Gimp-Forum.net
ofn-dial-marks feature request? - Printable Version

+- Gimp-Forum.net (https://www.gimp-forum.net)
+-- Forum: GIMP (https://www.gimp-forum.net/Forum-GIMP)
+--- Forum: Extending the GIMP (https://www.gimp-forum.net/Forum-Extending-the-GIMP)
+---- Forum: Scripting questions (https://www.gimp-forum.net/Forum-Scripting-questions)
+---- Thread: ofn-dial-marks feature request? (/Thread-ofn-dial-marks-feature-request)



ofn-dial-marks feature request? - eepjr24 - 11-27-2020

This is not a huge deal since you can accomplish it in a variety of other ways, it would just remove a step. I looked at the code to see if it would be trivial to accomplish but I am not familiar enough with either Python or paths to dare to tinker. =P 

Could check boxes be added to remove the inner or outer curve and leave the dial marks?  This would be useful for some types of dials like a speedometer, etc.

Trying to read the python, it seemed to me maybe it could be done with a conditional in the range loop on marksCount? I don't know how to tell which section of the path is being drawn though, so back to having to tinker and not sure what the effects of that would be. If I am way off base, just tell me to pack sand. 

- E


RE: ofn-dial-marks feature request? - rich2005 - 11-27-2020

Quote:Could check boxes be added to remove the inner or outer curve and leave the dial marks?

Just a clarification please. Is it just the curve that is the problem ? Do you want to use a regular rectangle ?

There is a script that will rotate a path (any shape)  https://i.imgur.com/x2rR7RR.jpg In the attachment here: https://www.gimp-forum.net/Thread-Making-a-watch-face

Alternative, ofnuts plugin ofn-path-to-shape.py Two stages, make some spokes / adjust the spokes https://i.imgur.com/6obS8lR.jpg


RE: ofn-dial-marks feature request? - eepjr24 - 11-27-2020

I'll use the speedometer as an example here.

[attachment=5242]

For this I used 5 ofn-dial-marks, one each to create the blue and red marks and 3 more to subtract the out the curves. If I had been more awake this morning, I could have done it with 2 subtractions and an offset to not erase marks but it was too early for degree maths. 

With the ability to uncheck inner or outer curves from the path output this would take 2 uses of ofn-dial-marks.

Either way there would still be a bit of clean up where the marks intersected or where the I did not want the smaller red marks on the bottom. The first I think Ofnuts has a tool to accomplish (subtracting the contents of one path from another) and the second is really just one lasso selection and delete.

None of this is really difficult, I just thought it was a nice feature to add if it was not difficult and something that Ofnuts wanted to work on. Who knows, in 6 months or a year I might even be familiar enough with python to give it a whirl. =)

PS: Please ignore the fact that my car is run by someone who uses base 12 instead of base 10, my degree maths were even worse than I thought this morning.

- Ernie


RE: ofn-dial-marks feature request? - Ofnuts - 11-27-2020

I don't quite get what you are removing. A dial mark as generated is a polygon, if you remove a side Gimp will still add one to close the mark if you use is to create a selection. Can you post the image you have before you start to remove things?


RE: ofn-dial-marks feature request? - eepjr24 - 11-27-2020

(11-27-2020, 06:06 PM)Ofnuts Wrote: I don't quite get what you are removing. A dial mark as generated is a polygon, if you remove a side Gimp will still add one to close the mark if you use is to create a selection.  Can you post the image you have before you start to remove things?

I think that the way you responded may have answered my question. So if they are generated as polygons then the lines are shape sides. Okay. I was thinking that the dial paths were generated as a pair of circles and then the joining lines were added by connecting coordinate pairs. 

Anyway, I will go ahead and illustrate because I could be wrong above (happens a lot) and maybe I will learn something else in the process. 

The 3 paths I created to mark the dial (See earlier post for the layer labels):

[attachment=5247]

The first "subtraction" path (remove innermost circle):

[attachment=5248]

The second "subtraction" path (remove outermost circle):

[attachment=5249]

The third "subtraction" path (remove innermost circle):

[attachment=5250]

I hope at least what I meant is making more sense now, even if I did have a fundamental misunderstanding of how you were generating the marks in the first place. 

- E


RE: ofn-dial-marks feature request? - Ofnuts - 11-28-2020

The marks in ofn-dial-marks are assumed to have a significant thickness/area, when what you are producing are really thin lines. So it looks more like a job for the "spokes" function in my ofn-path-to-shape (see the clock dial example in the doc). Otherwise you can also produce thin tick marks by given them a very narrow angle (here .5° for the short ticks and 1° for the long ones):

As paths:
[attachment=5256]

Rendered:
[attachment=5255]



RE: ofn-dial-marks feature request? - eepjr24 - 11-28-2020

(11-28-2020, 12:20 AM)Ofnuts Wrote: The marks in ofn-dial-marks are assumed to have a significant thickness/area, when what you are producing are really thin lines. So it looks more like a job for the "spokes" function in my ofn-path-to-shape (see the clock dial example in the doc). Otherwise you can also produce thin tick marks by given them a very narrow angle (here .5° for the short ticks and 1° for the long ones):
<snip>
Ahhhh, yes, using the right tool for the job will probably make things much easier! Thanks again, adding another tool to my toolbox. I think at this point I need to just download all your scripts and spend a few days playing with them just to get to know all my options. Thank you for creating a great set of tools.

- Ernie