利用者:RedTriangle/Reports

提供: wiki
< 利用者:RedTriangle
2018年6月29日 (金) 05:49時点におけるYamyam (トーク | 投稿記録)による版 (1版 をインポートしました)
(差分) ← 古い版 | 最新版 (差分) | 新しい版 → (差分)
移動先: 案内検索

Week 1 Report

What you did this week

According to advice of my mentor I start my work from bridge tools. I started to work to port the 'looptools'(http://blenderartists.org/forum/showthread.php?148728-LoopTools-for-BMesh-new-version-May-21st) bridge command into C.

What you plan to do next week

Continued work on porting the code bridge command.


Week 2 Report

What you did this week

In this week I working on porting 'LoopTools' python plugin on C. These functions was rewrited :

  • get input loops
  • calculate new geometry

What you plan to do next week

I plan continue work on bridge command, test function to calculate new geometry (calculate new vertex, and new faces) Also rewrite next functions:

  • remove faces
  • create new vertices
  • create new faces

Week 3 Report

What you did this week

  • The calculation of loop error was Fixed. We have an error("Selection must only contain edges from two edge loops") when we select few adjacent faces. It was be released as in 'LoopTools'.It was necessary to exclude from consideration the adjacent edges.
  • Was developed algorithm for calculate bridge geometry. We construct a line between the nearest points of the two input loops (this will make a bridge for the case when we have different number of vertices in the input loops).

Vertices created on the line according to the segmentation parameters. In conclusion, new faces are generated.

What you plan to do next week

  • Continue coding and testing bridge calculate geometry algorithm.
  • It is necessary add parameter of interpolation (linear or cubic).

Week 4 Report

What you did this week

This week I’ve kept on working on bridge tool.

  • I sent a code of my version of bridge to my mentor (about 350 lines)
  • I've prepared a detailed wiki-page on issues of "bridge" which were rewrited. And added some examples of working of bridge (http://wiki.blender.org/index.php/User:RedTriangle/Bridge#Bridge)
  • I fixed some errors of determination of the nearest edges and vertexes.
  • I fixed error of determination of sequence of vertices on the creation of new quads.

What you plan to do next week

  • Finish work on the bridge tool
    • Add GUI for parameters of the bridge
    • Refactor code of my version of the bridge
  • Start working on bevel tool. Analyse existing bevel code and discuss it with my mentor.

Week 5 Report

What you did this week

Bridge:

  • I did a small refactoring bridge. I made ​​some changes accordance with the comments of my mentor.
    • replacement of some operations on the vector function
    • renaming functions
    • adding the correct comments
  • I solved the problem with duplicate vertices in the new polygons.
  • I did a more detailed description of bridge on my wiki page. (http://wiki.blender.org/index.php/User:RedTriangle/Bridge)

Bevel:

  • Analyse existing bevel code

What you plan to do next week

Bridge:

  • Make initial commit of my bridge in sushi branch
  • Add GUI for input parameters of the bridge

Bevel:

  • I plan work on bevel code. Try to port code of mesh_bevel_round.py on C.
  • I will continue testing Bevel

Week 6 Report

What you did this week

Bridge:

  • I prepared my code to be sent to SVN.
  • I continue to experiment with test cases for the bridge. I was able find some test cases in which the bridge is buggy.

Bevel:

  • in this week I worked with Python code bevel. I started working on a version of the C function to determine the geometry and the calculation of new polygons.

What you plan to do next week

Bridge:

  • Bug fixes to the bridge
  • Refactoring bridge. Rewrite bridge code from 'BLI_array' on 'ListBase'

Bevel:

  • Testing functions calculate a new bevel geometry.

Problems

  • I could not send my code to SVN, because of problems with authorization


Week 7 Report

What you did this week

Finally succeeded in solving all the problems with SVN and I made ​​its first commit for the bridge. In This week I worked only on the bevel. An analysis of the existing code and code BevelRound, I developed the algorithm, and engaged in its implementation. I posted the first version of the algorithm on the svn, but it is not over yet.

What you plan to do next week

I plan to continue work on the bevel. I'll work on functions related to the rebuilding of the selected polygons. Also modification bevel algorithm for round shape of bevel.

Week 8 Report

What you did this week

Bridge:

  • I fixed the problem with the normals.I discovered this problem when I switch to texture mode.
  • I fixed a bug with generation of double vertexes.
  • I added a GUI for specifying parameters of the bridge.
  • For the cubic interpolation I implemented Hermite-form, instead of the full spline interpolation. I think it gave a good result, the bridge looks intact.

Bevel:

  • I worked on the implementation algorithm for calculating additional construction around the vertex.

What you plan to do next week

Bridge:

Bevel:

  • I will continue work on the generation new faces and regeneration old face for selected edges.

Week 9 Report

What you did this week

I worked a lot on the bevel on this week. I finished the implementation of a linear bevel, it works rather well for an arbitrary number of input edges at the vertex. You can see some examples of how it works in my thread, to the blenderartist. (http://blenderartists.org/forum/showthread.php?255869-GSOC-2012-Modeling-Tools-Bevel-Bridge-Boolean.&p=2163289&viewfull=1#post2163289) I also implemented round vevel for simple cases (no more than two edges on vertex). Unfortunately I could not find the time the bridge on this week.

What you plan to do next week

I plan to focus on the implementation round bevel of the construction of curved surface near the top for any occasion. Also, on the advice of one of the blender artist, I plan to pay attention to the bug tracking and analyzing a list of existing bugs for my tasks.

Tasks for the bridge are the same as last week.

Week 10 Report

What you did this week

I implemented aloritm of generation n-gons for round bevel, in the case of rebuilding the existing faces. Develop and implement a generator for complex surfaces round bevel.

What you plan to do next week

I plan to continue working on the implementation and debugging of the algorithm for constructing complex surfaces around the selected points. After last merging we have a problems with run blender in the sushi branch. I plan to fix the conflicts associated with my code. I will describe my algorithm to work Bevel on the wiki page.

Week 11 Report

What you did this week

This week I worked on the Bevel. I finished coding the algorithm for constructing smooth surfaces around vertices with an arbitrary number of input edges. The surface is constructed by sectors and we have a complex problem with the fact that adjacent vertices of the sectors did not match. I was able to find a solution to cope with it. I add in the user interface input parameters of the partition (segmentation) Fixed a bug with the construction of a round profile for non-orthogonal input edges. Fixed a bug with the construction of the intersection of of round profiles in the case of two input Edge at the vertex. Fixed a bug with the construction of the intersection of circular profiles in the case of two input Edge at the vertex. When implementing a round bevel become clear how to build it for the plane the selected faces. Added a correct handling of a flat location selected faces.

What you plan to do next week

I plan to finish the work on Bevel, finish a UI, clear unnecessary parameters. Continue to debug and fix small bugs, such as a bug with the calculation of normal for new polygons. Develop "the preprocessor" to determine the limits of variation of the indent parameter of the bevel. Prepare the code to be sent to SVN, refactor it, delete unused pieces, etc.

Week 12 Report

What you did this week

Currently, planar Bevel works well in all cases. I finished the development of round Bevel. An example of how the Bevel: http://wiki.blender.org/uploads/b/b7/Round_bevel_example.png

But during testing I found that in some examples, the algorithm does not work correctly for non-orthogonal cases. To fix this required a global intervention in what had been done. I rewrote about 2000 lines of code. I seriously revised functions of calculation of the round profiles. functions of search of various intersections and function of generating smoothed surfaces was rewritten. Changed method of calculating additional points. Added the data structures used to construct smooth surfaces. Added rules for handling cases with complex input data in the vertex. Fix bug with duplicate vertices when generating a smoothed surface.

What you plan to do next week

Unfortunately still round Bevel does not always work well. I plan to finish the final tuning of the algorithm, I also finally give you the source code in order and send it to SVN.

Week 13 Report

What you did this week

This week I worked on the code Bevel. I have fixed many small bugs, also I added a lot of checks, this greatly increases the stability of the Bevel.

Many users want to see an option that will automatically limit the allowable size of indenting. I implemented this feature.

I started to write documentation for their work. (http://wiki.blender.org/index.php/User:RedTriangle/Bevel)

What you plan to do next week

The remaining few days I plan to spend on work on my code to the "pencil down". I plan to continue working on documentation and more detailed description of the result.