Smooth transitions in Bézier paths
by Diederik van Lierop
I hope you don't mind if I increase the traffic on inkscape-devel even
At some point an idea popped up to only snap to nodes at discontinuities
in paths, and not to nodes at smooth transitions from one Bézier segment
to another. This is more intuitive (less "Hey, what it is Inkscape
snapping to now?") and might increase performance.
Problem is that in the object-snapper I only have access to SPCurve,
which is a wrapper for "old-style" NArtBpath. It's easy to find out
whether a point is inbetween two Bézier segments, but how do I find out
if a transition is smooth? I guess that should be done by comparing the
control points, or is there an easier way? With the transition to
lib2geom, will the NArtBpath be deprecated in favor of the Path class?
(that would make this rather exercise useless in the long run).
Any pointers would be appreciated here,
14 years, 6 months
question regarding wishblade robocraft support...
Sorry for posting here if the user list if better. And sorry about the
subject if it has been talked about before. I couldn't get sourceforge
to add me to the inkscape user mail list and sourceforge was timing out
on all my archive searches with no results.
I remember reading that there was a patch or something to inkscape to
support wishblade and robocraft users. Has that been integrated into
the inkscape application? Is that what the DXF is all about?
Wishblade users can not use DXF directly (needing the robocraft software
to import DXF and change it to some sort of native (and maybe
proprietary) GSD). Is there a way to get inkscape to save images in GSD
in order to simplify the process?
Ideally, I would prefer to skip bringing up the wishblade application.
In my set up, I do all my work in linux (GIMP/INKSCAPE) and only the
plotting in Windows. If I could do everything in linux it would
simplify the flow of work greatly. Has anyone figured out how the
craftrobo is communicated to in order to use it with common open source
linux plotting software?
Better yet, has anyone figured out how the wishblade has been crippled
in order to use it with common open source linux plotting software?
14 years, 9 months
inkscape git mirror?
I'm interested in setting up a git mirror of Inkscape SVN; who's the person in charge of our server currently? We'll need to have a fairly current version of git, libsvn, and the libsvn perl bindings installed to do this.
14 years, 11 months
How to change notification settings in Launchpad?
by Maximilian Albert
sorry if this is a dumb question but how can I change my Launchpad
settings so that I do not receive an email each time someone works on
one of the bugs? Is this the default behaviour for members of the
"Inkscape Bug Team" group? Currently I am sent a notification for each
and every change, but I'd prefer not to get my inbox flooded and rather
browse the list myself when appropriate. It may be just my infamiliarity
with Lauchpad (and lack of time for detailed investigation), but I
wasn't able to find out how turn this off. Thanks in advance.
15 years, 3 months
New win32 lib bundle with Dwarf2
by Bob Jamison
I just uploaded a new lib bundle on the server.
It has been over a year since I last updated the C++ libs.
I also have wanted for months to move from the existing
setjump/longjump (sjlj) exception handling to the new
(to MinGW) Dwarf2 EH. Gcc has had Dwarf2 for years, but
only recently has MinGW started providing it. It should
allow better backtraces and slightly smaller binaries.
I made a build with it manually this morning:
I assembled some MinGW binaries to make this working compiler:
...and here is the new bundle.
Note that the C++ compiler, library bundle, and Inkscape build
must all be either sjlj or Dwarf2. You can't mix them, else
there will be a linking error.
I also added a version of build.xml to the trunk to handle this,
build-dw2.xml. This should keep from breaking everyone else's
build. To use it:
./btool -f build-dw2.xml
Of course there is a corresponding cross-compiler:
...which made this build, which looks ok, but I haven't run yet:
It looks larger, but none of the DLLs have been stripped yet.
From now on, all of the "autobuilds" there will be Dwarf2.
Probably in a couple of weeks when everyone catches up, it will become
15 years, 3 months
by Alexandre Prokoudine
Here are several more notes on current implementation of raster effects.
1. Terminology and hints. In some cases terminology used in effects
coming from ImageMagick is not quite obvious. For instance, you might
want to reread a course on math to understand what "Sigma" stands for
in "Sharpen"/"Unsharp Mask" effects ;-) We might either rename some of
the labels (not good for experienced ImageMagick users) or provide
hints to explain the meaning of each option.
2. Raster effects are not the only ones to suffer from it, but here
goes: effect dialogs inherit caption from menu item name. In several
cases the caption is too long and is shrinked. This is not quite
visually appealing. Two solutions comes to mind: 1) automatically
resize dialog to match caption's length and 2) provide a separate
string to dialog's caption which translators could try to make
3. Cancelling isn't working reliably. I run "Reduce noise" on a
2200x1500 picture and when I click "Cancel" nothing really happens.
4.Should we probbaly implement a "Resize" effect too?
15 years, 3 months
Re: [Inkscape-devel] pdf-input.cpp:544: error: '_preview_output_dev' was not declared in this scope
by Joshua Facemyer / Impressus Art
Jon A. Cruz wrote:
> On Oct 20, 2007, at 6:49 PM, Joshua Facemyer / Impressus Art wrote:
>> Okay, will the same pdf functionality be present then? (Also, is
>> this a workaround to get the thing compiling at a loss of
>> functionality, or is it something that should / will be addressed in
>> a future update to make it compile in an environment that it should
>> without difficulty?)
> Someone familiar with the specifics would have to look into it, but I
> think it's a work-around in losing a bit of functionality...
> But from what I recall, it's probably due to us using internals of
> poppler that are not part of the public API. Get and build poppler
> yourself and you might have the needed .h, but use standard packing
> and you probably won't.
> Part of me only seeing this recently was that only recently had I
> gotten a new enough poppler in macports.
I use Gentoo, so my poppler should be "my own" build. Same with Jeff.
(Had you figured this out yet, Jeff?)
Is it maybe that some switch needs to be passed to poppler when building it?
15 years, 4 months
BUG HUNT - Dec 29 Scoreboard
by Bryce Harrington
Over a quarter of the way to our goal!
Pts BugID Assignee Description
3 171593 Diederik Snapping menu items and shortcuts
3 168912 Johan Engelen axonometric grid angle changes with...
12 168588 Johan Engelen assertion failed, file sp-conn-end.cpp
3 171783 Bulia Byak rclk on palette should change outline color
3 171710 JonCruz Sum color dragging color inother
3 171292 --- Transparency reduced pdf export
9 168865 Johan Engelen Crashes applying "Path along path" effect...
3 171803 Johan Engelen Angled Guidelines!
12 168570 Johan Engelen Crash when recreating grid using 'redo'
9 178004 Max Albert assert when saving as plain svg
3 170374 --- perspective/warp transformation envelope
6 168574 --- Grid setup truncate significant figures
12 178803 theAdib Save As... -> filetype "PDF via Cairo" crashes
9 178312 Diederik Global 'Enable snapping' option non-functional
9 177891 Johan Engelen feGaussianBlur crashes when setting fill paint
9 174475 Niko Kiirala feImage renderer crashes on non-external image
12 167416 Diederik Shrinking a pattern causes lock-up
9 172778 Johan Engelen Grid settings not respected when reopening doc
6 167640 Bryce Escaped HTML chars - tooltips show incorrectly
135 Goal: 500
Note: Points are not given for bugs fixed as invalid, dupe, or
worksforme, and only for issues that require code changes to fix (not
website, translation, or documentation changes).
Note: Scoring for bugs is based on their Importance: Crit=12, High=9,
Med=6, Low=3, Undecided=3, Wishlist=3.
15 years, 4 months
Undo looses selection (in Tweak tool)
One other glitch which the tweak tool revealed: undoing looses
draw a circle, switch to tweak, tweak, CTRL-Z, tweak again = you
can't because the circle is not selected anymore.
draw a circle, draw another, select all, switch to tweak, tweak
circle nb 1, CTRL-Z, switch to select = only circle nb 2 is selected
In fact this happens in all tools:
draw a circle [-> it is selected], draw another [-> the second one
gets selected], CTRL-Z = nothing is selected anymore
This is not particularly annoying in all other tools but it is in the
tweak tool, because undoing the first tweak deselects and an
additional tool switch is necessary before one can tweak again. I
already noticed that when the too was in its embryonic stage in the
calligraphy tool... but failed to report it :(
It would be nice if:
- the selection could be kept in Undo history. this would solve the
issue for all tools and in addition I found myself bitten by the lack
of selection history several times. Ex: you try to select several
small objects, it takes a while, you add another object by mistake,
the only way to subtract it from the selection is to click on it
again... which cool get messy if there are already several other
objects selected, or if you added it via a select under or in group
- there was a way to select objects directly from the tweak tool.
this would, in addition, allow to avoid constant switches between the
select tool and the tweak tool (not that s and and w are far apart on
the keyboard but still). ALT is not already used as a modifier in the
Tweak tool ;) Of course the selection behavior won't be as powerful
as the select tool (since the select tool modifiers is already very
complex and its modifiers are all full) but just ALT+click/drag =
selects and SHIFT+ALT+click/drag = adds to selection would be nice.
15 years, 4 months