Add optional custom pinnumbers to connector

[kimmoli]

One possible solution for custom pin numbering, e.g. fits needs in #4

• Pincount is detected automatically if not given.
• If both, pinout and pinnumbers, are given, their count must match.
• Freedom to give pin numbers in any order, in alphabetical and mixed notation.

Sample

connectors:
  X1:
    type: D-Sub
    subtype: male
    pincount: 25
    pinnumbers: [1,14,3,16,5,18,7,20,9,22,11,24,13]
    pinout: [ SENSE_P_1, SENSE_N_1, SENSE_P_2, SENSE_N_2, SENSE_P_3, SENSE_N_3, SENSE_P_4,SENSE_N_4, SENSE_P_5, SENSE_N_5, SENSE_P_6, SENSE_N_6, GND ]
  X2:
    type: F48
    subtype: female
    pincount: 48
    pinnumbers: [ z2,b2,d2,z4,b4,d4,z6,b6,d6,z8,b8,d8,z10,b10,d10,z12,b12,d12,z14,b14,d14,z16,b16,d16,z18,b18,d18,z20,b20,d20,z22,b22,d22,z24,b24,d24,z26,b26,d26,z28,b28,d28,z30,b30,d30,z32,b32,d32 ]

cables:
  W1:
    gauge: 0.25 mm2
    length: 0.2
    color_code: DIN
    wirecount: 16
    shield: true

connections:
  -
    - X1: [1,14,3,16,5,18,7,20,9,22,11,24]
    - W1: [2,1,4,3,6,5,8,7,10,9,12,11]
    - X2: [d4,z2,d10,z8,d16,z14,d20,z18,d26,z24,d32,z30]
  -
    - X1: 13
    - W1: s

[formatc1702]

This is a fantastic addition, I'll take care of it in the next few days!

[formatc1702]

Unfortunately, this breaks the minimalist way of specifying only a pincount (e.g. Tutorial 01)... fixed in dev branch 8bf296b

[formatc1702]

Additional idea: Perhaps, instead of having pinout vs pinnumbers, using pin_labels or pin_IDs vs pin_descriptions, pin_functions or similar might be less confusing/ambiguous? Not sure if pinnumbers works when it's not only numerical labels...

[kimmoli]

Thanks for your efforts on this project.

Additional idea: Perhaps, instead of having pinout vs pinnumbers, using pin_labels or pin_IDs vs pin_descriptions, pin_functions or similar might be less confusing/ambiguous? Not sure if pinnumbers works when it's not only numerical labels...

IMHO pinnumber is pin number, regardles is it number, letter or mixed.

pinout could be pinname

i would avoid underscores ...

[formatc1702]

pinnumber and pinname could work. I totally agree on the underscore thing. Thanks for helping out!

[kvid]

IMHO pinnumbers might be confusing when the values might be non-numeric. Why not use pins as suggested in issue#4 if these new pin identifiers will be the only way to reference pins? I would also consider a future possability to reference pins in alternative ways (in addition to the plain identifier - numeric or not), like #3 for the third pin and $GND for the pin named GND in the pinout, etc., but this should be raised as a separate issue. The pinIDs and pinlabels alternatives can also be considered without the underscore.

[kvid]

I have not tested the proposed code to see if it is possible to specify s as pin identifier and how then a conflict with the shield identifier is handeled. Is this an issue?