[Plugin] RichSection V0.5.8 UPDATED march 2015

pgarmyn

Hello,
i install the plugin but it doesn't appear any where , i searched in toolbar menu , any idea ?
I am using sketchup 2013.

A new plugin must be activated in the menu 'window/preferences',
a panel opens, select 'extensions' and activate 'RichSection'.

doush-007

thanks a lot

pgarmyn

@micione said:

Report a conflict with "offset.rb."
attached file.

Offset.rb is not a standalone plugin (Can be called by other plugins). It is from Rick Wilson, created in 2004.
What's happening How do you know there is a conflict
What other plugin is using offset.rb ? (Probably anther plugin from Rick Wilson)

TIG

Because your code is all encrypted we don't know what it does.
If it has any 'offset; methods within it that are not encapsulated in your module/class methods, then your 'offset()' method might well clash with the bald scripted 'offet.rb' by RickW, if users have that loading with other scripts.
Your code must not interfere with base class/module/methods or other's scripts that load as helpers - these are things like offset.rb, progressbar.rb and so on... It's easy enough to recast your own versions within your own modules so there is no overlap with other scripts...

pgarmyn

@tig said:

Because your code is all encrypted we don't know what it does.
If it has any 'offset; methods within it that are not encapsulated in your module/class methods, then your 'offset()' method might well clash with the bald scripted 'offet.rb' by RickW, if users have that loading with other scripts.
Your code must not interfere with base class/module/methods or other's scripts that load as helpers - these are things like offset.rb, progressbar.rb and so on... It's easy enough to recast your own versions within your own modules so there is no overlap with other scripts...

About my code:

Every method, variable, ... is declared within module PiGa ; module RSection ; end ; end
No basic classes or modified or extended.
And this is the case for the loader PiGa_RichSection.rb, and the main ruby RichSection.rbs

That wasn't yet good practice in 2004 (Looking in code of offset.rb)
I scanned my code : there is no method 'offset' or anything else named 'offset'

So, if there is something The problem is not there.
Please, micione tell me....
What's happening ? How do you know there is a conflict ?
What other plugin is using offset.rb ? (Probably anther plugin from Rick Wilson)

micione

pgarmyn,

"offset.rb" is used by other tools of Rick Wilson (eg "windowizer").
What happens:
Your tool applies the first section, then it stops, stuck.
Any attempt to use the "move tool" on the section, it causes BugSplat.

Only by removing "offset.rb" your tool starts working again.
In order to identify the conflict, must be eliminated (one by one) all installed plugins. And this is what I did.
I hope I was helpful.

P.S. perhaps seek the advice of Rick Wilson?

pgarmyn

@micione said:

pgarmyn,

"offset.rb" is used by other tools of Rick Wilson (eg "windowizer").....

Thanks
I will test it out. Tomorrow

pgarmyn

@unknownuser said:

A futur hit for sure!
When you say Demo, will be commercial at the end ?

Maybee commercial... . But I will also keep a FREE version .
Until March 2014, everybody can have the full 'DEMO'(Thanks for the correction Bob James)
As 'RichSections' are standard SU-groups, they only need the plugin for update.

TIG

The issue might be that offset.rb [ill advisedly] adds a new method to Array.offset(distance)* this clashes with the Point3d.offset(vector, distance) because an Array of three values [x,y,z] can be offset just as it it were a Point3d, but the native method is expecting a Vector3d [or an Array of three [xv,yv,zv]] and an optional second argument of a distance that overrides the vector's length...

*In my copy of offset.rb I have renamed it as offsetArray(dist) to avoid the clash [it's never used anyway!].

So to avoid clashes with the offset.rb method Array.offset() never use .offset() on point that is expressed as an Array - make that array into a Point3d before the offset is applied, then the clash is avoided.
So it should be
point=Geom::Point3d.new(0,1,2) [or ...new([0,1,2])]
then
point.offset(vector)
or in one step
Geom::Point3d.new([0,1,2]).offset(vector)
BUT NEVER
[0,1,2].offset(vector)
or
pt=[1,2,3] pt.offset(vector)

gbing

Amazing!!
This plugin cover one of the CAD missing function in sketchup.
Now I need to check it in layout.

Thank you pgarmyn!

Elibjr

Wow, this is a amazing! This is starting to make Sketchup look more like a CAD modeler! Well done. I-Render Nxt also renders section cuts similar to Indigo. I will also try this with this and Layout later!

pgarmyn

@tig said:

The issue might be that offset.rb [ill advisedly] adds a new method to Array.offset(distance)* this clashes ...

Thanks TIG and micione, for the analysis

Anyway, I don't think Array.offset is the reason for the incompatibility. As I said, no method in my plugin uses offset.

I just installed Windowizer-3 (W3) beside RichSection (RS) in SU8.
In this version of Windowizer there is no def offset(dist) for class Array.
I confirm Bugsplat when combining both plugins. When using RS or W3,
But after uninstalling RS (also the files), I have also have Bugsplats in W3 for complex models (many nested groups).

Bugsplat in RS (combined with W3) takes always place at the end, after a commit operation,

TO BE CONTINUED....

ArCAD-UK

Impressed so far

Feedback:

Why create two scenes? I appreciate they are different but do you really need two when the change is simply a toggle of the section plane?

RICHSECTION NAME is strange when using custom names e.g. the resulting names aren't as expected

It might be useful to have a simple continuous line in the Projection line styles.

pgarmyn

@arcad-uk said:

Impressed so far

Feedback:.....

Why create two scenes? The first scene is for editing: The camera is not saved, the sectionplan is visible. The second scene (name+p : for parallel view) is for those who really use it (directly) in layout: camera saved, sectionplan not visible. But i can put some swiches in the RichSection settings.

RICHSECTION NAME is strange ...... Dont understand. Is it the 'p' at the end?

It might be useful to have a simple continuous line.... : In what case? Exemple?

Thanks for your interest.

ArCAD-UK

@pgarmyn said:

RICHSECTION NAME is strange ...... Dont understand. Is it the 'p' at the end?

Last night my first section entry [XL01] was replaced with [XL0001] today everything seems ok I may have used "undo" and messed things up.

There is a lot going on when creating an RS, can I do a complete undo with 1 click?

@pgarmyn said:

It might be useful to have a simple continuous line.... : In what case? Exemple?

I was thinking in LO I might want to explode the scene and apply a different line style but I now see that RichSections don't display in vector mode.

jolran

@unknownuser said:

I was thinking in LO I might want to explode the scene and apply a different line style but I now see that RichSections don't display in vector mode.

You do know that you can do this in Hatchfaces ? If you want real edges as hatches from section slice that is.

ArCAD-UK

@jolran said:

You do know that you can do this in Hatchfaces ? If you want real edges as hatches from section slice that is.

I will have a look at that thanks! I'm not a fan of raster hatching...

jolran

Check out the help files, if that does not work for you, you can PM me and fire the questions. Or if you rather post a question in Hatchafaces topic thats fine too.

pgarmyn

@arcad-uk said:

Last night my first section entry [XL01] was replaced with [XL0001] today everything seems ok I may have used "undo" and messed things up.

There is a lot going on when creating an RS, can I do a complete undo with 1 click?

I was thinking in LO I might want to explode the scene and apply a different line style but I now see that RichSections don't display in vector mode. (

Undo : I can't undo scenes and styles loaded (from file). The SU API for Ruby has no functions to create them from scratch. I have to load them from a file. No way, at my knowledge, to include this in the undo DB (within start and commit_operation).

Controle vector output (with color and line style) to Layout : I can only dream of it.

pgarmyn

Hi TIG, micione and others...

As I said before (see 'TO BE CONTINUED'), I loaded the free version (Windowizer3, W3) and did some testing.
As the Bugsplat takes place after a commit_operation in the RichSection plugin (without 'other' errors), I tried to find the conflict in the W3 code.
I didn't find anything directly in common with the RichSection plugin, but I found these anomalies:

def getMaterials is a root method declared twice : once in windowizer,rb, once in getMaterials.rb; Even if the 'require getMaterials,rb' is comment out; the file is automatically loaded because it is directly in the Sketchup/plugins
The following classes are extended : Array, Sketchup::Face, Sketchup::Entity; Sketchup::ArcCurve; Sketchup::Curve
many Global variables
the global variable $wName is once a string [=Wdw.new.to_s.slice(6..14)] and later an empty class object [=Wdw.new]…..
I tried to clean up the code for testing, but I can't see the end of it.

As there is nothing specificly in common with RS (I mean : in common with RS but not with other plugins), I cleaned up the plugin directory and installed only Windowizer3 + Fredo's RoundCorner

It works fine for a (very) small model.
But for more a more complex models (50 identical component-instances: in total (nested) about 9500 edges, 3300 faces), I had directly a Bug-splat when creating round corners around one face within the component.
I uninstalled Windowizer3 (the files) and did the same manipulation : Eventing worked fine.
I think that the plugins of Fredo are a reference for the Sketchup community.

CONCLUSION : Windowizer 3 is not a stable plugin. I didn't spent my money to test the payed version ….

Windowizer is not distributed on Sketchucation, but maybee it is important to tell people that it causes Bugsplats in other plugins.