New Ticket     Tickets     Wiki     Browse Source     Timeline     Roadmap     Ticket Reports     Search

Ticket #9923 (closed defect: duplicate)

Opened 7 years ago

Last modified 6 years ago

BUG: inkscape-0.44 fails on 10.3 with bad patch

Reported by: lfk@… Owned by: rhwood@…
Priority: Normal Milestone:
Component: ports Version: 1.2
Keywords: Cc:
Port:

Description

The patch in the inkscape-0.44 port for 'platform darwin 7' is malformed. It wouldn't apply even if it had the proper number of lines and hadn't had all the leading whitespace squashed.

$ port version Version: 1.211 $ port info inkscape inkscape 0.44, gnome/inkscape (Variants: darwin_7) http://www.inkscape.org

Inkscape is an open source SVG editor with capabilities similar to Illustrator, CorelDraw, Visio, etc. Supported SVG features include basic shapes, paths, text, alpha blending, transforms, gradients, node editing, svg-to-png export, grouping, and more.

Library Dependencies: libgnomeprintui, boehmgc, gtkmm, gnome-vfs, libxslt, loudmouth Platforms: darwin Maintainers: gnome-darwinports@…

I fixed the patchfile so that it would apply, but the build still failed: object-edit.cpp: In function `void sp_spiral_outer_set(SPItem*, const

NR::Point&, const NR::Point&, unsigned int)':

object-edit.cpp:921: error: `isinf' undeclared (first use this function) object-edit.cpp:921: error: (Each undeclared identifier is reported only once

for each function it appears in.)

make[2]: * [object-edit.o] Error 1 make[1]: * [all-recursive] Error 1 make: * [all] Error 2

There is also a dependency (configure type) on lcms which isn't resolved.

I'll probably be upgrading to Tiger this weekend where I can only assume this builds since there aren't any other relevant bugs.

Attachments

inkscape_build.log (110.0 KB) - added by lfk@… 7 years ago.
Complete build output via 'script'.

Change History

comment:1 Changed 7 years ago by rhwood@…

  • Owner changed from darwinports-bugs@… to rhwood@…

Did you try it with just commenting out the patch?

comment:2 Changed 7 years ago by lfk@…

Yes, I did try without the patch at all and it failed in the same area that the patch is supposed to change. I did it again just this morning so I can paste the messages in here (didn't upgrade to Tiger yet :-).

if g++ -DHAVE_CONFIG_H -I. -I. -I.. -I/opt/local/include/freetype2 -I/opt/local/include -I/opt/local/include/libgnomeprint-2.2 -I/opt/local/include/libart-2.0 -I/opt/local/include/glib-2.0 -I/opt/local/lib/glib-2.0/include -I/opt/local/include/libxml2 -I/opt/local/include -I/opt/local/include/pango-1.0 -I/opt/local/include/libgnomeprintui-2.2 -I/opt/local/include/libgnomecanvas-2.0 -I/opt/local/include/freetype2 -I/opt/local/include/gtk-2.0 -I/opt/local/lib/gtk-2.0/include -I/opt/local/include/atk-1.0 -I/opt/local/include/cairo -D_REENTRANT -DORBIT2=1 -I/opt/local/include/gnome-vfs-2.0 -I/opt/local/lib/gnome-vfs-2.0/include -I/opt/local/include/bonobo-activation-2.0 -I/opt/local/include/libbonobo-2.0 -I/opt/local/include/gconf/2 -I/opt/local/include/glib-2.0 -I/opt/local/lib/glib-2.0/include -I/opt/local/include/orbit-2.0 -I/opt/local/include/loudmouth-1.0 -I/opt/local/include/glib-2.0 -I/opt/local/lib/glib-2.0/include -DXTHREADS -I/opt/local/include -I/opt/local/include/freetype2 -I/usr/X11R6/include -DPOTRACE=\"potrace\" -D_REENTRANT -I/opt/local/include/gdkmm-2.4 -I/opt/local/lib/gdkmm-2.4/include -I/opt/local/include/glibmm-2.4 -I/opt/local/lib/glibmm-2.4/include -I/opt/local/include/pangomm-1.4 -I/opt/local/include/gtk-2.0 -I/opt/local/lib/gtk-2.0/include -I/opt/local/include/sigc++-2.0 -I/opt/local/lib/sigc++-2.0/include -I/opt/local/include/glib-2.0 -I/opt/local/lib/glib-2.0/include -I/opt/local/include/pango-1.0 -I/opt/local/include/cairo -I/opt/local/include/gtkmm-2.4 -I/opt/local/lib/gtkmm-2.4/include -I/opt/local/include/atkmm-1.6 -I/opt/local/include/atk-1.0 -I/opt/local/include/libxml2 -I/opt/local/include -I../cxxtest -L/opt/local/lib -I/opt/local/include -no-cpp-precomp -Wall -W -Wpointer-arith -Wcast-align -Wsign-compare -Woverloaded-virtual -Wswitch -Wno-unused-parameter -g -O2 -MT object-edit.o -MD -MP -MF ".deps/object-edit.Tpo" \

-c -o object-edit.o test -f 'object-edit.cpp' || echo './'object-edit.cpp; \

then mv -f ".deps/object-edit.Tpo" ".deps/object-edit.Po"; \ else rm -f ".deps/object-edit.Tpo"; exit 1; \ fi object-edit.cpp: In function `void sp_spiral_outer_set(SPItem*, const

NR::Point&, const NR::Point&, unsigned int)':

object-edit.cpp:921: error: isfinite' undeclared in namespace std' make[2]: * [object-edit.o] Error 1 make[1]: * [all-recursive] Error 1 make: * [all] Error 2

Error: /opt/local/bin/port: Status 1 encountered during processing.

comment:3 Changed 7 years ago by lfk@…

Additional data: I did build inkscape on an Intel based mac running Tiger and other than the hidden lcms dependency it built without a hitch.

comment:4 Changed 7 years ago by rhwood@…

  • Status changed from new to assigned

Fix is in CVS.

The patch was removed entirely as inkscape had buried in src/isnan.h this line: # define isFinite(_a) (isfinite(_a)) /* MacOSX/Darwin definition < 10.4 */ which should take care of that problem.

If it still refuses to build on Mac OS X 10.3.x, please attach the build log. Thanks.

Changed 7 years ago by lfk@…

Complete build output via 'script'.

comment:5 Changed 7 years ago by lfk@…

It still fails just as I pasted in before. There must be some other hidden dependency to get the isfinite() stuff sorted out. I'm attaching the entire build log as requested (and yes, I did sync and verify that I had the new Portfile).

comment:6 Changed 7 years ago by markd@…

  • Summary changed from inkscape-0.44 port install fails with bad patch to BUG: inkscape-0.44 fails on 10.3 with bad patch

comment:7 Changed 6 years ago by markd@…

  • Status changed from assigned to closed
  • Resolution set to duplicate

This is now a duiplicate of a ticket #11447 filed on inkscape 0.45.

Note: See TracTickets for help on using tickets.