- Inkscape Devel - Inkscape Mailman

NEW: dragging out handles
by bulia byak 16 Mar '05

16 Mar '05

Here's another Node tool improvement: - If a node does not show one or both handles (i.e. they are retracted), you can drag a handle out by dragging away from the node with Shift. This is often more convenient than using the "Convert node to smooth" or "Convert segment to curve" buttons. Unlike before, dragging a node without modifiers, or with any modifiers but Shift, will always move the node itself, not its retracted handle. Retracted handles are now not displayed at all, so for the user there's no difference between a node on a line segment (without handles) or a curve node with retracted handles. They behave the same, and you can Shift+drag out a handle easily from both.

3 2

interface discussions
by Kees Cook 16 Mar '05

16 Mar '05

I love that the core features of Inkscape are so far along that the devel list has dived into long discussions of how to make subtle tweaks to the interface to make things _even better_. :) -- Kees Cook @outflux.net

1 0

NEW: small improvement in drawing tools
by bulia byak 16 Mar '05

16 Mar '05

Inspired by http://www.angelfire.com/mi/kevincharles/inkscape/aiwf.htm, I added a small improvement to the drawing tools: - When you switch to the Pen or Pencil tool, they now start in the "add" mode: the selected path displays endnodes to attach to, and any path you draw becomes a single object with the selected path. To start a new path object, deselect the selected path by pressing Esc. Previously you had to press "a" to go to the add mode, and I'm sure many people were not aware of this. You can still turn the add mode off by pressing "a", but I'm thinking of removing that because I don't see what advantage that may have. If you don't want to continue the path, just deselect it. Removing the "non-add" mode altogether will make the interface simpler. Any comments?

5 16

NEW: Unsetting paint and painting clones
by bulia byak 16 Mar '05

16 Mar '05

New in CVS: * Unsetting paint and painting clones: Both Fill and Stroke tabs of the Fill&Stroke dialog have a new mode button, "Unset paint" (with a question mark icon). Pressing this button removes the fill or stroke property from the style of selected objects, which has the visible effect of painting them black. The usefulness of this is that clones of such an object can redefine the unset fill or stroke, which means you can have clones painted differently from their original and from each other. Just use the Fill&Stroke dialog on a clone to assign it any kind of fill or stroke paint (flat color, gradient, etc.). Moreover, you can unset paint on some of the objects in a group, clone the group, and paint the clone; only those objects with unset paint will accept the paint, while others will retain their original paint.

2 2

Inkscape Status 3/15/05
by Bryce Harrington 16 Mar '05

16 Mar '05

Tuesday, Mar 15, 2005 The last two weeks have seen a number of new features added, the fruit of some quite heated (and fruitful) discussions on the list. Bulia's work on redesigning gradients has generated a lot of good feedback from the community and should really prove to be a big step forward for the application. Mental has been continuing his work on converting C code to C++ and attending to some of the more complex bugs as they pop up. The new gtkmm-based gui code is now accessible via 'inkscape --new-gui'. Jon Cruz is adding a new 'panel' system to replace the current 'dialog' focused design, that will give people more flexibility in how the user interface looks and works. There's still a lot more work needed, though, and we desperately need more people working on it. We're also seeing the impact of the involvement of a number of new contributors. Cyreve is working on improving text flow and wrapping. Kevin Wixson has been plugging away at improving documentation. Statistics Feb 1 Feb 15 Mar 1 Mar 15 ========== ===== ====== ===== ====== Lifetime Rank on SourceForge: 569 * * * (96.0%) * * * Max Week's Rank on SourceForge: * * * * * * * * Total SF Page Views 2,296,458 * * * Total SF Downloads 115,645 * * * Total Freshmeat URL Hits 15,108 16,448 16,885 17,186 Total Freshmeat Subscriptions 156 169 173 173 Lines of Code in src/: 323,324 333,139 338,993 342,852 Code lines 227,872 234,651 238,641 241,177 Comment line 56,438 58,239 59,589 60,354 Blank 45,624 46,859 47,484 47,978 Lines of Docs in doc/: 17,918 26,848 26,848 26,848 Lines of content in website: 9,486 9,776 9,979 10,067 Size of the Inkscape wiki: 23,576 24,925 25,094 25,900 Bugs open/total: 148/917 155/972 177/1019 182/1047 Features open/total: 352/556 370/579 384/ 602 397/ 619 Patches open/total: 7/154 5/167 6/ 170 7/ 173 CVS Commits (as per inkscape-cvs) 9,975 10,312 10,600 10,868 Inkscape-devel membership: 143 151 152 158 Inkscape-announce membership: 166 169 171 177 Inkscape-user membership: 181 193 201 220 Num Translations: 33 33 33 33 Ave Translation Ratio: 44.7 47.8 45.0 45.0 * - Statistics unavailable

1 0

Compiler warnings
by Bob Jamison 15 Mar '05

15 Mar '05

I spent a few hours yesterday compiling the tree with gcc4.0.0, and removing the warnings, so it should compile fairly cleanly now. Since warnings are a useful tool to indicate possible bugs (and I found one yesterday), can people please make sure that their committed code compiles without them? I know that "unused variable" and things like that can be a pain to avoid, especially in a "work in progress." But if we allow ourselves to become desensitized to warnings, we will likely ignore problem conditions and possible bugs. Bob

2 1

Re: [Inkscape-devel] translated win32 installer
by Adib Taraben 15 Mar '05

15 Mar '05

Ciao Luca, Do you want a screenshot? all the other translation is all done with the include of the language file. I already checked this with the latest CVS at the office (Torino). Only the GPL is not translated. The german GPL translation project refers to the original english version as the only official and legally binding. I would see in the installer an option to select, if the installation is visible to the login that did the installation or to all users of the computer. HTH, Adib. Luca Bruno schrieb: > Bob Jamison <rjamison@...357...> scrisse: > > >>>>I did some work on the .nsi for the win32 installer. I included a >>>>german >>>>translation. It should be very simple to add more languages. > > >> > Luc, that is wonderful. Thanks for adding this section. You >> > probably >> > only need to change the type to ${LANG_ITALIAN}, but that is >> > the only thing I see.. > > > >>I fixed that, and it is committed. Thanks again. > > > Thank you for the commit and the check (sorry, my mistake, I forgot to > change the label)... > > Now I use only Linux, but if I remember correctly, there also other > steps in a normal Win installer (such as program dir chooser, license, > etc...) that I think could be translated. > I'm not aware of technical details, so I'm asking: Is it possible? How? > If somebody show me an hack about it, I'd like to help with a full > Italian translation... > > Please, let me know :) > > Ciao, Luc@ > > P.S. I use to sign all my mail with "Ciao, Luc@", but my correct (in > Italian) name is "Luca"... :) >

1 0

Inkscape for AI users: a revert war :)
by bulia byak 15 Mar '05

15 Mar '05

Recently the wiki user Kwixson has significantly expanded and rewritten the "Inkscape for AI users" document on our wiki. Some of his edits seemed strange to me, and I proposed my variants. Unfortunately we could not reach a consensus, so after several mutual reverts we decided to post here and ask for the help and judgement from the community. The latest revision with our extensive comments is here: http://inkscape.org/cgi-bin/wiki.pl?IllustratorUsers Below I briefly list the things I disagree with. 0. To begin, I really appreciate the input on this document from the user who obviously knows AI well. This is some kind of expertise which I personally lack, so I would not be able to write such a document all by myself. Obviously it must be written from an AI user's point of view and use AI terminology and concepts. 1. However, Kwixson's approach seems to go further than that. He apparently wants this document to describe _only_ those things which you can do in Illustrator, and describe them exactly _the same way_ as one would do them in Illustrator, even if Inkscape offers more convenient methods. I think this is wrong. 2. With Kwixson's approach, Inkscape would always seem inferior to AI, because you can't win just by copying. You can't be a better Illustrator than Illustrator, you need to be different to be better. And indeed Kwixson's text has a generally condescending tone and phrases like "is much less responsive", "is not as intuitive" etc. We need to be honest about Inkscape's weaknesses, but these judgements were often not based on any real weaknesses. 3. After all, who is this document for? Those who like AI the way it is will never switch, so it seems stupid to target them in such a document. I think we need to target those who are used to AI but are looking for something different and better. And therefore we must stress our strengths and differences, explaining them in a way which is easy to understand for AI users. Now on to specific examples. A. Kwixson has removed my mention of keyboard accessibility, in particular keys for screen-pixel-sized transformations, claiming this is not important. I've seen this attitude before from other AI users; they tend to dismiss this because they don't have it. Those who are really using Inkscape (or Xara, where I got the idea from, though by now Inkscape's keyboard is superior even to Xara) will disagree. B. In the section on shapes, Kwixson has removed my explanation of the difference between a shape and a path and the unique features that shapes offer. Instead he inserted an advice to do Ctrl+Shift+C (convert to path) as soon as you created a shape, to be able to node-edit it! This is because AI does not have shapes as such, treating everything as paths. I think this is plain stupid. Inkscape's shapes are clearly superior to those of any other program I know, and we must present them as such. C. In the section on Node tool, Kwixson provided some very cumbersome descriptions of how to convert a segment from curve to straight line and how to continue a path. When I proposed much simpler and more straightforward ways to do the same, he insisted that his descriptions closely match the way AI does this and therefore must stay. Once again, I don't see why one should go through all this when there's a much simpler way. Disclaimer on this point: I cannot even claim to completely understand Kwixson's descriptions, so I may have missed something important in them. Please anyone who knows AI's path editing, review these paragraphs and let us know what you think. Any feedback will be appreciated. I'm crossposting this to the user list too.

9 29

CVS compiling for the uninitiated
by unknown＠example.com 15 Mar '05

15 Mar '05

I compiled CVS successfully a while back and was very excited. So today I tried: cvs -z3 update -Pd make and got: Making all in ui make[3]: Entering directory `/usr/local/src/inkscape/share/ui' make[3]: *** No rule to make target `all'. Stop. make[3]: Leaving directory `/usr/local/src/inkscape/share/ui' make[2]: *** [all-recursive] Error 1 make[2]: Leaving directory `/usr/local/src/inkscape/share' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/usr/local/src/inkscape' make: *** [all] Error 2 If I were to take a stab in the dark I would guess my makefile is too old and it doesn't include rules for the new ui stuff. I could be very wrong, but this makes me wonder when and how often should I use commands like "make clean", autogen.sh and configure when working with Inkscape CVS. I don't think I saw these sorts of instructions in the wiki. (I do think I read a short blurb about using autogen.sh when new folders enter the tree.) Sorry for such rudimentary questions. Education appreciated, Aaron Spike

3 4

Experimenting with the gcc4.0.0 compiler
by Bob Jamison 15 Mar '05

15 Mar '05

I downloaded/installed a prerelease snapshot of gcc4.0.0 today, and did a clean build with it. I expected it to produce better warnings than 3.4, since 3.4 was more strict than 3.3. And it does! Here is an example: ./xml/node.h:42: warning: ‘class Inkscape::XML::Node’ has virtual functions but non-virtual destructor ./xml/document.h:23: warning: ‘struct Inkscape::XML::Document’ has virtual functions but non-virtual destructor ./xml/sp-css-attr.h:20: warning: ‘class SPCSSAttr’ has virtual functions but non-virtual destructor The planned release of 4.0.0 is Apr 15, so it would be wise to keep up with developments. Bob

3 4