[postgis-users] Problems Compiling GEOS 3.0

Bill Thoen bthoen at gisnet.com
Thu Jun 19 08:36:01 PDT 2008


Excellent! Thanks, that solved my problem. Apparently gcc ver 4.3.0 that 
comes with FC 9 has "improved" header files, so now developers are going 
to have to explicitly include header files that used to be part of other 
header files. This is going to cause a lot of grief for a while I bet!

- Bill Thoen

Luigi Castro Cardeles wrote:
> hi,
>
> here is the solution. i found that at gentoo maillist.
> https://bugs.gentoo.org/attachment.cgi?id=149838&action=view 
> <https://bugs.gentoo.org/attachment.cgi?id=149838&action=view>
>
> best regards,
>
> 2008/6/18 Frank Warmerdam <warmerdam at pobox.com 
> <mailto:warmerdam at pobox.com>>:
>
>     Bill Thoen wrote:
>
>         This probably is the wrong list to whinge about this, but I
>         don't know if there is a geos-users list. Please point me in
>         the right direction if so.
>
>         Anyway, I'm having lots of problems running make on GEOS 3.0.
>         Errors saying "'DoubleBits.cpp memcpy' was not declared in
>         this scope. and the same for strtod in StringTokenizer.cpp. It
>         looks like there are some #include statements missing or I
>         have a non-standard set of *.h files (I'm using Fedora Core 9,
>         Linux).
>
>         Anyone else have problems like this and how do I correct it?
>
>
>     Bill,
>
>     I presume this is an issue with a very new gcc version that is
>     stricter
>     about something.  I've heard the gcc team is "improving" their include
>     files to avoid including stuff that isn't required so perhaps
>     memcpy()'s
>     include file needs to be pulled in deliberately now.
>
>     There is a geos mailing list, and trac at:
>
>      http://trac.osgeo.org/geos/
>
>     I would suggest you file a ticket, including information on the gcc
>     version, and a capture of the build errors. You also might want to try
>     building from "svn trunk" to see if this problem has already been
>     dealt
>     with.
>




More information about the postgis-users mailing list