GNOME Bugzilla – Bug 436148
Dia mis-handles xfig arcs
Last modified: 2011-03-30 20:42:40 UTC
a Debian user reported the following:
In the attached fig file (which was produced by xfig 3.2.5 and then edited
by deleting the second line, "Portrait"), dia puts the arcs on the wrong sides.
In xfig, the arc goes from "1", past "2", to "3",
while in dia, the ends of the arc are (more or less correctly)
near "1" and "3", but the middle is far away from "2".
I could reproduce this. Does the mentioned manual preparation of fig files (and dia complaining about the xfig import file contents anyway) imply a general xfig import problem?
I'm not sure if this is somehow related to #59941.
Created attachment 87610 [details]
Mentioned fig example file
I also get problems on opening other unmodified xfig files as well, see new attachment.
(Also reported as a Debian bug at http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=405833)
Created attachment 87612 [details]
Second example of fig import problem
*** Bug 436350 has been marked as a duplicate of this bug. ***
Bug reproduced. Indeed, there are a number of problems with Fig import, which I fortunately will have some time to work on at work soon.
can be reproduced with render-test.dia from the Dia distribution as well ...
Seems as if this is completely fixed in SVN now:
2008-05-04 Hans Breuer <email@example.com>
* plug-ins/xfig/xfig-import.c(fig_read_arc): acccept pie-wedge
although not completely supported. Better than refusing the whole file
(fig_read_text) : fixed probable typo 3.54/72.0
* lib/geometry.[hc] lib/libdia.def: three_point_circle() calculate
center point and radius given 3 points on a circle, fixed prototype and
implementation to ...
* plug-ins/xfig/xfig-export.c : corrected arc export for both draw_arc
and fill_arc methods
* plug-ins/xfig/xfig-import.c lib/create.h(create_standard_arc) : the
fifth parameter was named and treated as radius alsthough it set the
* lib/create.c(create_standard_arc) : also the second point passed in
was simply ignored. Now it moves the second handle. Part of bug #436148
unfortunately, dia still produces errors with my "Second example".
Thanks for considering,
I've just pushed a fix to master which makes Dia read the file without complaints.
But I fail to correlate the files content with this bugs title. It does not contain an arc, but a ployline. Also it seems to be generally broken - maybe because it is created by an alpha version of XFig?
From my understanding every compound has to be closed by -6, so the number of 6 and -6 at lines beginning should match. See:
Declaring as fixed. Please provide a non-alpha-version created file, if you still are not satisfied ;)
*** Bug 646262 has been marked as a duplicate of this bug. ***