Compatibility

[JLundberg65]

Hello,
I really like the feasibility of your slicing routine as a hatch fill. I have a project I'm playing with that's in Visual Studio C and C++. Since I'm not using Ubuntu and its dependencies, is it possible to make a version that would compile in Visual Studio?

John.

[CaseySanchez]

It ought to sans Cairo and the main.cpp program provided; the core Mandoline library itself should compile so long as you can manage to link your program to Eigen, which can be found here: https://eigen.tuxfamily.org/index.php?title=Main_Page

Note that I have not tried to build on Windows, but I am certain it should.

[JLundberg65]

Casey, I appreciate your reply. Thank you. Just a couple of questions if you wouldn’t mind. How fast would you say your process is? Let’s say this was passed 30 small objects about .75” height and .25” width over a 10” x 10” page and the line spacing was about .010” How difficult would it be to set a main hatch angle of say, 20deg, 45deg or 90deg? Of course all id mute if I can’t get it and Eigen to compile in Visual Studio 2013 Thanks again, John From: Casey Sanchez [mailto:notifications@github.com] Sent: Saturday, February 20, 2021 8:19 AM To: Tannz0rz/Mandoline Cc: JLundberg65; Author Subject: Re: [Tannz0rz/Mandoline] Compatibility (#1) It ought to sans Cairo and the main.cpp program provided; Mandoline itself should compile so long as you can manage to link your program to Eigen, which can be found here: https://eigen.tuxfamily.org/index.php?title=Main_Page Note that I have not tried to build on Windows, but I am certain it should. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#1 (comment)> , or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAJGZM6IYAOGH7WN2WNWS7DS77AFNANCNFSM4W5HVZGQ> . <https://github.com/notifications/beacon/AAJGZM35R6DOYEQI6X6FMQTS77AFNA5CNFSM4W5HVZG2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOF2TMLDQ.gif>

[CaseySanchez]

By "height" and "width" are you referring to two dimensions? This is dependent upon the complexity of the objects, increasing calculation time given the number of total sides to be compared against. When I get the chance I'd be happy to create & slice 30 20-sided polygons embedded in a rectangle at those specifications and report to you my findings.

EDIT: Also, changing the angle is simply a matter of applying an affine transformation via the "Transform" heuristic to the "Graph" to be sliced.

[JLundberg65]

Correct, 2 dimensional objects. Will it also work with Bezier curves as well? Here are some objects in my thought process. Others could be objects like your example. How long did that take to process? Not actually draw but process the job in memory? The arcs could be Bezier or made from vectors. The time it takes to fill the inside of 30 of these be great. If a box was put around the extents of the word and the box was filled leaving the text hollow would be another type of object. I appreciate you offering to run a test. If the calculations of a single object, “T” takes a while when multiplied out, maybe start up different threads to handle multiple objects simultaneously. Just rambling J The objects XY Dim are say .250, .750 Thanks again. Maybe we could email privately? Take care and stay safe. . From: Casey Sanchez [mailto:notifications@github.com] Sent: Saturday, February 20, 2021 1:51 PM To: Tannz0rz/Mandoline Cc: JLundberg65; Author Subject: Re: [Tannz0rz/Mandoline] Compatibility (#1) By "height" and "width" are you referring to two dimensions? This is dependent upon the complexity of the objects, increasing calculation time given the number of total sides to be compared against. When I get the chance I'd be happy to create & slice 30 20-sided polygons embedded in a rectangle at those specifications and report to you my findings. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#1 (comment)> , or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAJGZM2DIECYNLAWMHBULY3TAAHBHANCNFSM4W5HVZGQ> . <https://github.com/notifications/beacon/AAJGZM4SZGHKRU7JWIU6X6DTAAHBHA5CNFSM4W5HVZG2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOF2T2SMI.gif>

[CaseySanchez]

Certainly, please contact me at [redacted]

https://i.imgur.com/Kk5hQ7F.png

This took 1283ms, which certainly is not the most performant, nor does it create an optimal path. It could stand to be optimized in various aspects I'm sure. Bear in mind this was designed for primitive robot path planning.

Mandoline works by utilizing the even-odd rule (https://en.wikipedia.org/wiki/Even%E2%80%93odd_rule), which requires a set of line segments. It may be possible to approximate Bezier curves using line segments to your desired precision to achieve a similar effect.