https://github.com/g0mb4/simdatabase
A MySQL/WebAPI(JSON) extension for Simulink.
https://github.com/g0mb4/simdatabase
cmex json mysql s-function sfun simulink webapi
Last synced: 7 months ago
JSON representation
A MySQL/WebAPI(JSON) extension for Simulink.
- Host: GitHub
- URL: https://github.com/g0mb4/simdatabase
- Owner: g0mb4
- License: apache-2.0
- Created: 2018-07-26T19:53:53.000Z (over 7 years ago)
- Default Branch: master
- Last Pushed: 2020-12-30T09:59:24.000Z (about 5 years ago)
- Last Synced: 2025-03-14T03:11:10.090Z (11 months ago)
- Topics: cmex, json, mysql, s-function, sfun, simulink, webapi
- Language: C
- Homepage:
- Size: 491 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# SimDatabase
A MySQL/WebAPI(JSON) extension for Simulink.
## MySQL block
The MySQL database block connects to a MySQL server in order to request the desired output data, this is achieved by a custom CMEX S-Function written in C using the **MySQL Connector C API version 6.1.11**.
The block can interpret a predefined format of SQL entry:
| date | time | data |
|:-----------:|:--------:|:--------------------:|
| YYYY-MM-DD | hh:mm:ss | N1=v1;N2=v2;N3=v3; … |
| ... | ... | ... |
This way a specified number of *vs* can be extracted from the database for the given *Ns* based on the time interval.
The precision required for the simulation is, currently, achieved by linear interpolation, which can be turned off.
The connection and the query take a long time compared to the other operations. To solve this problem the block saves the results in a unique file and only uses the database when this file does not exist, otherwise reads the data from the file, which takes significantly less time.
## WebAPI (JSON) block
The WebAPI (JSON) block connects to a TCP/IP server in order to request the desired output data in the form of a JSON response. The block implements a TCP/IP client and uses [DaveGamble's cJSON library](https://github.com/DaveGamble/cJSON) to parse the response.
An example of a request:
```
http://127.0.0.1/api/api.php?startDate=2019-05-
17&startTime=06:00:00&endDate=2019-05-
17&endTime=20:30:00&sensors=TW,I
```
An example of a response (if the query was successful):
```
{
"data":[
{
"date":"2018-04-17",
"time":"06:00:00",
"TW":20.98,
"I":53.33
},
{
"date":"2018-04-17",
"time":"06:40:00",
"TW":21.95,
"I":103.33
}
]
}
```
An example of a response (if the query was unsuccessful):
```
{
"error": "startTime is needed"
}
```
The precision required for the simulation is, currently, achieved by linear interpolation, which can be turned off.
The connection and the query take a long time compared to the other operations. To solve this problem the block saves the results in a unique file and only uses the database when this file does not exist, otherwise reads the data from the file, which takes significantly less time.
## Requirements
- MATLAB (>= R2018b) + Simulink (trial can do the trick)
- Visual Studio 2017 Community Edition (free)
## Compilation
Setup ```mex``` with the ```mex -setup``` command in MATLAB.
Run ```sfun/sql/compile.m``` or ```sfun/webjson/compile.m```.
## Usage
Use the database example ```example/sql/sim_test.sql``` to setup the database.
In order to use the JSON block, an example implementation of the API can be found in ```example/webjson```, it is called **api**. It uses the previously installed database scheme.
## Notes
- Compiled with MS Visual Studio 2017 Community.
- Tested on Windows 10 x64.