kicad-library-utils
kicad-library-utils copied to clipboard
KiCad
Handle footprint name as string
If the footprint name contains only numeric characters, it should be converted to string to ensure the proper rules checking.
That is not a valid footprint name for this library. See our footprint naming scheme at https://kicad-pcb.org/libraries/klc/F3.6/. And these check scripts are for those guidelines/rules. So as @cpresser said, this doesn't hurt anything but it's not needed when building footprints that conform to the rules of this library.
I think this is a use full addition anyway. The scripts should not throw a cryptic error message just because the user put in stupid input.
The next logical step is to create a F3.py that actually checks that rule. I just skimmed over the F3 section of KLC, and it seems there is no way a footprint can only contain numbers. So we could check that.
And perhaps throw a warning if its number of characters is <3.
And throw errors if it contains evil code-points (thinking of emojis, german umlauts, ...)
Bad jokes aside, we might have to (and be able to) relax naming rules. See https://github.com/KiCad/kicad-symbols/pull/2938
Nevertheless, we can have a script output a warning if non latin chars are used. A good check would verify that any char is present in the kicad font. Basically an explicit whitelist.
Somewhat off-topic
Footprint-names are also file-names. Yay!
I have no clue how the Pi would do as filename for windows/linux/mac users. With different locales even more complication could arise.
Yes, I'm not opposed to this PR.
OT: True, the π example is a symbol. But all supported platforms can handle multibyte filenames and assuming there have already been users calling their schematics MyCool80'sHeävÿMëtälPCB.sch or something without issue, I guess KiCad's codebase isn't too brittle.
Even more OT: I wonder how many ERP systems these π parts broke along the supply chain ;)
Edit: Not to mention Cyrillic/Greek/Arabic/whatever users in general.