Skip to content

A hardware monitoring project that provides high-performance tools, Python libraries, and C++ header files for retrieving and displaying detailed hardware information.

License

MIT, GPL-3.0 licenses found

Licenses found

MIT
LICENSE
GPL-3.0
LICENSE.GPL3
Notifications You must be signed in to change notification settings

gafoo173/HardView

PyPI Stable Version Version Supported Python versions License License Platforms

Temperature Monitoring

HardView Logo

HardView - Hardware Information Project

HardView is a project that includes Python, C++, and C libraries, Windows drivers, and tools for monitoring hardware and displaying its information through various sources, whether from the system or other libraries. It provides a unified interface for developers to access information via libraries and a user interface for end-users through the tools.

HardView Project Overview

This table summarizes the libraries and tools included in the HardView project, describing their purpose, language, and features.

Library Name Description Language Purpose / Features
HardView The core library providing static hardware information for Windows and Linux. Supports JSON or Python objects in the Python version. C Main library for retrieving static system information.
LiveView A monitoring library for both static hardware info and real-time data. Supports CPU temperature and regular usage on Windows and Linux C++ Real-time monitoring of hardware metrics, integrates static info and CPUID functions.
HardwareWrapper An internal library wrapping LibreHardwareMonitorLib with simple functions through C++/CLI, allowing use from C++. Primarily used by LiveView on Windows for temperature readings. C++/CLI Simplifies access to LibreHardwareMonitorLib, providing easy C++ usage for Windows sensor data.
cpuid An internal, header-only C++ library providing easy helper functions to access most CPUID information. Used by LiveView for CPUID-related functionality. C++ Lightweight, easy-to-integrate CPUID helper library for detailed processor information.
C++/Headers A folder containing header-only C++ libraries like SMART.hpp (for SMART info) or Live.hpp (C++ header-only version of LiveView), and others. C++ Header-only C++ modules for advanced hardware access and monitoring.
Drivers A set of Windows kernel drivers granting access to low-level hardware functionality useful for monitoring. Each driver comes with a header-only C++ library for easier integration. These drivers are not used by the main HardView libraries (Python or C++) since they are unsigned. They are provided for those who wish to sign and use them, or for personal use with local build and test signing. C/C++ Optional drivers for advanced hardware access under Windows. Not required for standard HardView usage.
Tools A collection of CLI and GUI Python tools that rely on HardView to display hardware information. Python Command-line and GUI utilities for interacting with hardware info provided by HardView.

Key Features

  • Comprehensive Hardware Data: BIOS, System, Baseboard, Chassis, CPU, RAM, Disks, Network,GPU.
  • Advanced Storage & SMART: Detailed disk, partition, and SMART attributes.
  • Performance Monitoring: Real-time and interval-based CPU, RAM, and system performance.
  • C/C++ Implementation: High performance native code.
  • Python Integration: Easy-to-use Python API.
  • Temperature & Sensors Monitoring:(3.1.0+) Retrieve real-time temperature, voltage, and fan speed readings from system sensors.
    Windows: Uses LibreHardwareMonitor for comprehensive sensor data via native integration.
    Linux: Uses lm-sensors for direct hardware monitoring.
    Licenses: LibreHardwareMonitor — MPL-2.0 (licenses/LICENSE.MPL-2.txt), lm-sensors — LGPL-2.1-or-later (licenses/LICENSE.LGPL-2.1.txt)

Installation (Python)

From PyPI

pip install hardview

From Source

git clone https://github.com/gafoo173/hardview.git
cd hardview
pip install .

For supported platforms and full setup instructions, see docs/INSTALL.md.


Dependencies

Python (Windows):
Requires LibreHardwareMonitorLib.dll and HidSharp.dll.
These DLLs are included in the package, so no separate installation is needed.

Windows Temperature Information Features

The temperature information features in Windows specifically require the MSVC Runtime, namely the following DLLs on 64-bit systems:

  • msvcp140.dll
  • vcruntime140.dll
  • vcruntime140_1.dll

If you place these DLLs alongside HardwareWrapper.dll, the temperature-related functions will likely work properly even if you haven't installed the full MSVC runtime.
(This applies whether you are using the Python LiveView or the HardwareTemp.dll from the SDK; in all cases, these libraries are required.) In HardView Python versions 3.2.0+, these DLLs are already included alongside the package, so you don’t need to place them manually.)

Python (Linux):
Requires the lm-sensors library to be installed for hardware monitoring.

C++ Libraries:
Check the top of each library header file for listed dependencies.
Most libraries have no external dependencies.
Exception: SPD.hpp requires InpOutx64.dll.
It is recommended to review the header file beginning for any dependency notes.


Potential Issues on Windows

HardView.LiveView Temperature Features

The temperature monitoring features in HardView.LiveView rely on LibreHardwareMonitorLib, which in turn uses on WinRing0.
WinRing0 is an old and well-known driver used for reading from MSR, physical memory, and other low-level hardware resources.

The Problem

Unfortunately, WinRing0 is now blocked by Windows.
This means you may encounter alerts from Windows Defender similar to this one:

Windows Defender Alert

You’ll notice that python.sys is reported as the suspicious driver.
This has nothing to do with Python itself. What actually happens is:

  • LibreHardwareMonitorLib, when creating its driver, names it as <ProgramName>.sys.
  • Since Python scripts run under python.exe, the driver ends up named python.sys.
  • In reality, this file is just the WinRing0 driver renamed.

Is It Dangerous?

  • WinRing0 is just a driver that grants access to resources that normally require kernel-mode from user-mode. The danger only arises if a malicious program abuses it.
  • The driver created by LibreHardwareMonitorLib is temporary. It will be stop automatically when you restart your system.

It’s recommended to delete the driver using sc delete or remove its file after the program finishes, to prevent any malicious software from exploiting it.

Manually Stopping or Removing the Driver

If you want to close or remove the driver manually after running your script/program:

  1. Open CMD as Administrator.
  2. Run the following command to stop the driver:

sc stop R0<ProgramName>

  • For Python scripts:
    sc stop R0Python
    
  • For an executable program named X.exe:
    sc stop R0X
    
  1. To permanently delete the driver, run:
sc delete R0Python


Usage Examples

HardView (Not recommended for monitoring in 3.1.0+. It's better to use LiveView)

import HardView
import json

# JSON output
bios_json = HardView.get_bios_info()
cpu_json = HardView.get_cpu_info() #In Linux all outputs N/A in this function 

# Python objects output
#You must pass the parameter `false` in versions prior to 3.0.3, e.g. `HardView.get_bios_info_objects(false)`.

bios_objects = HardView.get_bios_info_objects() 
cpu_objects = HardView.get_cpu_info_objects() #On Linux, all outputs of this function show N/A It is recommended in 3.1.0+ to use the cpuid function from LiveView.PyLiveCPU.

# Performance monitoring
cpu_usage_json = HardView.get_cpu_usage()
ram_usage_objects = HardView.get_ram_usage_objects()

# Monitor over time
cpu_monitor_json = HardView.monitor_cpu_usage_duration(5, 1000)
ram_monitor_objects = HardView.monitor_ram_usage_duration_objects(3, 500) 

# Pretty print CPU info
import pprint
pprint.pprint(json.loads(cpu_json))

LiveView

from HardView.LiveView import PyLiveCPU, PyLiveRam, PyLiveDisk, PyLiveNetwork
import time

# Initialize system monitors
cpu_monitor = PyLiveCPU()         # CPU usage monitor
ram_monitor = PyLiveRam()         # RAM usage monitor
disk_monitor = PyLiveDisk(mode=1) # Disk R/W speed monitor (mode 1 for MB/s)
net_monitor = PyLiveNetwork()     # Network traffic monitor

print("System Monitor - Single Reading")
print("-" * 40)

# Get system metrics with 1-second sampling interval
cpu_usage = cpu_monitor.get_usage(1000)           # CPU percentage
ram_usage = ram_monitor.get_usage()               # RAM percentage
disk_rw = disk_monitor.get_usage(1000)            # Returns [(Read MB/s), (Write MB/s)]
net_traffic = net_monitor.get_usage(1000, mode=0) # Total network MB/s

# Display current system status
print(f"CPU: {cpu_usage:5.1f}% | RAM: {ram_usage:5.1f}% | "
      f"Disk R/W: {disk_rw[0][1]:4.1f}/{disk_rw[1][1]:4.1f} MB/s | "
      f"Network: {net_traffic:6.3f} MB/s")
      
print("Monitoring complete.")

LiveView (temperature) - Requires admin privileges

#!/usr/bin/env python3
import sys

# Check CPU temperature - single reading
if sys.platform == "win32":
    # Windows CPU temperature
    try:
        from HardView.LiveView import PyTempCpu
        cpu_temp = PyTempCpu()  # Auto-initialize
        temperature = cpu_temp.get_temp()
        print(f"CPU Temperature: {temperature:.1f}°C")
    except Exception as e:
        print(f"Windows temperature error: {e}")
        
elif sys.platform == "linux":
    # Linux CPU temperature  
    try:
        from HardView.LiveView import PyLinuxSensor
        sensor = PyLinuxSensor()
        temperature = sensor.getCpuTemp()
        if temperature > 0:
            print(f"CPU Temperature: {temperature:.1f}°C")
        else:
            print("CPU temperature not available")
    except Exception as e:
        print(f"Linux temperature error: {e}")
        
else:
    print("Unsupported platform")

SDK Temperature (Rust) - Requires admin privileges

//This code will work on Windows only.
use libloading::{Library, Symbol};
use std::os::raw::{c_double, c_int};

type InitFn = unsafe extern "C" fn() -> c_int;
type ShutdownFn = unsafe extern "C" fn();
type GetTempFn = unsafe extern "C" fn() -> c_double;
type UpdateFn = unsafe extern "C" fn();

// Check if required DLLs exist next to the executable
fn check_dependencies() -> Result<(), String> {
    let required_dlls = ["HardwareTemp.dll", "HardwareWrapper.dll", "LibreHardwareMonitorLib.dll", "HidSharp.dll"];
    
    let exe_dir = std::env::current_exe()
        .map_err(|e| format!("Failed to get executable path: {}", e))?
        .parent()
        .ok_or("Failed to get executable directory")?
        .to_owned();

    let mut missing = Vec::new();
    for dll in &required_dlls {
        if !exe_dir.join(dll).exists() {
            missing.push(*dll);
        }
    }

    if !missing.is_empty() {
        return Err(format!("Missing DLLs: {}", missing.join(", ")));
    }
    Ok(())
}

fn main() {
    // Check dependencies first
    if let Err(error) = check_dependencies() {
        eprintln!("Error: {}", error);
        return;
    }

    // Load the library from executable directory
    let exe_dir = std::env::current_exe().unwrap().parent().unwrap().to_owned();
    let dll_path = exe_dir.join("HardwareTemp.dll");
    
    let lib = unsafe { 
        Library::new(&dll_path).expect("Failed to load HardwareTemp.dll") 
    };

    unsafe {
        // Load required functions
        let init: Symbol<InitFn> = lib.get(b"InitHardwareTempMonitor\0").expect("InitHardwareTempMonitor not found");
        let get_cpu_temp: Symbol<GetTempFn> = lib.get(b"GetCpuTemperatureTemp\0").expect("GetCpuTemperatureTemp not found");
        let update: Symbol<UpdateFn> = lib.get(b"UpdateHardwareMonitorTemp\0").expect("UpdateHardwareMonitorTemp not found");
        let shutdown: Symbol<ShutdownFn> = lib.get(b"ShutdownHardwareTempMonitor\0").expect("ShutdownHardwareTempMonitor not found");

        // Initialize hardware monitor
        let init_result = init();
        if init_result != 0 {
            eprintln!("Failed to initialize hardware monitor. Error code: {}", init_result);
            return;
        }

        // Update and get CPU temperature
        update();
        let cpu_temp = get_cpu_temp();

        // Display result
        match cpu_temp {
            -1.0 => println!("CPU Temperature: ERROR - Run as Administrator or sensor not supported"),
            -99.0 => println!("CPU Temperature: ERROR - Missing dependencies"),
            temp => println!("CPU Temperature: {:.1} °C", temp),
        }

        // Cleanup
        shutdown();
    }
}

MSR.hpp Example (C++) - needs MsrDrv.sys installed and running

//this code will work in intel only
#include "MSR.hpp"
#include <iostream>

int main() {
    try {
        // Create MSR driver instance
        MSR::MsrDriver driver;

        if (!driver.IsValid()) {
            std::cerr << "MSR driver not available!" << std::endl;
            return 1;
        }

        // Read CPU TjMax (maximum temperature)
        int tjMax = 0;
        if (MSR::Thermal::TryGetTjMax(driver, tjMax)) {
            std::cout << "CPU TjMax: " << tjMax << "°C" << std::endl;
        } else {
            std::cerr << "Failed to get TjMax." << std::endl;
        }

        // Read current CPU temperature
        int currentTemp = 0;
        if (MSR::Thermal::TryGetCurrentTemperature(driver, currentTemp)) {
            std::cout << "Current CPU Temperature: " << currentTemp << "°C" << std::endl;
        } else {
            std::cerr << "Failed to read CPU temperature." << std::endl;
        }

        // Read a specific MSR register (IA32_PLATFORM_ID)
        try {
            UINT64 platformId = driver.ReadMsr(MSR::Registers::IA32_PLATFORM_ID);
            std::cout << "IA32_PLATFORM_ID MSR: 0x" 
                      << std::hex << platformId << std::dec << std::endl;
        } catch (const MSR::MsrException& ex) {
            std::cerr << "Error reading MSR: " << ex.what() 
                      << " (code: " << ex.GetErrorCode() << ")" << std::endl;
        }

    } catch (const MSR::DriverNotLoadedException& ex) {
        std::cerr << "MSR driver not loaded: " << ex.what() << std::endl;
        return 2;
    } catch (const MSR::MsrException& ex) {
        std::cerr << "MSR Exception: " << ex.what() 
                  << " (code: " << ex.GetErrorCode() << ")" << std::endl;
        return 3;
    }

    return 0;
}

PhysMemDrv.hpp Example (C++) - needs PhysMemDrv.sys installed and running

#include "PhysMemDrv.hpp"
#include <iostream>
#include <iomanip>

int main() {
    try {
        PhysMemDriver::DriverHandle driver;
        uint8_t firstByte = driver.ReadPhysical<uint8_t>(0xF0000);
        std::cout << "BIOS first byte: 0x" 
                  << std::hex << std::setw(2) << std::setfill('0') 
                  << static_cast<int>(firstByte) << std::dec << std::endl;
    } catch (...) {
        std::cerr << "Failed to read BIOS ROM" << std::endl;
    }
}

SMART.hpp Example(C++) - requires Admin privileges

#include "SMART.hpp"
#include <iostream>
#include <vector>
#include <memory>

int main() {
    try {
        // Scan all available drives (0-7)
        auto drives = smart_reader::ScanAllDrives(8);

        for (const auto& drive : drives) {
            std::cout << "Drive: " << drive->GetDrivePath() << "\n";
            std::cout << "Type: " << drive->GetDriveType() << "\n";
            std::cout << "Temperature: " << drive->GetTemperature() << " °C\n";
            std::cout << "Power-On Hours: " << drive->GetPowerOnHours() << "\n";
            std::cout << "Power Cycle Count: " << drive->GetPowerCycleCount() << "\n";
            std::cout << "Reallocated Sectors: " << drive->GetReallocatedSectorsCount() << "\n";

            if (drive->IsProbablySsd()) {
                std::cout << "SSD Life Left: " << drive->GetSsdLifeLeft() << "%\n";
                std::cout << "Total Bytes Written: " << drive->GetTotalBytesWritten() << "\n";
                std::cout << "Total Bytes Read: " << drive->GetTotalBytesRead() << "\n";
                std::cout << "Wear Leveling Count: " << drive->GetWearLevelingCount() << "\n";
            }

            std::cout << "-------------------------------------\n";
        }
    } catch (const std::exception& e) {
        std::cerr << "Error reading SMART data: " << e.what() << std::endl;
    }
}

info.hpp (C++) - Linux Only

#include "info.hpp"
#include <iostream>
using namespace LinuxInfo;
int main() {
    // Get CPU information
    auto cpuInfo = getCPUInfo();

    std::cout << "=== CPU Info ===\n";
    for (const auto& [key, value] : cpuInfo) {
        std::cout << key << ": " << value << "\n";
    }

    return 0;
}

Documentation

Full documentation is available on the GitHub Pages website:

https://gafoo173.github.io/HardView/

All documentation is in the docs/ folder:

  • What.md: API Reference & Output Examples
    Full explanation of every function, what info it returns, how to use it from Python, and real output samples.
  • INSTALL.md: Installation Guide
    Supported platforms, installation methods, and troubleshooting tips.
  • FAQ.md: Frequently Asked Questions
    Solutions to common installation, usage, and troubleshooting issues.
  • LiveViewAPI.md: LiveView API Reference
    Detailed explanation of the LiveView module API, including functions, usage, and examples.
  • LiveViewErrors.md: LiveView Errors & Exceptions
    Guides and examples for handling errors and exceptions in the LiveView module.

API Reference (Python)

Function (JSON) Function (Python Object) Description
get_bios_info() get_bios_info_objects() BIOS vendor, version, release date
get_system_info() get_system_info_objects() System manufacturer, product name, UUID
get_baseboard_info() get_baseboard_info_objects() Motherboard info
get_chassis_info() get_chassis_info_objects() Chassis/computer case info
get_cpu_info() (Windows Only) get_cpu_info_objects() (Windows Only) Processor details
get_ram_info() get_ram_info_objects() Memory modules and totals
get_gpu_info (Windows Only) get_gpu_info_objects() (Windows Only) GPU information
get_disk_info() get_disk_info_objects() Storage devices
get_network_info() get_network_info_objects() Network adapters
get_partitions_info() get_partitions_info_objects() Disk partitions (advanced)
get_cpu_usage() get_cpu_usage_objects() Current CPU usage
get_ram_usage() get_ram_usage_objects() Current RAM usage
get_system_performance() get_system_performance_objects() Combined CPU/RAM usage
monitor_cpu_usage_duration(d, i) monitor_cpu_usage_duration_objects(d,i) Monitor CPU usage over time
monitor_ram_usage_duration(d, i) monitor_ram_usage_duration_objects(d,i) Monitor RAM usage over time
monitor_system_performance_duration(d,i) monitor_system_performance_duration_objects(d,i) Monitor system performance over time

LiveView Classes & Methods

API Reference

Class.Method Aliases Description
PyLiveCPU.get_usage(interval_ms) --- Get total CPU usage % over a given interval.
PyLiveCPU.cpuid() cpu_id() Get CPU details via CPUID instruction.
PyLiveCPU.CpuSnapShot(...) (Windows) cpu_snapshot(...) Get raw CPU time counters for a specific core or number of cores.
PyLiveRam.get_usage(Raw=False) --- Get total RAM usage % or raw [used_bytes, total_bytes].
PyLiveDisk(mode) --- Create disk monitor (mode=0 % usage [Windows], mode=1 R/W MB/s).
PyLiveDisk.get_usage(interval) --- Get disk usage as % or {Read MB/s, Write MB/s}.
PyLiveDisk.HighDiskUsage(...) high_disk_usage(...) Check if disk R/W exceeds threshold.
PyLiveNetwork.get_usage(interval, mode=0) --- Get total MB/s (mode 0) or per-interface MB/s (mode 1).
PyLiveNetwork.getHighCard() get_high_card() Get name of network adapter with highest usage.
PyLiveGpu.get_usage(interval_ms) (Windows) --- Get total GPU usage %.
PyLiveGpu.get_average_usage(interval_ms) (Windows) --- Get average GPU usage %.
PyLiveGpu.get_max_usage(interval_ms) (Windows) --- Get maximum GPU usage %.
PyLiveGpu.get_counter_count() (Windows) --- Get number of GPU counters monitored.
PyTempCpu.get_temp() (Windows) --- Get current CPU temperature.
PyTempCpu.get_max_temp() (Windows) --- Get max CPU core temperature.
PyTempCpu.get_avg_temp() (Windows) --- Get average CPU core temperature.
PyTempCpu.get_fan_rpm() (Windows) --- Get CPU fan RPM.
PyTempCpu.update() --- Refresh CPU temperature & fan RPM.
PyTempCpu.reget() re_get() Re-read CPU temperature & fan RPM.
PyTempGpu.get_temp() (Windows) --- Get current GPU temperature.
PyTempGpu.get_fan_rpm() (Windows) --- Get GPU fan RPM.
PyTempGpu.update() --- Refresh GPU temperature and fan RPM.
PyTempGpu.reget() re_get() Re-read GPU temperature and fan RPM.
PyTempOther.get_mb_temp() (Windows) --- Get motherboard temperature.
PyTempOther.get_Storage_temp() (Windows) get_storage_temp() Get storage temperature.
PyTempOther.update() --- Refresh other temperatures.
PyTempOther.reget() re_get() Re-read other temperatures.
PySensor.GetData(init=False) (Windows) get_data(init=False) Fetch sensors & fan data.
PySensor.GetValueByName(name) (Windows) get_value_by_name(name) Get sensor value by name.
PySensor.getAllSensors() (Windows) get_all_sensors() List all sensor names.
PySensor.getAllFanRPMs() (Windows) get_all_fan_rpms() List all fan RPM readings.
PySensor.update() --- Refresh sensors & fans data.
PySensor.reget() re_get() Re-fetch sensors & fans data.
PyManageTemp.Init() (Windows) init() Initialize temperature monitoring.
PyManageTemp.Close() (Windows) close() Shutdown temperature monitoring.
PyManageTemp.Update() (Windows) update() Update all temperature data.
PyRawInfo.RSMB() (Windows) rsmb() Get raw SMBIOS table bytes.
PyLinuxSensor.getCpuTemp() (Linux) get_cpu_temp() Get CPU temperature.
PyLinuxSensor.getChipsetTemp() (Linux) get_chipset_temp() Get chipset temperature.
PyLinuxSensor.getMotherboardTemp() (Linux) get_motherboard_temp() Get motherboard temperature.
PyLinuxSensor.getVRMTemp() (Linux) get_vrm_temp() Get VRM/memory temperature.
PyLinuxSensor.getDriveTemp() (Linux) get_drive_temp() Get storage temperature.
PyLinuxSensor.getAllSensorNames() (Linux) get_all_sensor_names() List all sensor names.
PyLinuxSensor.findSensorName(name) (Linux) find_sensor_name(name) Search for a sensor name.
PyLinuxSensor.GetSensorTemp(name, Match) (Linux) get_sensor_temp(name, Match) Get sensor temperature by name.
PyLinuxSensor.GetSensorsWithTemp() (Linux) get_sensors_with_temp() Get all sensors with their temperatures.
PyLinuxSensor.update(names=False) (Linux) --- Refresh sensor readings.

Sensor Value Fetch Flow (LiveView)

(Uses mermaid.js diagrams in compatible renderers)

classDiagram
    class LiveView {
        Request To Read
    }

    %% Linux path
    class LinuxPath {
        Search sensor name in lm-sensors
        If found → return value
        If not found → return -1
    }

    %% Windows path
    class WindowsPath {
        Check if monitoring library is initialized
        If initialized → ask HardwareWrapper
    }

    class HardwareWrapper {
        Forward request to LibreHardwareMonitorlib
        If value available → return value
        If not available → return -1
    }

    %% Relations
    LiveView --> LinuxPath : "Linux"
    LiveView --> WindowsPath : "Windows"
    WindowsPath --> HardwareWrapper

Loading

Platform Support

Feature Windows Linux
BIOS Info yes yes
System Info yes yes
Baseboard Info yes yes
Chassis Info yes yes
CPU Info yes yes (by LiveView)
RAM Info yes yes
Disks yes yes
Network yes yes
Advanced Storage / SMART yes No
Performance Monitoring yes yes
Sensors yes (by LiveView) yes (by LiveView)

Development

Build Instructions

Windows:

python setup.py build_ext --inplace

Linux:

python setup.py build_ext --inplace

License

Core Project (HardView)

All core project files, including project-specific libraries and header files are licensed under the MIT License. They are free for both personal and commercial use.

Tools Folder

All tools in the Tools folder are licensed under: GNU GENERAL PUBLIC LICENSE (GPL-3).
You can read the full license terms in the file: LICENSE.GPL3.


Quick Start And Tests

For a quick and visually appealing colorful test in the console using the rich library, run the script:
test.py

To test all functions normally without coloring, use:
TestAll.py

LiveView Tests
The LiveView test files are located in tests/units


Contribution

Contributions are welcome!

  • Fork and submit pull requests
  • Report issues or feature requests

HardView — Your Window into Hardware Information

See HardView API: For the full HardView API

See LiveView API: For the full LiveView API