Comparison of pos setting behaviours by PixPos() & tcas_main() function

面麻

Right as what the title shows, I found the differences of position setting behaviors between PixPos() and tcas_main() function.
We could found the original codes of PixPos() function at tcaxPy.py, which has been listed below:

1. def PixPos(PIX, x, y):
   
2.     return ((x, y), PIX[1], PIX[2])

复制代码

It returns a PIX structure with a change of position info only.
About tcas_main() function, I could merely provide its parameter info by its definition:

1. def tcas_main(TCAS_BUF, pix, start, end, offsetX, offsetY, layer):

复制代码

offsetX and offsetY stand for the offsets in the horizontal and vertical direction respectively.
At first, I estimated that their effects would be the same, but a different result is showed by the preview.
The test script is (Note: When using PixPos() function, we must set offsetX and offsetY as 0 in tcas_main()):

1. from tcaxPy import *
   
2. 
   
3. 
   
4. def tcaxPy_Init():
   
5.     global font
   
6.     global font_out
   
7.     global font_size
   
8.     font_file   = GetVal(val_FontFileName)
   
9.     face_id     = GetVal(val_FaceID)
   
10.     font_size   = GetVal(val_FontSize)
    
11.     spacing     = GetVal(val_Spacing)
    
12.     space_scale = GetVal(val_SpaceScale)
    
13.     font = InitFont(font_file, face_id, font_size, spacing, space_scale, 0xFFFFFF, 0, 0)
    
14.     font_out = InitFont(font_file, face_id, font_size, spacing, space_scale, 0xFF0000, 2, 1)
    
15. 
    
16. 
    
17. def tcaxPy_Fin():
    
18.     FinFont(font)
    
19.     FinFont(font_out)
    
20. 
    
21. 
    
22. def tcaxPy_Main(_i, _j, _n, _start, _end, _elapk, _k, _x, _y, _a, _txt):
    
23. 
    
24.     TCAS_BUF = []        # used for saving TCAS FX raw data
    
25. 
    
26.     # an5 -> an7, TCAS uses an7
    
27.     ux = _x - int(_a / 2 + 0.5)
    
28.     uy = _y - int(font_size / 2 + 0.5)
    
29. 
    
30.     text = TextPix(font, _txt)
    
31.     bord = TextPix(font_out, _txt)
    
32.     bord = PixBlur(bord, 2)
    
33.     main = CombinePixs(bord, text)
    
34.     #main = PixPos(main, ux, uy)
    
35. 
    
36.     ts = 10 * _start
    
37.     te = 10 * _end
    
38. 
    
39.     tcas_main(TCAS_BUF, main, ts, te, ux, uy, 0)
    
40.     #tcas_main(TCAS_BUF, main, ts, te, 0, 0, 0)
    
41. 
    
42.     return (None, TCAS_BUF)

复制代码

We could find the difference clearly from the samples below:
pixpos():


tcas_main():


I prefer the effect of position setting behavior by tcas_main().
It seems that the effect of PixPos() function has a little problem.

Looking forward to your answer and explanation.

milkyjing

I think it's the misunderstanding of PixPos, check the source code, find the difference between PixPos and PixPosShift (the behavior of which the tcas_main is like).

面麻

milkyjing 发表于 2016-5-2 11:51
I think it's the misunderstanding of PixPos, check the source code, find the difference between PixP ...

Reference of function definition in tcaxPy.py:

1. def PixPos(PIX, x, y):
   
2.     return ((x, y), PIX[1], PIX[2])
   
3. 
   
4. def PixPosShift(PIX, dx, dy):
   
5.     x = PIX[0][0]
   
6.     y = PIX[0][1]
   
7.     return ((x + dx, y + dy), PIX[1], PIX[2])

复制代码

Still, I could not grasp my wrong.
Do you mean that in this script PixPos() ought to be replaced with PixPosShift()?
However, I see the parameters x&y in PixPos() and offsetX&offsetY in tcas_main() as the same.

milkyjing

every PIX object has its initial position specified in PIX[0], (a tuple of two dimensional x, y pair)

PixPos resets the initial position by (x, y), while PixPosShift added to it

http://www.tcax.org/forum.php?mo ... =1079&fromuid=2

面麻

Thank you for replying me during your busy working.
This confusion counts a lot for me.
After PixPos()'s being replaced with PixPosShift(), the positions have become normal in preview.