Review Request 118785: Segment Axis object in KIG ( GSoC 14 )

[David Narváez]

> On June 18, 2014, 11:35 a.m., Maurizio Paolini wrote:
> > The old segment_axis macro required as argument a segment, whereas the new construction
> > depends on two points... this could lead to some confusion, and actually, the name
> > SegmentAxisABType is a little bit incorrect.
> > A user that is accustomed to the old segment_axis would find the use of the new 
> > SegmentAxisABType different.
> > Question: what is the reason to add this new construction?
> 
> Aniket Anvit wrote:
>     The macro one could only construct the perpendicular bisector for segments. Since the object one takes two points as inputs, it can construct the perpendicular-bisector of any two points ( even if there is no segment constructed with them ). Also, perpendicular-bisector is an important constructionso we should have it directly implemented as an object rather than as a built-in macro. It will be better...
> 
> Maurizio Paolini wrote:
>     I am aware of that, however changing the parents *will* change how the user interacts with kig.  Moreover, there are situations
>     (in particular when constructing a new macro) where it is important for the user to understand exactly what is the hierarchy of
>     the constructed object, although in this particular case letting the axis depend on the two points is probably better.
>     
>     Please, let me clarify that I am *not* contrary to the new Object construction!
>     
>     Anyway, let me make a proposal: the creation of a "special constructor" should allow the user to construct the "perpendicular bisector"
>     in both ways:
>     
>     - either by clicking on the two points (then the new object is directly used)
>     - or by clicking on the segment, in which case "kig" should construct the two end-points and then use the new object
>     
>     I guess that the present implementation of "mid-point" is exactly of this sort!

I only read these comments today, sorry about that. Maurizio is right about not changing the behavior, and the last suggestion about making it like mid point is what we need. I'll work on that fix now.


- David


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/118785/#review60377
-----------------------------------------------------------


On June 19, 2014, 5:43 a.m., Aniket Anvit wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/118785/
> -----------------------------------------------------------
> 
> (Updated June 19, 2014, 5:43 a.m.)
> 
> 
> Review request for KDE Edu and David Narváez.
> 
> 
> Repository: kig
> 
> 
> Description
> -------
> 
> This patch adds the segment axis object to KIG. This new object will take two points as input and generates a line which would perpendicularly bisect the segment constructed with those two points.
> 
> 
> Diffs
> -----
> 
>   macros/CMakeLists.txt 8abf249 
>   macros/segment_axis.kigt e86b90c 
>   misc/builtin_stuff.cc 28846c8 
>   objects/line_type.h 3ea252b 
>   objects/line_type.cc dbf575c 
> 
> Diff: https://git.reviewboard.kde.org/r/118785/diff/
> 
> 
> Testing
> -------
> 
> Took 6 points and constructed the perpendicular-bisectors choosing random pair. Tested with semgents too.
> 
> 
> Thanks,
> 
> Aniket Anvit
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kde-edu/attachments/20140624/fbbcff3d/attachment.html>