https://github.com/elecgeek/healthmeter

Converts the binary file (.DAT) into a more readable and makes some statistics for health or sport meters.
https://github.com/elecgeek/healthmeter

cplusplus cplusplus-17 data-processing healthcare

Last synced: 7 months ago
JSON representation

Converts the binary file (.DAT) into a more readable and makes some statistics for health or sport meters.

• Host: GitHub
• URL: https://github.com/elecgeek/healthmeter
• Owner: ElecGeek
• License: gpl-3.0
• Created: 2024-07-07T11:28:39.000Z (about 2 years ago)
• Default Branch: main
• Last Pushed: 2024-08-08T20:04:48.000Z (almost 2 years ago)
• Last Synced: 2025-03-16T06:12:24.326Z (over 1 year ago)
• Topics: cplusplus, cplusplus-17, data-processing, healthcare
• Language: C++
• Homepage:
• Size: 48.8 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
For the whose has an health or sport meter with the properties:

* can be connected with a (micro) usb that exposes a disk containing a data file.

* the file is a concatenation of measurements, until the meter's memory is full (and has to be erased from the device).

* the measurement binary files:

  * contains some data

  * contains records separated by a series of $ff (odd or even)

    * the header

    * the record number

    * some data

    * the date and time as plain char numbers (only from 0 to 12, 23, 31 or 59 are used)

    * some data

    * the sampling rate as the time, in seconds, between two samples.

    * some data

      * contains a data zone that can read as a raw file 8bits, signed, 2 channels

      * this data is between 0 and 100

  * terminates with a set of $ff

  * contains some data

This project is for him.

The project process:

* the parsing of the arguments to get the files names

  * if no file-name is provided at all, the default one is used

  * if at least one is provided, it or they are processed

* the reading of these binary files in the same order

* some statistics. They are sent to the standard output as:

  * the record number

  * some data in debug mode only

  * the date

  * the time

  * the record length in minutes

  * the min, max and average

  * many percent of the time the value is under a certain threshold

* a file-name based on the first record time-stamp for each input file.

* a raw file, of that file-name, intended to be open by most of the music editing software.

  * a record limit containing the time-stamp in the 2 channels. It has been design to not be confused with the record themselves

    * 100% sample

    * date

    * 0%

    * time

    * 100% sample

	The user has to zoom at the maximum to see the samples individually.

    The numbers (between 0 and 59) are coded as 2 samples. One is for the left digit, one for the right. They are multiplied by 10%. e.g. 52 is 50% 20%.

  * the data converted from 0 to 100% into 0 to 127. By this way, the editing software shows 100% for a 100% measure.

An additional feature applies if a record follows the previous one exactly one hour later:

* the statistics display the time only, not the date.

* the record limit, in the raw file, are omitted.

All the interfaces with the machine are done via the C++11 features. The project should compile by a simple "make" invocation.

The Makefile supports the SOURCEDIR, BUILDDIR, DESTDIR, CXX, CXXFLAGS and LDFLAGS variables for cross-compilation. 

It compiles on Ubuntu, Windows and MacOS-X

There is a work-flow to build standard versions. For specific versions, such as default names, one has to build himself.

TODO or never: handle properly the new year and fix the bug that does not merge the records in the new year night.