Super cool, already tested a few scenarios - works! Merci Bastien!
Oct 2 2019
Dec 14 2016
Dec 13 2016
Nov 7 2015
Oct 20 2015
Hey Bastien - so quick, thank you so much for the effort!
All these different properties were quite a hassle to work with anyway, so them going away is a good prospect in two senses even :)
Oct 17 2015
Oct 14 2014
I finally found time to experiment with a first use-case: Building an interactive hoverable svg for web use, as directly as possible, and in this case overlayed above a rasterized freestyle render.
Oct 5 2014
Perfect - builds fine now; Gave the exporter a super quick shot today too - magic! :)
Thanks a ton! I'll let you know my findings when I start experimenting with this in more detail in the next days!
Oct 4 2014
First of all I'm super happy to see this happening, wasn't aware how far this has progressed already; Thanks so much! :)
So ... I'll try this out on a project I'm currently working on and provide you with some real production feedback! ... But I got a minor build problem first ;)
Jun 29 2014
Thanks for looking into it!
May 12 2014
Mar 6 2014
Oh whoops, that explains a lot ... Still I don't see why the axis lock is discarded on pressing 'c', but for the time being (RC bugfixing and all) this is probably neglectable. Thanks for the quick response!
Jan 19 2014
Fixed in D212
Jan 14 2014
Jan 13 2014
Here's two little python snippets for triggering the issue directly from the console:
So the error occurs when the importer generates the geometry inside blender. (That is, after parsing, which seems to work fine)
Inside class SVGGeometryPATH an object (curve) is created, a bezier spline is added to it, then the script starts adding bezier points, copying over the coordinates from the parsed data.
When printing the spline's bezier_points array length it becomes apparent that the error occurs right when an overflow happens:
Thought i'd run the file through http://validator.w3.org ... and it validates.
That makes it less likely an Inkscape related problem, i guess.
Back to reading code ...
I'll look a bit into it, feel free to beat me to it in case you're familiar with the code and able to provide a fast fix.
Especially let me know if you think my svg file is the actual problem here ... Not that I created anything fancy there, but you never know ... :)
Jan 11 2014
@Paweł Łyczkowski (plyczkowski) Pawel, if you somehow got the impression that my approach to this was to bluntly discard all existing material in order to redo everything from scratch, that's not the case. I did experiment exactly with those things you mention here! And it just so happened that in my process I eventually found this approach to be unrewarding and non-conclusive for me. Which is not to say that it can't work at all! But it's simply not where I arrived at. If you collectively decide that you want a different solution than any of those I propose, then that's totally fine, but please don't ask me to work in some way you would want me work, because that's not going to lead anywhere. I like Andrzej's work, I think his icons are great, but I am me, and that's how my work will look like, so please treat it as such, ok? :)
@Paweł Łyczkowski (plyczkowski)
Here is a draft for your suggestions:
Went over them again to see how much further I could push this approach, here's the updated proposal :)
Jan 10 2014
I absolutely agree on the "wheel rolling forward" problem now that you mention it; for this I would propose either of these two tweaks as a fix:
Phew! ... Communication with 16x16 pixels is brutally challenging :), but I'm happy to present you with another proposal:
Jan 8 2014
@Paweł Łyczkowski (plyczkowski) Reading that i just thought of blender with all icons animated ... Imagine that! ... so. trippy. :)
Jan 7 2014
Ok if the arc from the ogl logo is unrecognizable or even unknown that clearly excludes it from being an option ... so that's that. :)
Thanks for the feedback!
Jan 6 2014
Jan 4 2014
@William Reynish (billrey) Oh cool that's good news, must have missed out on that, will try that in a second :) thanks!
I might have unfortunately discovered a problem about this proposal: If it works the way i understand it ...
Jan 3 2014
I'd find this very useful in these occasional "Whoops how do I navigate in 3D without a middle mouse button?" situations.
Not really a professional concern in that sense, but it's happening now and then i guess. (Can't say if it outweighs the visual clutter argument though)
Maybe a placement near the "other viewport clutter" (axis gimbal, object name) would be more subtle?
Definitely something I could have benefitted from in the past, I'll look into it!
Jan 2 2014
Dec 31 2013
Just a quick interjection from my side: @Paulo José Oliveira Amaro (pauloup)'s suggestion to update the lists to the new list style could possibly already be happening in the course of my work on T37881 - File Browser renaming of bookmarks. I'm still mostly reading through the code and assessing the situation, but so far my understanding is that the lists in the file browser are all coded in pure C and all fancy, new lists that allow renaming and such utilize the bpy.types.UIList python template. Thus a rewrite of the file browser sidebar layout in python might be a necessary and/or favorable decision for the sake of a good and DRY implementation.
Dec 21 2013
I'm on it!
I added yet another revision, some things about this were still bothering me ... see the comments for the revision!
Dec 20 2013
New diff is up, thanks for the hints, that helped a great deal!!
Dec 19 2013
I'm kinda stuck implementing the toggle - i could not yet find any way to modify the icon via C, neither could I find a way to change the operator name on exec or something (although that's probably a bad idea anyway because of possible messup with localization? still need to read up on how localization is implemented actually, maybe I'm wrong :)) EDIT: Did my reading up on the internationalization thing: If I understand it correctly, changing the name on the fly should not pose a problem regarding internationalization. Correct me in case I'm wrong here!
Dec 18 2013
Ok sure, performance matters as well! :)
Just out of dev curiosity: When are these poll functions run? Timer based or callback based? (on operation invoke/exec something?)
Fully agree with the single menu item toggle solution!
Ok I looked into it - here's my current implementation (for @Brecht Van Lommel (brecht)'s feature suggestion):
Dec 17 2013
cool, i'm on it!
Dec 16 2013
Oh i see, well then your suggestion sounds good to me, and @Scott Petrovic (scottpetrovic)'s does as well (as an additional feature to have) ... If you think this thing is realistic to tackle for a newcomer in blender core development, I'd be willing to give it a shot! (after proposing the exact plan for the changes and having them approved, of course)
I'm probably going to feel really stupid in a moment, but where can you actually enable auto-pack in the first place?
I looked all over the place in blender itself, but neither that, nor googleing and searching in the wiki didn't return anything for me.
One last thing, so obvious that I even forgot to mention it:
As a matter of fact we already have such a convention widely used inside the Node Editor: "Fac"
So in that sense we can also look for and pick up such already existing conventions,
and rethink if they could be applied at a wider scale inside the application, transgressing the confined borders of the Node editor.
As a user as well as a developer I very much agree on all the points @Brecht Van Lommel (brecht) brought up against introducing percentages instead of 0-1 ranges.
Consequently, for the sake of consistency I think that it would be desirable to adhere to this 0-1 standard as much as possible, and only fall back to alternatives in places where it is absolutely the only logical choice to use something other than 0-1.
Dec 14 2013
Cool thanks, learned a lot here. On to new tasks ... :)
Dec 13 2013
I just uploaded the diff and attached it to this task, hope I did this the correct way too.
thanks for the feedback, i just tried implementing the simplified approach @Campbell Barton (campbellbarton) suggested, and it works perfectly! (also got rid of the breakage on consecutive runs this way)