[cairo] cairo-1.6.4 build failing on solaris-10....help needed

sam sirlin samuel.w.sirlin at jpl.nasa.gov
Tue Jul 1 13:07:59 PDT 2008


On Mon, 9 Jun 2008 19:08:22 +0900
"Hung Nguyen Vu" <vuhung16plus+shape at gmail.com> wrote:

> On Mon, Jun 9, 2008 at 4:27 PM, Hung Nguyen Vu
> <vuhung16plus+shape at gmail.com> wrote:
> >
> >
> > On Mon, Jun 9, 2008 at 4:48 PM, Malay HALDER <malay.halder at alcatel-lucent.com> wrote:
> >>
> >> /users/toolsrd/ws/si_ctoolrd/gtk/production/sources/cairo-1.6.4/src/cairo-ps-surface.c: In function '_cairo_ps_surface_emit_header':
> >> /users/toolsrd/ws/si_ctoolrd/gtk/production/sources/cairo-1.6.4/src/cairo-ps-surface.c:110: error: too few arguments to function 'ctime_r'
> >> gmake[3]: *** [libcairo_la-cairo-ps-surface.lo] Error 1
> >
> > If you don't use ps ( PostScript), you can disable it with.
> >
> > make clean; ./configure --enable-ps=no; make
> >
> > Possibly, this is a bug that has not been tested. The funtion ctime_r is different under Solaris[1:.
> > ATM, I don't have Solaris to verify the bug nor fix it it is likely to be.
> >
> > [1] http://www.openldap.org/lists/openldap-bugs/199812/msg00110.html
> Openldap has a fix for this bug like this.
> FYI for anyone interested in a patch.
> http://www.openldap.org/lists/openldap-bugs/199812/msg00105.html

Yes. the manpage for ctime_r on solaris says

char *ctime_r(const time_t *clock, char *buf, int buflen);

so this allows compilation in cairo-ps-surface.c:
                                 /* sws solaris needs len ctime_r
(&now, ctime_\ buf), */
                                 ctime_r (&now, ctime_buf, 25),

probably the hard number should be a #define...


-- 
sam sirlin <samuel.w.sirlin at jpl.nasa.gov>


More information about the cairo mailing list