I just discovered that OpenMP support (used to parallelize Gaussian
blur) is not functional on Linux and OS X because of a missing
AC_DEFINE. Here is a patch that fixes this problem. Admins: please
apply to 0.48 branch also.
We are still using the internal Poppler API to import PDFs. This is
very bad, since the internal Poppler API is not stable, and will keep
breaking. (We already don't build with poppler > 0.12.1)
How about using the public Glib API to render to a Cairo SVG surface
instead? This is how pdf2svg works, and to be honest I had to use that
utility several times to augment Inkscape, because it handles text
much better. What are the advantages to using internal Poppler API
Chris Lilley wrote:
> On Saturday, January 16, 2010, 10:32:16 PM, knowone wrote:
> k> How can I save the file as PNG with the drop shadow intact?
> Make sure you are exporting as PNG. Do not 'Save As' Cairo PNG.
> (I came across this a couple of days ago, and the symptom was that blur filters were not being applied. Also, the pixel dimensions of the two methods differ, if the graphic bleeds over the edge of the 'page').
Can someone tell me why we have a non-vector format in the save as list?
One that seems redundant and apparently with less functionality than the
normal export. Is this for testing purposes?
Selector tool has a new mouse cursor (arrow with an open hand) for
when your mouse is over a selectable object, and another (arrow with
clinched hand) for when you're dragging an object. This improves
precision of selection and UI consistency (previously, the mouse
cursor over a selectable object was different across platforms, e.g.
hand icon on Linux or four-way arrow on Windows).
Inkscape. Draw Freely.
I just built the inkscape trunk from bzr and ran this script
bus = dbus.SessionBus()
inkdoc1 = bus.get_object('org.inkscape', '/org/inkscape/desktop_0')
doc1 = dbus.Interface(inkdoc1, dbus_interface="org.inkscape.document")
im = doc1.image(100 ,100, '/home/ktenney/Firefox_wallpaper.png')
Which worked on a dbus branch a while back.
Now it starts inkscape, but does not place the image
- was the dbus code merged?
- if so, how do I enable/use it?
- if not, where is dbus capable code?
Because my GSoC student had no progress at the midterm evaluation
deadline, I decided to fail him for GSoC. But since I was screaming so
loud at LGM that this summer we would have a PowerStroke LPE, I couldn't
escape coding it myself. Too bad, I should have applied for GSoC as a
For those interested: I think the PowerStroke LPE is ready for some
playing and testing. You have to build Inkscape 0.48+devel with
experimental LPEs enabled. 
Cusp nodes behave funny. To see clearly what happens, you have to apply
a stroke to the path, and no fill.
I do not yet know whether I am going to take the effort to properly
handle cusp nodes... ;)
I know it is buggy, so please only mail *ideas* for improvement, etc.
 You have to build Inkscape with "#define LPE_ENABLE_TEST_EFFECTS",
otherwise the LPE won't show in the LPE dialog. (uncommenting line 9 in
src/live_effects/effect.cpp will do the trick)
Thanks to Jasper van de Gronde and Jon Cruz for their answers on :
This weekend, I made some *GTKmm code* in order to design the NFT Dialog.
The window is fully functional, but I need more informations about how can I
read informations from an SVG file, how to generate a preview image and a
lot of other stuffs.
I give you my dirty code.*
Please affect me a tutor*.
See ya ! :-)
On Thu, Jul 8, 2010 at 6:14 PM, bulia byak <buliabyak@...400...> wrote:
> 2010/7/8 Nicolas Dufour <nicoduf@...48...>:
> > 1. Removing the Uniconvertor scripts from the devlibs. UniConvertor
> import or export scripts warn users to install UniConvertor manually (and
> give the site address) when the libs are missing. (Note to self: fix the
> gettext import bug in uniconv_output.py...).
> I think that will be perceived as a regression. Lots of formats that
> used to work will suddenly stop working. We had a lot of pain with
> displaying messages like this - most people simply perceive any such
> message as an error and complain or give up without reading. Those who
> read it and act on it will be a minority.
>From other side a lot of win32 users ask us "Why you announce XXX feature
but i have not this in Inkscape?".
We cannot synchronize our project development cycles, it's normally. But we
could provide for user a way updating UniConvertor.
External UniConvertor installation is a best solution because current bundle
updating is a task rather for programmer than for
> What's such a big problem with including a new UC into devlibs?
Actually there are not impossible tasks. It's just a subject of additional
development time and resulted bundle size. Latest version
has several new native extensions and more complex structure. And in next
versions we will add new native resources because users
ask for more and more complex features.
As a simple solution we can provide for users an
Inkscape-0.48-with-UniConvertor.zip archive with Inkscape installer,
and small REAME file inside. To simplify UniConvertor download the best way
is not a message "You need downloading from bla-bla-bla site..."
but just a button "Download" which opened browser on required page. Last
option is for those users who download Inkscape exe installer only.
Another valuable reason is planned preferences for translation. We are going
provide a lot of translation options (we already have the code
but we still not expose it as a command line options) and naturally such
preferences could be managed better from GUI than command line.
So we will add preferences dialogs for each format into UniConvertor. It's
not an easy task so for Inkscape team will be better utilizing ready
dialogs from UniConvertor than creating the same but on Inkscape side. And
as I have wrote UniConvertor translation progress dialog
would be also a good additional feature for Inkscape.
It seems like a decent number of tweets and complaints I've seen around
the community (understandably) are about inkscape crashing.
Doing basic searches for bugs tagged w/ crash gives the following info:
All statuses is over 200 bugs and Confirmed or In Progress have 84.
84 crash bugs that can be reproduced seems unreasonably high for
something intended for production work and code-wise there's a chance
that some are caused by shared underlying issues.
If anyone is interested in taking a look into these to see if there are
any you might be able to fix, that would be awesome.