- Inkscape Devel - Inkscape Mailman

[CAD] Status
by Sebastian Götte 17 Aug '13

17 Aug '13

Hi, This week I mainly squashed some bugs. The current code [0] still requires -std=c++11 in the CXXFLAGS. Connector creation is working properly now, without stray points left over afterwards. Points can be selected in groups which can be transformed (scale, shear, rotate, translate etc.). Simple orthogonal routing and straight routing work. Libavoid routing works as far as there are no objects to be avoided (I did not yet test it with objects to be avoided). Multiple connectors can be connected to the same endpoint. Remaining problems are: * When moving a connector in selection mode ([F1]), sometimes a stray point appears somewhere inside the bounding box of the original connector (strange one!). * Connector aborting does not work yet. I deactivated these 10 lines for now. * There are some open UI questions: - How to disconnect connectors? - How to re-connect connectors to different endpoints? - How to edit SPPoints as parts of groups? How to move them into groups? - Probably some more which I forgot * Libavoid routing is still untested. * Sporadically, I manage to get the UI into some infinite recursion. I am looking into these issues. Best regards, Sebastian [0] https://code.launchpad.net/~h-e-6/inkscape/connector-wip

3 2

[Templates] Report
by Jan Darowski 17 Aug '13

17 Aug '13

Hi! Now all the static templates support is ready and working. I've been working for some time now on understanding extensions. Soon first commits for this part should appear. Slagvi, Jan Darowski

2 1

[CAD] Comments and SVG Connector proposal
by Tavmjong Bah 16 Aug '13

16 Aug '13

Sebastian, Good to see you're making some progress squashing bugs. I've played around a little bit with the code. Here are a few observations: 1. Points (ports) need to be shown on the screen independent of whether or not they are attached to connectors. 2. Creating a connector should use a pre-existing point (port) if the cursor is near a point at the start or create a new point if it is not. The same thing should happen for the end point. 3. A distinction needs to be made between dragging a freely movable start/end point and moving a connector end from one port to another. 4. In your current code, drawing a simple connector creates a start point and an end point at the beginning of creating a connector. At the end, a new end point is created but doesn't seem to be actually used. 5. There needs to be one final redraw at the end of creating a connector. Since getting something out of Doug is like pulling teeth, I've put together a page on what I envision SVG 2 connectors to look like.[1] The key point (no pun intended) is that connection points can be defined using percentage values related to the objects they are part of or to the connector end points themselves. The one problem I have is how to specify a connection point inside a <symbol>. I hope to discuss this at the next SVG Working Group meeting. Your feedback (or anybody else's feedback) is welcome. Tav [1] http://tavmjong.free.fr/SVG/CONNECTORS/index.xhtml

2 1

crash revovry
by younes mesbah 16 Aug '13

16 Aug '13

Hi, I've made a dialog box in GTK,i need to knew from wich file i can call my dialog box (inkscape's code files) with "dt->_dlg_mgr->showDialog("mydialog")".thinks.

1 0

inkscape crash recovery
by younes mesbah 13 Aug '13

13 Aug '13

hey everyone , We want to create a dialog box to revover files in case of crash and we still don't know how to open the crash-recovery box at the start of inkscape . thanks for your help .

2 1

[CAD] Status
by jaseg 13 Aug '13

13 Aug '13

Hi, This week, again I feel I spent way too much time in gdb I moved the code from using SPKnot for handles to ui/tool/ControlPoint. I did this not for dragging (SPKnot has built-in dragging) but for the selection feature of SelectableControlPoint s. Connector creation by dragging or clicking both ends works now, live preview works, saving connectors works. Remaining are a couple of segfaults: One when closing inkscape without saving just after a Connector has been created while the connector tool is still active, and one when I try to drag an endpoint of the connector. This one is actually quite tricky, I have not yet been able to track down its cause properly. Another issue is that when a Connector is created with two clicks (but *not* when it is created via dragging), I have a leftover SPPoint at its start. I think I know where this comes from, but I have not yet found out how to properly delete a SVG object in inkscape without segfaulting. And finally, though I can save SPPoints and SPConnectors just fine to SVG, I can't load them yet. I would appreciate some eyes on the connector creation code in SPConnector.cpp with respect to the SPPoint dragging segfaults. I am not very confident of my work in ref/unrefing things in the connector context. The code can be found at [0]. To compile it, add -std=c++11 to your CPPFLAGS. It still needs that for a little variadic template I wrote. I will change this to a bunch of non-variadic ones for pre-C++11 before integrating. Best regards, Sebastian [0] https://code.launchpad.net/~h-e-6/inkscape/connector-wip

2 2

Gtk+ 3 test-binaries now in PPA
by Alex Valavanis 12 Aug '13

12 Aug '13

Hi All, Daily builds of Inkscape with experimental support for Gtk+ 3 are now available in the trunk PPA. I've named the package "inkscape-trunk-gtk3". So... 1. We now get early warnings about build-failures with Gtk+ 3. However, please remember to do a local test build with Gtk+ 3 support before pushing any major GUI changes to trunk. 2. It's now easy to test the Gtk+ 3 version on Ubuntu without needing to manually build from source Note that there's a bug tag "gtk3" for tracking the (numerous) runtime errors [1]. If you spot any new issues, or know how to fix anything in the tracker then please get involved! Cheers, AV [1] https://bugs.launchpad.net/inkscape/+bugs?field.tag=gtk3

2 1

Boolean ops (again!)
by Eric Greveson 09 Aug '13

09 Aug '13

Hi devs, I'm looking at Boolean operations (particularly Difference ops) on real-world SVG files again, and I've come across another bug, which I've reported here for posterity: https://bugs.launchpad.net/inkscape/+bug/1209281. This comes with a sample SVG file which might reproduce the bug. Note that I say "might" because it is reproduced on the current trunk on my x86 Ubuntu 12.04, but not on the latest Windows build I'm running (0.48.4). I aim to have a look into what's going on here and hopefully come up with a patch - I've looked through the livarot code in a bit of detail now, and I think I understand at a basic level how the various path-intersection algorithms are implemented (particularly wrt quantization of coordinates and the implications for the sweep-line method). Meanwhile, a couple of questions: 1) While I expect I'll be able to fix the current unstable behaviour (occasional copying of topmost path's coords into bottom-most path during difference operations, for too-small paths that "fall through" the quantization grid), the whole requirement for quantization in livarot is a bit unsatisfactory. Longer-term, it might be nice to switch to a path-intersection algorithm without the quantization restriction, and that deals better with degenerate cases - perhaps something like the CGAL 2D intersection methods ( http://www.cgal.org/Manual/beta/doc_html/cgal_manual/packages.html#part_VII… I believe this part of CGAL is GPL licensed, is this an issue? Thoughts? 2) I keep coming across code in Inkscape that could be improved with the RAII pattern (particularly, things like: { Thing *t = new Thing(); ... do stuff with t ... delete t; } When writing new code, I'd much prefer to use smart pointers, e.g. { std::unique_ptr<Thing> t(new Thing()); ... do stuff with t ... } Obviously things like std::unique_ptr and std::shared_ptr require (part of) C++11. I note that parts of inkscape use boost::shared_ptr - is use of boost the way to go in this case, to keep support for older (non-C++11) compilers? Thanks Eric

5 5

Pre-merge checks
by Alex Valavanis 09 Aug '13

09 Aug '13

Hi All, A minor error in yesterday's merge of the (otherwise great) unit-refactoring branch was picked up by the automated testing in the PPA. Basically, a couple of dead files were not removed from the POTFILES.in list. Can I, therefore, suggest that we always run a "make check" on branches before merging them to avoid this happening again. Thanks, AV

3 2

Staging website blueprints up
by Jared Meidal 06 Aug '13

06 Aug '13

All the current webpages are now reflected in blueprints for the staging website: https://blueprints.launchpad.net/inkscape-web I've used the nomenclature of "+" to indicate child pages, and to include the parent page in the title. There are a number of blank pages still existing: *about+overview *about+screenshots (at least not much) *community *community+gallery (limited) *learn *contribute *contribute+contributor-stats *develop (not much more than descriptions and links to the child content) What stands out to be in this review is that _many of the parent pages are blank_. I also understand that this is the staging site which has not been completed. I would prefer we not simply fill a void by adding content, or simply add links to the content on the child pages. Rather, is there a strong reason for these parent pages to exist at all, beside simply adding a category for the necessary child pages? If there is a vision for these pages I would like to see that reflected in our blueprints. Otherwise I think we may see need to consider a revision of the menu structure to keep a simple site and not loose users in the multiplicity of pages. I would love to contribute more in filling content, if suggestions can be made within these blueprints or through the mailing list! Thanks. --Jared

2 4