[QUESTION] CPU name logic

Is the following code block the entire logic in bpytop determining the CPU name that gets displayed?

I’m just curious at this point because I have a CPU that reports as “Intel® Core™2 Duo CPU” and bpytop just ends up displaying “Duo”, which is not that helpful since it could also refer to a Core Duo CPU.

I’m wondering if anyone out there has already built a function to more reliably return a useful and accurate (but shortened) CPU name that would better suit the bpytop use case. Failing that, maybe I can add to the filtering logic here. For instance, filtering out “Processor” from “ARMv7 Processor rev 3 (v7l)” on the Raspberry Pi.

This function did pretty good with a Ryzen 3600X processor, but maybe that was just because it called itself “Ryzen 3600X”. I didn’t check.

def get_cpu_name() -> str:
	'''Fetch a suitable CPU identifier from the CPU model name string'''
	name: str = ""
	nlist: List = []
	command: str = ""
	cmd_out: str = ""
	rem_line: str = ""
	if SYSTEM == "Linux":
		command = "cat /proc/cpuinfo"
		rem_line = "model name"
	elif SYSTEM == "MacOS":
		command ="sysctl -n machdep.cpu.brand_string"
	elif SYSTEM == "BSD":
		command ="sysctl hw.model"
		rem_line = "hw.model"


	try:
		cmd_out = subprocess.check_output("LANG=C " + command, shell=True, universal_newlines=True)
	except:
		pass
	if rem_line:
		for line in cmd_out.split("\n"):
			if rem_line in line:
				name = re.sub( ".*" + rem_line + ".*:", "", line,1).lstrip()
	else:
		name = cmd_out
	nlist = name.split(" ")
	if "Xeon" in name and "CPU" in name:
		name = nlist[nlist.index("CPU")+1]
	elif "Ryzen" in name:
		name = " ".join(nlist[nlist.index("Ryzen"):nlist.index("Ryzen")+3])
	elif "CPU" in name and not nlist[0] == "CPU":
		name = nlist[nlist.index("CPU")-1]


	return name

[aristocratos]

Added more detection logic in v1.0.17

[aristocratos]

@viv-4 Fixed in v1.0.32