- Inkscape Devel - Inkscape Mailman

Drawing algorithm
by Tom Moore 05 May '08

05 May '08

Hi to everyone in the development mailing list - this is my first post, so sorry if I'm not asking the right people. I use Inkscape on a regular basis and have always been really impressed with the freehand drawing tool, specifically the efficient resulting curve (very few points to describe the curve). Does anyone know if it is viable to write a similarly efficient algorithm in Flash using ActionScript 3? I've given it a go on a number of occasions over the past 6 months to decent effect, but not half as impressive as Inkscape. Because ActionScript and Flash are pretty slow compared to C, I tend to use fairly simple algorithms to draw curved lines through points. I can only assume that the Inkscape algorithm is pretty sophisticated and I can't begin to try and work out how it was written from the svc, having had no experience with the language. Has there ever been any documentation, high-level schema or flowchart produced to explain the steps taken to calculate the optimised curve? I'm sure that someone must have produced some pseudo code or white paper to describe the best practice for a vector line drawing tool. If anyone has any ideas I'd be really grateful. Tom

1 0

Re: [Inkscape-devel] File import
by Tavmjong Bah 05 May '08

05 May '08

On Mon, 2008-05-05 at 12:21 +0200, momo@...1386... wrote: > Here are TGA test files: > > There are 16,24 and 32 bits with and without RLE compression (details are in > the file names). All generated by photoshop. > > molumen Thanks again! The 16 bit files don't open while the 24 and 32 bits files do. A good predictor is if an image is shown in the preview window. I take it the same GDK code is used for both importing and generating the preview. Tav PS. I am testing with v0.46 with Fedora 7.

1 0

Re: [Inkscape-devel] Gobject or C++???
by SF Markus Elfring 05 May '08

05 May '08

> > What itŽs the best method to new GTK software?? > > In most cases, avoid C and C++ as much as is practical. Use a higher-level > language with GTK bindings and use the object model of the language you choose > rather than GObject. You may still need to dip into C occasionally for special > functionality, but minimize your use of C or C++. Neither are good languages > for application development. I find this wording confusing. Each programming language has got its strengths and weaknesses. I assume that you prefer scripting languages. There are more design options available. It is not clear what "the best" will mean. What are the main factors for qualification of development results? ;-) Regards, Markus

2 1

Newest go-ooo.org has built in SVG support
by rygle 04 May '08

04 May '08

Don't know if you know about the community built version of OpenOffice.org, but it's at http://go-ooo.org. It boots quicker than the standard OOo, and has a few other great enhancements, like apparently now the ability to import SVG's directly. Still haven't tested this, as it's a 130Mb download, but the previous go-ooo.org version was noticeably quicker than the default on my computer, and a little more stable too I think. Any comments? Cheers, Rygle. -- View this message in context: http://www.nabble.com/Newest-go-ooo.org-has-built-in-SVG-support-tp17034415… Sent from the Inkscape - Dev mailing list archive at Nabble.com.

2 2

compile fails, linux, rev 18505:
by bulia byak 03 May '08

03 May '08

make[2]: Entering directory `/home/d/ink/inkscape/src' make[2]: *** No rule to make target `dom/cssparser.cpp', needed by `dom/cssparser.o'. Stop. please fix :) -- bulia byak Inkscape. Draw Freely. http://www.inkscape.org

2 1

Batch export feature idea/request.
by Brian Vanderburg II 02 May '08

02 May '08

I have a set of several objects I'd like to batch export. I export them all to a single directory. To have multiple color sets I have a few key items and all of the objects are updated (mostly clones/etc), but then the problem is that batch export wants to write over the previous files so I must manually specify new files: directory layout: items.svg red/item1.png red/item2.png red/item3.png blue/item1.png blue/item2.png blue/item3.png etc With batch exports, after I save the red items and then change the key items to me the stuff blue, batch export still wants to save as red/item?.png A simple solution is in the batch export dialog provide an additional option 'Batch export to directory: '. When enabled, any existing export directory information would be ignored and only the filename portion of the exported file would be used, or if not available the object id. Then I can make the items red, batch export ,select 'Put all files in dir' and choose 'red' for the directory and export. Then change the key items to make it all blue, batch export but choose 'blue' as the export directory, etc for other colors. Something like the attached image Brian Vanderburg II

1 0

Re: [Inkscape-devel] Buildtool -- new
by Bob Jamison 02 May '08

02 May '08

When you say "btool something" that something is one of the targets in build.xml. The project has a default target listed at the top (dist). This is the one that will be executed, after all of its dependent targets, unless you specify another one. You can pick any target to execute. So you can say: btool compile ...and only run a compile, with no linking or copying. I guess you could add a pair of targets, something like this: <target name="refresh_prepare" description="delete the build.dep file to force a regeneration"> <delete file="build.dep"/> </target> <target name="refresh" depends="refresh_prepare,compile" description="force the cc task to regenerate its cache">  </target> The second task doesn't need to do anything. It just says: "before me, run refresh_prepare, and then compile" I'm not sure how to work that into the scheme of the rest of the target hierarchy, though. :) bob -----Original Message----- >Nice, > >Can you alias that, so that -Drefresh=true is just refresh, so you can >simply go 'btool refresh'? >

1 0

Re: [Inkscape-devel] Buildtool -- new
by Bob Jamison 02 May '08

02 May '08

Just a reminder, for when you do an "svn update". If, in the left column, in addition to U and G, you see any A or D for added or deleted files, then you will need to regenerate the dependency list. Also, at link time, if you see errors about missing symbols, then you probably have too many or too few .o files in the /build/obj directory. This is another clue that you need to refresh the dependency cache. The "clean" target has a task for deleting build.dep, so "btool clean" always works, if all else fails and you need to take drastic measures. Another trick: if you need to force a rebuild of a certain set of files, but don't want to rebuild the whole tree, then look in /build/obj. The directory structure mirrors /src. Just delete the corresponding .o's and run btool again. bob

1 0

Buildtool -- new "refreshCache" flag
by Bob Jamison 02 May '08

02 May '08

After thinking about telling Rygle about deleting build.dep to force a recomp, it occurred to me how kludgey that is. So I added yet another flag to the <cc> task, refreshCache="true/false". Setting this to true will force a regeneration of the cache. But you say "Hey, I don't want to edit build.xml every time I want to clean the cache!" Fair enough. So I added another property to build.xml, "refresh=false". This value is used in the <cc> task as refreshCache="${refresh}" To change this, and force a refresh, merely redefine "refresh" on the command line: btool -Drefresh=true You should see this near the top of the output: overloading property 'refresh' = 'true' This is a workaround to the problem: how to specify a task-specific parameter on the command line. I need to figure this one out eventually. By the way, did you know that you can do this? Calling btool this way: btool -Ddist=c:/inkscape ...sends the completed Inkscape distro directly to c:\inkscape :) Since btool uses URIs, "\" or "/" doesn't matter. Let me know if you have any problems. But this is a good example of how btool can be modified, and new Tasks added, to perform whatever build functionality we need. bob

2 1

Build problems - Last two commits (18500 & 18501)
by rygle 02 May '08

02 May '08

Hi guys, Having problems building the last two trunk revisions on Win32; SVN 18500 gives me this; ===== Make error line 201: problem compiling: mingw32-g++: src/dom/svgparser.cpp: No s uch file or directory mingw32-g++: no input files ===== So I updated to SVN 18501, and someone changed main.cpp, which now gives me this error; ===== Make error line 201: problem compiling: src/main.cpp: In function 'void sp_do_ex port_png(SPDocument*)': src/main.cpp:1027: warning: 'sp_item_invoke_bbox' is deprecated (declared at src /sp-item.h:215) src/main.cpp: In function 'bool replaceArgs(int&, char**&)': src/main.cpp:1419: warning: missing initializer for member '_WIN32_FIND_DATAW::f tCreationTime' src/main.cpp:1419: warning: missing initializer for member '_WIN32_FIND_DATAW::f tLastAccessTime' src/main.cpp:1419: warning: missing initializer for member '_WIN32_FIND_DATAW::f tLastWriteTime' src/main.cpp:1419: warning: missing initializer for member '_WIN32_FIND_DATAW::n FileSizeHigh' src/main.cpp:1419: warning: missing initializer for member '_WIN32_FIND_DATAW::n FileSizeLow' src/main.cpp:1419: warning: missing initializer for member '_WIN32_FIND_DATAW::d wReserved0' src/main.cpp:1419: warning: missing initializer for member '_WIN32_FIND_DATAW::d wReserved1' src/main.cpp:1419: warning: missing initializer for member '_WIN32_FIND_DATAW::c FileName' src/main.cpp:1419: warning: missing initializer for member '_WIN32_FIND_DATAW::c AlternateFileName' src/main.cpp:1523: error: invalid conversion from 'const gchar*' to 'gchar*' src/main.cpp:1523: error: initializing argument 1 of 'void std::vector<_Tp, _A lloc>::push_back(const _Tp&) [with _Tp = gchar*, _Alloc = std::allocator<gchar*> ]' ===== I don't think the first error has yet been dealt with, but the second error simply occurs before the first error gets a chance. Cheers, Rygle. -- View this message in context: http://www.nabble.com/Build-problems---Last-two-commits-%2818500---18501%29… Sent from the Inkscape - Dev mailing list archive at Nabble.com.

5 7

LGM
by Alexandre Prokoudine 02 May '08

02 May '08

Hi, So, who is coming to LGM? :) Alexandre

4 3

New Extensions and Translation or other requirements
by rygle 01 May '08

01 May '08

Hi, Inductiveload has been busy fixing and making extensions. The version of Add Nodes in trunk can now divide into a given number of segments (see LP#171856), and there is also now a Render -> Polar Grid option in trunk (see LP#224522). Soon there will also be a Render -> Cartesian Grid extension, but that's got a slight problem being sorted (see LP#224955). I've recently added the new files to /share/extensions/Makefile.am - I realise that affects the Linux (Mac?) build process, even if it doesn't affect Win32. Does /share/extensions/Makefile.am also flag files for translation (seems to hint at that)? If not, is there a need to change something else for translations? Is there anything else that needs to be updated for new extensions? Cheers, Rygle. -- View this message in context: http://www.nabble.com/New-Extensions-and-Translation-or-other-requirements-… Sent from the Inkscape - Dev mailing list archive at Nabble.com.

3 2

Re: [Inkscape-devel] [Bug 171856] Re: Segment Division
by Aaron Spike 01 May '08

01 May '08

Rygle wrote: > Hi guys, just trying to get my head around all this. No problem. > I committed this and milestoned for release in 0.47, but I assumed that > Aaron and/or a release warden would check it out before the proper > release. I was trying to flag somehow that this is new and needs to be > approved. That's why I set it as Wish List and Fix Committed, not Fix > Released. Should I have instead marked it as Fix Released and nominated > it for release 0.47? Or just leave any milestoning/nomination because > it's automatically in? Seems like it won't get noticed by release > wardens then. There are no release wardens for 0.47 yet. Release wardens impose marshal law during release times to make sure things get out the door in a reasonable amount of time and without any avoidable nasty bugs. During the normal development cycle we average contributors are in charge. If you've been granted commit access the Inkscape community trusts you to decide what goes into the tree. > I'm also unsure, but based on Maximillian's comments, but do some people > feel that applying patches treads on toes if the bug is assigned to > someone else? I merely wanted to get it into trunk so that people would > find any problems with it. I've seen some good patches that didn't make > it into 0.46 because they didn't get attention. Should I not commit > patches when the bug is assigned to someone else? Please be frank, > because I don't want to offend others. I think sometimes applying a patch assigned to another person could possibly maybe tread on another person's toes perhaps. This time it didn't. I grabbed this patch because the author had written me individually about the change and because there doesn't seem like a lot of people who want to deal with python patches. But I knew that I wasn't going to get around to committing it until the weekend and it is very important to have quick turn around on patches. This shows that we value the work being done and encourages repeat contribution. Just think how discouraging it is when I sit on a two line python patch that is obviously fine. You're timing was excellent. Feel free to commit a patch. As long as you do due diligence. Meaning: You understand the issue it addresses. You test it locally to see that it does address that issue. You do not notice any major regressions as a result of it. Otherwise ask for a little clarification and help. And if you feel comfortable, go ahead and assign bugs to yourself. :-) And, yeah, it probably doesn't hurt to ask if the bug is assigned to someone else. I know in my situation sometimes I don't comment everything I know on the bug because I don't think of it. There might be some issue I want to check into. So if you ask I'd be more likely to write that in the bug comments. Not trying to hide info; I just don't think of these things all the time. Thanks for applying this patch! Aaron P.S. I'm copying Inkscape-devel to give people a chance to disagree with me. :-)

1 0

New tool added
by Jon A. Cruz 01 May '08

01 May '08

I've just committed the first cut of the new eraser tool. This first part only adds in the functionality to cut away parts of objects you have selected. It won't do anything if you don't have something selected, and it won't do anything if the first mode is selected (the sub-toolbar has toggles for the two modes). However it does commit all the infrastructure changes including the new icons, new drawing context, verbs, switching, etc. I had been wanting to get that in first to shake down bugs in that support work. Hopefully I'll be able to get the other mode committed this weekend. That's the main differentiator and works like touch-delete or perhaps pop-the-balloons. Please let me know if there is anything in that commit that needs fixing, etc. etc. Thanks.

10 23

Gobject or c++???
by Martin (OPENGeoMap) 30 Apr '08

30 Apr '08

Hi all: I want ask a question to inskape community because it´s expert in this matters... What it´s the best method to new GTK software?? - c++/gtkmm - GTK/ gobject - VALA - a mix of gtk, gobject and gtkmm like inskape and k-3d Regards.

2 2

W and H setting broken when geometric bounding box used
by qazxswedcvfrtgbnhyujmkiolp2000 qazxswedcvfrtgbnhyujmkiolp2000 30 Apr '08

30 Apr '08

Hi inkscape-devel, I notice that if i draw a rectangle, then try and adjust W or H (e.g. type 123.456 in the box), that W or H is not set correctly (e.g. it comes out slightly different to the value I typed)!. Works OK if visual bounding box is used, broken if geometric bounding box is used. Is this a bug? Thank you, qazxswedcvfrtgbnhyujmkiolp2000

1 0

Re: [Inkscape-devel] [lib2geom-devel] inkscape svn write-enabledaccess request
by Bob Jamison 29 Apr '08

29 Apr '08

Then I must be missing something. Isn't the whole point of linking directly to another repository the ability to always be referencing the current state of the subscribed-to tree? Not a snapshot. Or does it not work that way? bob >Please no. SVN externals are made of pain and fail. It should be >obvious why, if you think about it -- the external and local versions >are not locked in sync, so you can't get a "known good" configuration >when doing e.g. regression testing. > >Every project I've been in that has used svn externals has eventually >abandoned them for that reason. > >-mental >

2 2

GSoC tag-team for lib2geom integration
by unknown＠example.com 29 Apr '08

29 Apr '08

Hi all, Yesterday I had a chat with mgsloan about lib2geom integration for GSoC'08. We spoke about working together on it. He knows 2geom very well, I know Inkscape well and have worked with 2geom a lot for LPE. Because complete conversion to 2geom is a big task, where 2geom issues can become apparant, I think it is nice to have a 2geom developer working on it aswell (although I did commit some small things to 2geom, I do not count myself as a real 2geom developer). This way, we can both work and discuss things where we are uncertain. I hope to have a somewhat longer discussion with mgsloan within the next couple of days. How do people feel about this? Apart from having a 2-man team on lib2geom integration: I know I have been very brief in my gsoc proposal as blueprint. I am thinking about 1. writing test code (cxxtest) 2. integrating lib2geom in steps, for example per tool or per feature 3. switching to 2geom path representation in inkscape core is advanced task and should be tried later in the project 4. I'd rather *not* work in a branch, because then people that code new things will use 2geom instead of me/us having to recode new things to 2geom. (conflicts etc abound) 5. Release lib2geom after integration is complete. Not before integration is complete because of things that might be missing or bugged in 2geom. Thanks for your opinion, Johan

7 10

Win32 0.46.1 patches
by rygle 29 Apr '08

29 Apr '08

Hi guys, Just letting you know I've moved all previous Win32 patches for 0.46 to /packaging/win32/patches/0.46, and created a new 0.46.1 directory, just to keep things neat. Bryce and others have expressed concern about linux distros getting narky about unnecessary bug fixes, hence keeping the Win32 only patches separate. That way we can fix a bunch of stuff that was still broken in the win32 0.46 release and not bother the linux distro people at all. All these patches should already be in trunk, and so should automatically make it into 0.47 without remaining separate (I think). The Mac guys might like to do the same ?? Can we keep milestoning bugs for certain releases (unless they're not likely to be solved by 0.47) - with only bug fixes of existing features being marked of 0.46.1. Can we also start adding patches to the win32 0.46.1 section for stuff that's been tried in trunk? I've committed two patches that were in trunk, but there are a bunch more out there already. Thanks, Rygle. -- View this message in context: http://www.nabble.com/Win32-0.46.1-patches-tp16965858p16965858.html Sent from the Inkscape - Dev mailing list archive at Nabble.com.

1 1

export of grids and guides to XCF
by Alexandre Prokoudine 29 Apr '08

29 Apr '08

Hi, https://bugs.launchpad.net/bugs/222655 has an attached patch to add grid/guides export to XCF exporter. Could someone who knows Python well please have a look at it? Alexandre

2 1

new SVGEllipticalArc implementation merged into path.h
by Marco 29 Apr '08

29 Apr '08

I merged my implementation of the SVGEllipticalArc class into path.h substituting the old one. Some method lacks of a native implementation, yet and it relies on a conversion to SBasisCurve, but the class is fully working (there isn't any not_implemented exception thrown any more). Moreover I commit a new version of the svg-elliptic-arc-toy, with the ability of testing the portion method too. PAY ATTENTION ! I want to point out that I modified the src/CMakeList.txt file, adding some line of code in order to find out which version of ragel is installed. This is needed because starting from version 5.18 the ragel package substituted the command rlcodegen with the rlgen-cd command as specific code generator back-end for C, C++ and D languages. If you'll get into troubles when you build the library you'd look that you have the ragel package installed. Anyway in case of problem you can send me a mail or contact me on the lib2geom channel. Regards, Marco -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/

4 14

Learn SVG book.
by Marcin Floryan 29 Apr '08

29 Apr '08

Hi! I've no idea if this would be useful for anyone, but it showed up in one of my feeds so I thought it could be worth sharing: The English and French editions of the LearnSVG.com ebook is now available for free and can be downloaded here: http://www.learnsvg.com/ Regards, -- Marcin Floryan http://marcin.floryan.pl/ [GPG Key ID: 0D5581C5]

4 3

Re: [Inkscape-devel] New tool added
by Jon A. Cruz 29 Apr '08

29 Apr '08

On Apr 28, 2008, at 1:36 PM, Ronan Zeegers wrote: > I think that a button in the toolbox for the eraser tool is a good > idea. But it should just load the calligraphic tool (or his option > toolbar) with a different preset and a different boolean mode. > btw, this boolean mode with an auto union cloud be really helpfull > with the calligraphic tool too. It could provid the "paint feeling" > that Flash give... It also prevent to have lots of bezier lines > when you're using the calligraphic to do a complex draw. It's probably going to be close to that. Bbyak had already been working on such an additive and subtractive options for the calligraphic tool. There's a bit of a difference in some use cases, and it will be diverging more. From what I can see of things, though, the calligraphic tool will be very useful for a different set of operations. > An other thing; it would be really cool to be able to auto switch > to the eraser tool when using the eraser of a wacom tablet. Eeek! You got me! :-) Actually with the code in SVN it should be able to do almost that already. It should be detecting input devices and switching to the last tool used with a given device (tablet mouse/puck, stylus front, eraser back), so once you use the eraser end and use it to pick the eraser tool, it should switch tools as you go. I've just not tested that again recently since I'm on my box running Hardy and that has some wacom issues. And thanks for reminding me. I'll make that tool the default for any "ERASER" type extended input device.

1 0

Inline Function Removal
by Jimmy 28 Apr '08

28 Apr '08

I am currently working on Inline Function Removal and while replacing calls to sp_repr_set_attr with direct calls to the wrapped member functions I noticed that the wrapped function is defined as void. There are tests that are run to check if sp_repr_set_attr fails. sp_repr_set_attr always returns true although I suspect this was designed to change at some point If sp_repr_set_attr is replaced in one of these tests for example: /// \todo fixme: Exception if object is NULL? */ if (!sp_repr_set_attr(object->repr, key, NULL)) { No value will return and the compilation will fail. Should I just comment out these tests or remove them and leave the todo? As of now the tests just accept true from sp_repr_set_attr and they don't actually check anything. Jimmy Varvaro

2 1

Filter examples update
by Joshua Blocher 28 Apr '08

28 Apr '08

Ted, Mauve provided a newer version of inkscape-filters.svg with more filter examples in it. I simply updated it in svn. includes edge detection, sephia, age, blueprint, emboss, and many more. Everything below Patterned Glass is new. Let me know if there is something i can do. Joshua L. Blocher

1 0