Automatically generate/update unit tests using AI

The AI Test Case Generator, powered by Bito CLI, is a testament to how AI can enhance the quality of software. By automatically generating/updating unit tests, you can achieve 100% code coverage, leading to reliable applications. Basically, test cases provide a safety net, catching bugs and issues that might be overlooked in manual testing.

Developers enjoy coding but often find writing unit tests tedious. Bito’s AI Test Case Generator significantly cuts down the time spent on writing unit tests, speeding up the development process. It can be integrated into local development environment or CI/CD process. You can even Specify your test automation requirements like test framework, behaviors, and scope to customize automation to your needs.

Provide Bito CLI a file and it will automatically generate function/method tests along the happy path as well as boundary/edge cases. If it’s an API/interface, it will build test related to testing the interface, authorization, input validation, throttling, etc. It will generate mocks/stubs as necessary.

How to use

• Clone the Repository: Clone this repository to your local machine.
• Run the Script: Execute the script by providing the source code file as well as context file(s):

./generate_testcases.sh source_file context_file_1 context_file_2 ...

generate_testcases.sh

This code is a bash script that utilizes the “bito” command-line tool (https://github.com/gitbito/CLI) to generate test cases for a given code file. The script performs the following steps:

1. Check if “bito” is installed:
   • If “bito” is not found, it displays an error message and exits.
2. Check the number of arguments:
   • The script expects at least one argument, which should be the path to the code file.
   • The following arguments are code files that can be useful for the AI to have context.
   • If the number of arguments is incorrect, it displays a usage message and exits.
3. Delete the context.txt file if found.
   • This file can be useful if the user wants to keep talking with the AI after using the tool.
   • It is erased every time the tool runs to start from zero.
4. Extract the filename without the extension:
   • The script extracts the filename from the provided path and removes the file extension.
   • The extracted filename is stored for later use.
5. Ask the user for his/her preferred testing framework and save it in a variable.
6. Read the two promps into variables to be used by the bito command.
   • The script reads the prompts into variables.
   • Then replaces the desired framework, the file to cover name and pastes the concatenated context files at the end of the first prompt.
7. Run the bito command:
   • The script executes the bito command with the following options:
     • -p "$temp_prompt" : This is the first prompt with the replaced variables according to the user’s input.
     • -f "$1" : Specifies the path to the code file provided as an argument.
     • -c "context.txt" : the file where the context of the conversation with the AI is to be written.
8. Run the bito command:
   • The script executes the bito command with the following options:
     • -p "$temp_prompt" : This time, this variable is loaded with the second prompt, which tells the AI to complete the test and cover error/boundary paths.
     • -f "$1" : Specifies the path to the code file provided as an argument.
     • -c "context.txt" : the file where the context of the conversation with the AI is to be written.
9. Run extract_code.sh on the output file:
   • The script executes the extract_code.sh script on the temporary output file generated in the previous step.
   • The purpose and functionality of the extract_code.sh script is documented in section below.
   • If a code block does not specify a language or if the language is not one of the ones recognized by the script (python, javascript, js, bash), the script will save the code block to a .txt file. Update the code to add more languages based on your needs
10. Remove the temporary output file:
    • The script removes the temporary output file "${filename}.$$" to clean up after execution.

Usage Detail

The script requires at least one argument: the path to the source file. If the source file does not exist or if the number of arguments is incorrect, the script will display an error message and exit. The rest of the arguments are optional, but very useful. In addition to the file to cover with tests, you can pass more files to be used by the AI to have context about the code you want to cover with tests. You can pass as many files as you want, but for AI performance reasons, it is recommended not to pass more than 5 extra files.

Note: Please checkout and change mode to execute for all shell scripts in AI-Automation/unittests/gentestcase folder and run the script from the folder itself.

./generate_testcases.sh source_file context_file_1 context_file_2 ...

extract_code.sh

This is a bash script that extracts code blocks from a given source file and saves them as separate files. The script assumes that the source file uses Markdown syntax, with code blocks surrounded by triple backticks (“`). The language of the code block is determined by the string immediately following the opening backticks (e.g., “`python).

Usage Detail

The script requires one argument: the path to the source file. If the source file does not exist or if the number of arguments is incorrect, the script will display an error message and exit.

./extract_code.sh source_file

Main Functionality

The script reads the source file line by line. When it encounters a line that starts with triple backticks, it checks whether this line signifies the start or the end of a code block.

• If it’s the start of a code block, the script stores the language of the code block (the string immediately following the backticks) in block_lang and sets in_code_block to true.
• If it’s the end of a code block, the script determines the file extension based on the language stored in block_lang. It then increments the counter corresponding to that language. The content of the code block is saved to a file named test_case_<source_filename>_<counter>.<extension>, where <source_filename> is the name of the source file without the extension, <counter> is the counter for the corresponding language, and <extension> is the file extension determined by the language.

Edge Cases

The script assumes that the code blocks in the source file are correctly formatted according to Markdown syntax. If a code block does not specify a language or if the language is not one of the ones recognized by the script (python, javascript, js, bash), the script will save the code block to a .txt file. Update the code to add more languages based on your needs.

Example

Given a source file example.md with the following content:

```python
print("Hello, world!")
```

```shell
echo "Hello, world!"
```

The script will create two files: test_case_example_1.py and test_case_example_1.sh, each containing the corresponding code block.