and we need many new
features of AGG.


JG> ./configure warns, saying that I should upgrade to agg-2.4, but I'm already
JG> at agg-2.4.


You need to interpret this as the 2.4 source version (the one you can
download now from the AGG website).


JG> Now that agg-2.5 was released (in Oct 2006), configure should
JG> recommend upgrading to 2.5 if that header file is not present.

Agree. Maybe also point out that recent 2.4 version would do as well.


JG> There is a newer agg-2.4-2.2 in Fedora Core Updates that does compile,
JG> though it produces a metric ton of long scary warnings, like this:

[...]

JG> Most of these warnings are not fixed in agg-2.5.  Udo, have you
JG> debugged the warnings to make sure they're AGG's fault rather than
JG> gnash's, and reported them upstream?  I'll append all the ones I see
JG> when building with agg-2.5 below.  (I was unable to find a maintained
JG> agg CVS tree.)

I try to solve all warnings but as you can see the warnings come from
the AGG header files. This does not necessarily exclude that the root
of the "problem" sits in the Gnash source since AGG is based on
templates. However, I solved all warnings could. There are some left,
like AGG mixing integers and enums. I got no response from Maxim (AGG
author) when talking about the problem.

At the other hand, the AGG renderer itself uses heavily templates so
that the compiler emits lots of identical warnings.

Anyway, you can safely ignore these warnings.

Since there apparently are some warnings for render_handler_agg*, I'll
have a closer look at it once I get the opportunity.


Udo





reply via email to

[Prev in Thread] Current Thread [Next in Thread]