Difference between revisions of "Google Summer of Code"

From Inkscape Wiki
Jump to navigation Jump to search
(better structure)
Line 54: Line 54:
  
 
== Internal work & Performance improvements ==
 
== Internal work & Performance improvements ==
 +
 +
This is a laundry-list of things which need doing. A SoC project would involve doing a few related things from this list.
 +
 +
===General===
  
 
* Improve startup time and initialization performance (i.e. improve font cache).
 
* Improve startup time and initialization performance (i.e. improve font cache).
 
* Improve icon cache: Convert the icon cache (widgets/icon.cpp) to create the PNGs with a directory structure and index file matching the icon theme specification. Use this to get rid of customized classes InkAction, SPIcon, and so on.
 
* Improve icon cache: Convert the icon cache (widgets/icon.cpp) to create the PNGs with a directory structure and index file matching the icon theme specification. Use this to get rid of customized classes InkAction, SPIcon, and so on.
* Remove dom/ directory: This directory contains strange code which is barely used. The only class which is actually used is the URI class.
+
* Continue C++ migration:
 +
** Use Glib::ustring for UTF-8 data and std::string for filenames and raw byte strings instead of C-strings (char*).
 +
** Replace uses of GList and GSList with STL containers.
 +
* Use a different data structure for Inkscape::Selection. Inkscape::Selection currently uses GSList as its data structure. This is suboptimal, since a very common operation is checking whether some object is selected. Change Inkscape::Selection so that its underlying structure is a boost::multi_index container which implements the same semantics as Java's LinkedHashSet.
 +
* Extension system improvements: Refactor the extension API. Clean up the kludgy class hierarchy, possibly using multiple inheritance. Use GInputStream and GOutputStream as parameters instead of file paths, so that things like the clipboard can write data to memory without going through a temporary file.
 +
 
 +
=== XML tree ===
 +
 
 
* SAX parser: Convert the current document parser from DOM to SAX, so that it creates our XML tree right away, instead of creating the libxml2 DOM tree, creating our tree to match it, then freeing the libxml2 tree. This should improve performance and allow more robust fixes for some problems.
 
* SAX parser: Convert the current document parser from DOM to SAX, so that it creates our XML tree right away, instead of creating the libxml2 DOM tree, creating our tree to match it, then freeing the libxml2 tree. This should improve performance and allow more robust fixes for some problems.
 
* Typed XML tree: Improve XML tree so that it can store some attributes in parsed, binary form. The main target of this are the data URIs used to embed images, which could be stored as binary data only. This work should be done after completing the SAX parser, as this will make it easier.
 
* Typed XML tree: Improve XML tree so that it can store some attributes in parsed, binary form. The main target of this are the data URIs used to embed images, which could be stored as binary data only. This work should be done after completing the SAX parser, as this will make it easier.
 +
 +
=== 2Geom geometry library ===
 +
 
* Rewrite of Geom::PathVector and Geom::Path:
 
* Rewrite of Geom::PathVector and Geom::Path:
 
** Change PathVector to be a real object instead of a std::vector of Path, so that it can have useful methods, similar to curves.
 
** Change PathVector to be a real object instead of a std::vector of Path, so that it can have useful methods, similar to curves.
Line 67: Line 81:
 
*** Path -> Subpath
 
*** Path -> Subpath
 
*** PathVector -> Path
 
*** PathVector -> Path
 +
* Remove SPCurve: SPCurve is a thin wrapper around Geom::PathVector which exists for historical reasons. Its functionality should be added to PathVector, and SPCurve should be purged.
 
* Boolean operations and stroking: Add methods to PathVector objects:
 
* Boolean operations and stroking: Add methods to PathVector objects:
 
** Set operators (& | - ^), which perform the relevant boolean operation on the paths. Use the algorithm from CGAL or devise a new robust algorithm.
 
** Set operators (& | - ^), which perform the relevant boolean operation on the paths. Use the algorithm from CGAL or devise a new robust algorithm.
 
** stroke(double line_width, LineJoin join, LineCap cap, double miter_limit), which performs the stroke-to-path operation.
 
** stroke(double line_width, LineJoin join, LineCap cap, double miter_limit), which performs the stroke-to-path operation.
 
** stroke(double line_width, LineJoin join, LineCap cap, double miter_limit, std::vector<double> const &dasharray), which performs stroke-to-path with dashing.
 
** stroke(double line_width, LineJoin join, LineCap cap, double miter_limit, std::vector<double> const &dasharray), which performs stroke-to-path with dashing.
 +
 +
=== Object tree ===
 +
 +
* Improve the performance of layer visibility: Right now, toggling layer visibility causes massive changes in the display tree, because the entire toggled layer is invalidated at the XML level. This results in very bad performance for an action which should nearly instantaneous. Improve the control flow so that only the visibility of the object representing the layer is turned off, but the remainder of the display and object tree is leaved alone.
 +
* Robust ID handling: Currently ID clash resolution is implemented as a giant switch over object types, listing every possible dependency. Replace this with virtual methods on objects, which provide information on what each object refers to. Improve behavior when a dependency of an object is deleted, and when IDs are changed / deleted from the XML editor.
 +
* Replace the print functions in the SPxxx classes with a generic "Visitor" pattern and port at least one export type (e.g. EMF, PDF, PS) to the new routines.
 +
 +
=== User interface ===
 
* Shape manipulators: The idea here is to rewrite shape tools in the same paradigm as the node tool. Instead of storing all information about the shape in knots which differ only by their callbacks, allow to store information in a higher-level manipulator object. This would enable things like dragging the side of a rectangle and consistent outlining / update preferences for all shapes.
 
* Shape manipulators: The idea here is to rewrite shape tools in the same paradigm as the node tool. Instead of storing all information about the shape in knots which differ only by their callbacks, allow to store information in a higher-level manipulator object. This would enable things like dragging the side of a rectangle and consistent outlining / update preferences for all shapes.
* Improve the performance of layer visibility: Right now, toggling layer visibility causes massive changes in the display tree, because the entire toggled layer is invalidated at the XML level. This results in very bad performance for an action which should nearly instantaneous. Improve the control flow so that only the visibility of the object representing the layer is turned off, but the remainder of the display and object tree is leaved alone.
+
* Provide more actions in the context menu depending on the content of the selection, e.g. show "Put on Path", "Flow into Frame" etc. when the selection contains the appropriate objects.
 +
* When right-clicking a control point, show a transient dialog which would allow one to key in its coordinates.
 +
 
 +
=== Rendering ===
 +
* [https://blueprints.launchpad.net/inkscape/+spec/icc-for-cairo ICC Color Management for cairo outputs] - Would allow Inkscape to produce CMYK PDFs and PSs with Cairo. <!-- Isn't this more of a Cairo project? -->
 +
* Box blur: Currently we always use a very accurate method to compute the Gaussian blur filter. Add an alternate method which approximates Gaussian blur using three stacked box blurs (simple averages). This is detailed in the SVG 1.1 SE specification.
 +
 
 +
=== Project infrastructure and platform support ===
 +
 
 +
* Migrate argument parsing to GOption and remove the dependency on popt. Update the devlibs version of Glib to one where this bug is fixed: [https://bugzilla.gnome.org/show_bug.cgi?id=522131]. Once this is done, port argument parsing to GOption.
 
* Common build system for all platforms: Migrate the build system to Waf for all platforms. Alternatively, if there are important technical reasons why Waf is not suitable, port all necessary features to the CMake build system and remove Autotools.
 
* Common build system for all platforms: Migrate the build system to Waf for all platforms. Alternatively, if there are important technical reasons why Waf is not suitable, port all necessary features to the CMake build system and remove Autotools.
* Remove SPCurve: SPCurve is a thin wrapper around Geom::PathVector which exists for historical reasons. Its functionality should be added to PathVector, and SPCurve should be purged.
 
* Use a different data structure for Inkscape::Selection. Inkscape::Selection currently uses GSList as its data structure. This is suboptimal, since a very common operation is checking whether some object is selected. Change Inkscape::Selection so that its underlying structure is a boost::multi_index container which implements the same semantics as Java's LinkedHashSet.
 
* Box blur: Currently we always use a very accurate method to compute the Gaussian blur filter. Add an alternate method which approximates Gaussian blur using three stacked box blurs (simple averages). This is detailed in the SVG 1.1 SE specification.
 
 
* GTK3 on Windows: Rebuild the Windows devlibs so that they contain GTK3. Make the Windows port work with them, possibly sending the appropriate patches to the GTK maintainers.
 
* GTK3 on Windows: Rebuild the Windows devlibs so that they contain GTK3. Make the Windows port work with them, possibly sending the appropriate patches to the GTK maintainers.
* Migrate argument parsing to GOption and remove the dependency on popt. Update the devlibs version of Glib to one where this bug is fixed: [https://bugzilla.gnome.org/show_bug.cgi?id=522131]. Once this is done, port argument parsing to GOption.
 
* Extension system improvements: Refactor the extension API. Clean up the kludgy class hierarchy, possibly using multiple inheritance. Use GInputStream and GOutputStream as parameters instead of file paths, so that things like the clipboard can write data to memory without going through a temporary file.
 
* Robust ID handling: Currently ID clash resolution is implemented as a giant switch over object types, listing every possible dependency. Replace this with virtual methods on objects, which provide information on what each object refers to. Improve behavior when a dependency of an object is deleted, and when IDs are changed / deleted from the XML editor.
 
* Continue C++ migration:
 
** Use Glib::ustring for UTF-8 data and std::string for filenames and raw byte strings instead of C-strings (char*).
 
** Replace uses of GList and GSList with STL containers.
 
* [https://blueprints.launchpad.net/inkscape/+spec/icc-for-cairo ICC Color Management for cairo outputs] - Would allow Inkscape to produce CMYK PDFs and PSs with Cairo. <!-- Isn't this more of a Cairo project? -->
 
* Replace the print functions in the SPxxx classes with a generic "Visitor" pattern and port at least one export type (e.g. EMF) to the new routines.
 
  
 
== Rendering Improvements ==
 
== Rendering Improvements ==

Revision as of 20:55, 9 March 2014

Welcome to Inkscape!

For quite a few years Inkscape has been successfully participating in Google Summer of Code.

GSoC is a program where Google funds the development of specific features in open source software by university students. You don't need to be a Computer Science student to apply. Features to be developed are picked by Inkscape administrators from the pool of proposals submitted by students.

We've mentored about half a dozen students a year since GSoC started. Many students enjoyed their work and continue to be involved; perhaps your mentor will be a past GSoC student! We have a high rate of acceptance of student code into the core codebase. Indeed, GSoC projects have been a key source of some of Inkscape's best features in the past several releases.

If you are interested in joining us this summer, it is time to get your proposal ready. You can choose a proposal from our list of suggestions or come up with one of your own. In either either case, you must give us a detailed outline of what you plan to do. It is highly recommended that you discuss your idea as early as possible with Inkscape developers. They not only can give you guidance as you flesh out your proposal but ultimately you must convince them that you can do the work planned in the time allotted.

Student Applications

  • Google program information:
    • Home page.
    • Summer of Code Application form. Applications open at 19:00 UTC on March 10th.

Student Applications for GSoC 2014 must be submitted to the GSoC site by 19:00 UTC on March 21st 2014.

The "two patches" rule

We require two patches from each potential GSoC student, before accepting the student for GSoC participation (it is the same requirement as for obtaining rights to commit changes to the Bazaar code repository). The reason for this requirement is that you can show us that you have succeeded in building Inkscape on your PC, and that you have understood a little piece of Inkscape's code and are able to improve it. Inkscape is a large project, and you really should not try to understand all the code. Many (all?) developers know only parts of the program code! You can join our jabber/IRC channel and ask developers for help.

Suggested "easy" bug fixes or improvements

To get you started on Inkscape development, you can find (probably) easy-to-fix bugs or small improvements that require very little knowledge of the whole program by searching our bug-tracker for bugs tagged with 'easy-fix'.

Performance Evaluation

GSoC has two formal evaluation points, at the mid-term and at the end. These evaluations determine if you receive the stipend from Google. In order to receive a pass for the evaluations you will need to show adequate progress toward your project's goals.

To help you meet your goals and so that your mentor can better evaluate your progress you need to:

  • Have frequent, public discussions of your progress. (Don't rely on just your mentor for advice.)
  • Have a public Inkscape branch for your code to which you commit regularly.
  • Give weekly status reports.

For the final pass, you will normally be required to merge your code into Inkscape trunk.

Remember: we want you to succeed!

Suggested Ideas

Please note that your project proposal must be realistic: if you promise too much, we will most likely think you have not understood at all what it is about and will reject your proposal.


Internal work & Performance improvements

This is a laundry-list of things which need doing. A SoC project would involve doing a few related things from this list.

General

  • Improve startup time and initialization performance (i.e. improve font cache).
  • Improve icon cache: Convert the icon cache (widgets/icon.cpp) to create the PNGs with a directory structure and index file matching the icon theme specification. Use this to get rid of customized classes InkAction, SPIcon, and so on.
  • Continue C++ migration:
    • Use Glib::ustring for UTF-8 data and std::string for filenames and raw byte strings instead of C-strings (char*).
    • Replace uses of GList and GSList with STL containers.
  • Use a different data structure for Inkscape::Selection. Inkscape::Selection currently uses GSList as its data structure. This is suboptimal, since a very common operation is checking whether some object is selected. Change Inkscape::Selection so that its underlying structure is a boost::multi_index container which implements the same semantics as Java's LinkedHashSet.
  • Extension system improvements: Refactor the extension API. Clean up the kludgy class hierarchy, possibly using multiple inheritance. Use GInputStream and GOutputStream as parameters instead of file paths, so that things like the clipboard can write data to memory without going through a temporary file.

XML tree

  • SAX parser: Convert the current document parser from DOM to SAX, so that it creates our XML tree right away, instead of creating the libxml2 DOM tree, creating our tree to match it, then freeing the libxml2 tree. This should improve performance and allow more robust fixes for some problems.
  • Typed XML tree: Improve XML tree so that it can store some attributes in parsed, binary form. The main target of this are the data URIs used to embed images, which could be stored as binary data only. This work should be done after completing the SAX parser, as this will make it easier.

2Geom geometry library

  • Rewrite of Geom::PathVector and Geom::Path:
    • Change PathVector to be a real object instead of a std::vector of Path, so that it can have useful methods, similar to curves.
    • Move the copy-on-write idiom to the PathVector object, rather than using it in the Path.
    • Investigate whether it is possible to store subpath data in a more compact way and make the Curve objects only convenience facades. Right now, if the path has only linear segments, every point is stored twice.
    • Apply the following renames to match SVG terminology:
      • Path -> Subpath
      • PathVector -> Path
  • Remove SPCurve: SPCurve is a thin wrapper around Geom::PathVector which exists for historical reasons. Its functionality should be added to PathVector, and SPCurve should be purged.
  • Boolean operations and stroking: Add methods to PathVector objects:
    • Set operators (& | - ^), which perform the relevant boolean operation on the paths. Use the algorithm from CGAL or devise a new robust algorithm.
    • stroke(double line_width, LineJoin join, LineCap cap, double miter_limit), which performs the stroke-to-path operation.
    • stroke(double line_width, LineJoin join, LineCap cap, double miter_limit, std::vector<double> const &dasharray), which performs stroke-to-path with dashing.

Object tree

  • Improve the performance of layer visibility: Right now, toggling layer visibility causes massive changes in the display tree, because the entire toggled layer is invalidated at the XML level. This results in very bad performance for an action which should nearly instantaneous. Improve the control flow so that only the visibility of the object representing the layer is turned off, but the remainder of the display and object tree is leaved alone.
  • Robust ID handling: Currently ID clash resolution is implemented as a giant switch over object types, listing every possible dependency. Replace this with virtual methods on objects, which provide information on what each object refers to. Improve behavior when a dependency of an object is deleted, and when IDs are changed / deleted from the XML editor.
  • Replace the print functions in the SPxxx classes with a generic "Visitor" pattern and port at least one export type (e.g. EMF, PDF, PS) to the new routines.

User interface

  • Shape manipulators: The idea here is to rewrite shape tools in the same paradigm as the node tool. Instead of storing all information about the shape in knots which differ only by their callbacks, allow to store information in a higher-level manipulator object. This would enable things like dragging the side of a rectangle and consistent outlining / update preferences for all shapes.
  • Provide more actions in the context menu depending on the content of the selection, e.g. show "Put on Path", "Flow into Frame" etc. when the selection contains the appropriate objects.
  • When right-clicking a control point, show a transient dialog which would allow one to key in its coordinates.

Rendering

  • ICC Color Management for cairo outputs - Would allow Inkscape to produce CMYK PDFs and PSs with Cairo.
  • Box blur: Currently we always use a very accurate method to compute the Gaussian blur filter. Add an alternate method which approximates Gaussian blur using three stacked box blurs (simple averages). This is detailed in the SVG 1.1 SE specification.

Project infrastructure and platform support

  • Migrate argument parsing to GOption and remove the dependency on popt. Update the devlibs version of Glib to one where this bug is fixed: [1]. Once this is done, port argument parsing to GOption.
  • Common build system for all platforms: Migrate the build system to Waf for all platforms. Alternatively, if there are important technical reasons why Waf is not suitable, port all necessary features to the CMake build system and remove Autotools.
  • GTK3 on Windows: Rebuild the Windows devlibs so that they contain GTK3. Make the Windows port work with them, possibly sending the appropriate patches to the GTK maintainers.

Rendering Improvements

  • Unify the interactive and non-interactive renderers. (It would be desirable to have one codebase, but we need to investigate a little more whether this is practical.)
  • Pluggable renderers - allow writing rendering backends which use something other than Cairo, e.g, OpenGL, Skia, Mozilla Azure or GEGL.
  • OpenGL renderer - implement an OpenGL 3.x+ canvas which would render Beziers using this method: http://http.developer.nvidia.com/GPUGems3/gpugems3_ch25.html Investigate whether this generalizes to S-basis and circular arcs. Since the described method does not handle stroking, this work depends on implementing boolean operations and stroking. Also check how OpenGL path rendering is implemented in Qt, since apparently the performance there is very good. http://zrusin.blogspot.com/2011/09/nv-path-rendering.html

Better SVG standard coverage

SVG 2 and CSS3

See: SVG2.

This includes providing suitable GUI and SVG 1.1 fallbacks.


Note: Some of the above items are somewhat low hanging fruit and would need a few of them to make a good proposal, others alone are worthy of a full GSoC project.

Improve Inkscape SVG for use on Web

A proposal should fix most if not all of these problems:

  • Fix "plain" SVG to not remove <script> section, etc.
  • Add option to set viewbox attribute to Document Properties.
  • Easily add/edit hyperlinks.
  • Better CSS support.
  • Option to add title to SVG (searchability and accessibility).
  • Remove XML prolog/DOCTYPE.
  • Move sodipodi namespace items to Inkscape namespace.

Live Path Effects related

Tools and shapes related

  • Multi shape editing - ability to edit more than 1 shape (like ellipse or star) at once, with extra features like resizing rectangles by dragging their edges.
  • Transformation Anchors
  • Extend the Symbols dialog to include generic objects.
  • Connector tool improvements - Expose new libavoid functionality: orthogonal routing, connector ports, curved connectors. (Mentor: Michael Wybrow)
  • Fillet/Chamfer tool


Please do not feel limited to the above ideas - some of our best contributions have been unique ideas that students had in mind from other sources!

For more ideas, please see https://blueprints.launchpad.net/inkscape/

Successful SOC Projects from Previous Years

  • 2005
  • 2006
    • Support for SVG Filters
    • Filter Effects
    • PDF export
    • Inkboard Protocol Spec / Lib Conversion
  • 2007
    • Text Style Improvements
    • PDF import
    • Live Path Effects
    • 3D Box Tool
    • UI for SVG Filter Effects
    • Raster Functionality
    • Importing from, and Exporting to, a remote ccHost instance
  • 2008
    • SVG Fonts support
    • 2Geom refactoring project - port most geometry code to 2Geom
    • lib2geom: interactive applications showing off the power of lib2geom
    • Tech drawing abilities
    • A test suite
  • 2009
  • 2010
    • Cairo-based rendering
    • C++ification of SP Layer
  • 2011
    • Rendering caching
    • Javascript support improvements
    • CSS support improvements
  • 2012
  • 2013
    • Recolor Tool
    • Improved Units Support
    • Electronics CAD Support
    • New From Templates Dialog
    • New Raster to Vector Algorithm