dbq-blog

About Project

A quick project hosting all the codes I Wrote while writting the Blog about Dbq.

To Run the Project as is:

Setup up a Database (call it whatever you want)
Create a table called store in your database

CREATE TABLE store ( id INT PRIMARY KEY AUTO_INCREMENT, product VARCHAR(120) NOT NULL, quantity INT(10) NOT NULL, price DECIMAL NOT NULL, available BOOL, timing DATETIME NOT NULL);

copy the config-example.json file to a config.json file in the same directory and fill in your db connection details.
You can now Run codes in the project

Structure of the Project

Main

main.go file is the Entry point of the project and it runs all the functions created in different files. Before running go run main.go in you terminal you would have to uncomment any of the function you don't want to execute.

Others

The file naming of the other files are self explanatory.

Insert

insert_withdbq.go: Contains Sql Database Insert Code snippet that use Dbq for db insert.
insert_withoutdbq.go: Contains Sql Database Insert Code snippet that does not use Dbq for db insert.

Query

query_withdbq.go: Contains Sql Database Query Code snippets that use Dbq for db query.
query_withoutdbq.go: Contains Sql Database Query Code snippets that does not use Dbq for db query.

Script

For the sake of Benchmarking to compare the differences between when Query uses Dbq and does not use it, I wrote a DbScript which populates a database with up 100000 random name data (ofcourse you can increase or decrease the number if you want to investigate further).

To Run the script:

Setup up a Database (call it whatever you want)
Create a table called benchmark in your database

CREATE TABLE benchmark(id int primary key auto_increment not null, name varchar(255) not null, timestamp timestamp not null);

cd into the dbScript directory
make sure your db connection details are already set in a config.json file
run the go run main.go command Just wait and your table will be populated with up to 100000 rows at 50hops of 2000 data insert per hop using DBQ

Benchmark Test

In the dbq-blog_test file, I wrote about five benchmark tests

To Run the Benchmark Test:

make sure you are in the root directory
make sure you have already set up benchmark Table and populated it with the script as directed above.
make sure your db connection details are already set in a config.json file
Run go test -bench . (This runs all the benchmark tests). Go to Go's Documentation to see other options available for the benchmark command (e.g go test -bench . -benchtime=10x)

  goarch: amd64
  pkg: dbq-blog
  BenchmarkSingleRowQueryWithoutDBQ-2              	    2002	    530847 ns/op
  BenchmarkMultipleRowsQueryWithoutDBQ-2           	       1	1235999600 ns/op
  BenchmarkSingleRowQueryWithDBQ-2                 	    1916	    595392 ns/op
  BenchmarkMultipleRowsQueryWithDBQ-2              	       1	8387918970 ns/op
  BenchmarkMultipleRowsQueryWithDBQNoTimeParse-2   	       1	6083932760 ns/op
  PASS
  ok  	dbq-blog	18.119s

Please, Do Note that: Speed is not the only Factor to be considered. Although the Speed measurement can help guide your decision to use Dbq in any project or not but the NUMBER 1 reason to use dbq is for convenience and productivity - not for speed.

If you would Like to Help Adapt this repo to also work with PostgreSQL, a PR is highly welcome.

This of course is a good task for beginners. By the time you go through this project and understand it well enough you can exercise yourself by adapting it to PostgreSQL. DBQ is compatible for both MySQL and PostgreSQL.

If you have better ways I could have implemented some of the functions in this project

I would really love to learn from your PR. :D

propersam/dbq-blog