Infill is not generated when "Ensure vertical shell thickness" is on

[flannelhead]

Version

Version of Slic3r used goes here
git master @ 3fc57ba

Operating system type + version

What OS are you using, and state any version #s
Arch Linux

Behavior

• Describe the problem
  Infill is not generated when the "Ensure vertical shell thickness" option is on.
• Steps needed to reproduce the problem

1. Load STL
2. Select "Ensure vertical shell thickness"
3. Select infill <100 %

• Expected Results
  Infill is expected to be seen in preview.
• Actual Results
  Infill is not seen in preview.

The infill appears again after unchecking "Ensure vertical shell thickness" and changing infill density to trigger regeneration of infill.

Reverting 61d82b0 fixes the issue.

STL/Config (.ZIP) where problem occurs

Upload a zipped copy of an STL and your config (File -> Export Config)
config.zip
STL: http://www.thingiverse.com/download:648838

[bubnikv]

Sorry, I cannot reproduce the problem with g3fc57ba, built from scratch on a build server for win64.
Maybe you need to do a fresh rebuild? Or is the problem specific to some sequence of steps? Would you please provide screenshots?

[bubnikv]

By the way, ensure vertical wall thickness is a competing feature to the extra perimeters when needed. I think of the two as exclusive or.

[flannelhead]

@bubnikv Good to know that, thanks.

I also built the mentioned revision on Windows 64 from scratch. There I couldn't reproduce the issue no matter how hard I tried..

On Linux I also built from scratch and even created an entirely new profile. Still I could reliably reproduce the issue.

Here's one screenshot that might be of interest. It seems the solid infill between the perimeter and sparse infill is generated.

[bubnikv]

Thanks, I will try with Linux then.

On Fri, Nov 4, 2016 at 3:26 PM, Sakari Kapanen [email protected]
wrote:

@bubnikv https://github.com/bubnikv Good to know that, thanks.

I also built the mentioned revision on Windows 64 from scratch. There I
couldn't reproduce the issue no matter how hard I tried..

On Linux I also built from scratch and even created an entirely new
profile. Still I could reliably reproduce the issue.

Here's one screenshot that might be of interest. It seems the solid infill
between the perimeter and sparse infill is generated.
[image: 2016-11-04-162336_1366x768_scrot]
https://cloud.githubusercontent.com/assets/5001906/20008990/5cd3ac38-a2ab-11e6-8bb5-b8c000e2830a.png

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
#14 (comment), or mute
the thread
https://github.com/notifications/unsubscribe-auth/AFj5I5cjLxlG4OI5RfGNo5f_Yf0DAYRqks5q60CNgaJpZM4KpSU-
.

[flannelhead]

Still testing further. It appears that compiling with C++98 standard enabled (env CFLAGS='-std=c++98' perl Build.PL) the issue doesn't exist. So it might be something related to the CPPVER >= 11 code.

[flannelhead]

Ok, the traces seem to lead to this method. Disabling it seems to resolve the issue.

[bubnikv]

Thanks a lot! Thinking about it, this construct does not make sense and the compiler has no way to know about the ownership of the pointers stored in the array. The correct solution is simply to delete this function. Thanks again.

[bubnikv]

Fixed, thanks to @flannelhead
89cf290