TODO Notifier

• TODO Notifier
• Description
• How to use?
  • Method 1: As a pip package
  • Method 2: Directly cloning from GitHub
• Accessing generated summaries
• Technical Details
  • Working
  • Salient Features
  • Other Salient Features

Description

Find project on GitHub, PyPi
Medium Article detailing about TODO Notifier and how to use it (complements README): Medium

More often than not, we put some TODO items in code and forget about them. Sometimes, we think of coming back to a TODO item by some date but miss it being too busy with some other development.

TODO Notifier aims to solve this problem. It parses through any project, collects all the todo items, generates automated summaries and send automated reminders about them.

Recommended format to write TODO items**

TODO {2022-05-22} @user_name msg

Above format has following components

• TODO in capital (though users can make it case insensitive by setting the same in config.py). It need not to be starting word of the line. For a host of valid examples, please check tests/sample_test_file.py and tests/sample_test_file2.py
• [Optional] TODO is followed a date in YYYY-MM-DD format within curly brackets. The respective TODO item is expected to be completed by end of this date
• [Optional] Date is followed by a unique user name accompanied by @
• [Optional] User name is followed by the usual message/comment of the respective TODO item

However, the relative position of the TODO item, date inside {} brackets, username with @ and message should be as recommended above.

The framework is robust in the sense that if the TODO item misses some data like date and/or message and/or username etc., the respective TODO item will still be picked up by the TODO Notifier. However without relevant information, certain functionalities may not work. For e.g. without date, it cannot know if the TODO item has overshoot its expected date of completion.

How to use?

Before running TODO Notifier, check out the sample summary files produced by the TODO Notifier in directory sample_reports. It generates three default summaries. For more information, check out point #3 in Salient Features.

Method 1: As a pip package

Install TODO Notifier by running pip install todonotifier

Then you can use below code to try the TODO Notifier.

from todonotifier.config import DefaultConfig
from todonotifier.connect import CONNECT_METHOD, Connect
from todonotifier.driver import run as driver_run

git_url: str = ""  # Placeholder for HTTPS/SSH based git url. It could also be a local folder address
project_dir_name: str = ""  # Placeholder. Suggested to keep same as project name
sender_email = ""
password = ""
receivers = []
host = "smtp.gmail.com"  # Change accordingly
port = 465  # Change accordingly
notifier = EmailNotifier(sender_email, password, host, port, receivers)
config = DefaultConfig(save_html_reports=True, ignore_todo_case=True, notifier=notifier)  # Change flags as per need
connect = Connect(connect_method=CONNECT_METHOD.GIT_CLONE, project_dir_name=project_dir_name, url=git_url, branch_name="production")  # If using a local folder address in `git_url`, then change `connect_method` to `CONNECT_METHOD.DRY_RUN_DIR`

driver_run(connect=connect, config=config)

It will generate three files by default in folder .report under current working directory.

Method 2: Directly cloning from GitHub

Clone using git clone https://github.com/ashu-tosh-kumar/todo_notifier.git

Then you can use the user_driver.py file to run it. You can edit the user_driver.py to following to try running it on tests/sample_test_file.py.

git_url: str = "tests/sample_test_file.py"  # Placeholder for HTTPS/SSH based git url
project_dir_name: str = ""  # Placeholder. Suggested to keep same as project name
sender_email = ""
password = ""
receivers = []
host = "smtp.gmail.com"
port = 465
notifier = EmailNotifier(sender_email, password, host, port, receivers)
config: BaseConfig = DefaultConfig(save_html_reports=True, ignore_todo_case=True, notifier=notifier)  # Change flags as per need
connect = Connect(connect_method=CONNECT_METHOD.DRY_RUN_FILE, project_dir_name=project_dir_name, url=git_url, branch_name="production")  # If using a local folder address in `git_url`, then change `connect_method` to `CONNECT_METHOD.DRY_RUN_DIR`. For `CONNECT_METHOD.DRY_RUN_FILE` and `CONNECT_METHOD.DRY_RUN_DIR`, branch_name is not important

driver_run(connect=connect, config=config)

It will generate three files by default in folder .report under the main project directory.

This method is however not recommended and there's always a risk of overriding any local changes made when pulling latest changes. Please use Method 1.

Accessing generated summaries

After running above code to generate the summaries, the TODO Notifier stores these summaries as .html files if save_html_reports=True is passed in the configuration. All such reports are saved in directory .report in current working directory.

TODO Notifier also sends automated emails about the same if setup in the configuration.

However, if users require access to the generated summaries for some custom downstream integration, they can access the same as following:

# Access all the summary generators
config.config.summary_generators

# Access summary from (say) ByModuleSummaryGenerator
by_module_summary_generator = config.summary_generators[0]

# Check name of the respective summary generator
by_module_summary_generator.name

# Generated `list` of `TODO` objects
by_module_summary_generator.container

# Generated html summary of `TODO` objects
by_module_summary_generator.html

Technical Details

Working

• TODO Notifier copies/clones the respective repository into a temporary location to avoid the risk of modifying any file.
• It then reads through all the files in the project and collects all the TODO items.
• It then generates the summaries as specified in the configuration.
• Finally it sends the notifications (only Email notifications are supported as of now) to the configured email id.

Salient Features

• Allows excluding specific folders of the project via absolute address, relative address or regular expression from being scanned. It has a default list of folders that are not scanned: DEFAULT_EXCLUDE_DIRS in constants.py. But the same can be controlled using the flag flag_default_exclude_dirs in config/DefaultConfig

• Allows excluding specific files of the project via absolute address, relative address or regular expression from being scanned. It has a default list of files that are not scanned: DEFAULT_EXCLUDE_FILES in constants.py. But the same can be controlled using the flag flag_default_exclude_files in config/DefaultConfig

• Provides three default summary generators. Summaries are how TODO Notifier shares the information about TODO items. Each summary is essentially an HTML document that can be easily shared. The default summary generation can be controlled by flag: flag_default_summary_generators in config/DefaultConfig.

• Module wise list of all TODO items
• User wise list of TODO items expired already

• User wise list of TODO items that are supposed to expire in the upcoming week

• Flag save_html_reports can be used to control whether to save the generated summaries as files. If yes, it will store all generated summaries in folder .report locally

• Users can write their own summary generators and add pass the same in variable summary_generators in config/DefaultConfig. Each summary generator is a child of BaseSummaryGenerator in summary_generators.py

• Provides default implementation of sending notifications via Email as EmailNotifier in notifications.py. More ways of notifications can be added easily by inheriting from BaseNotifier

• Most of the features are configurable via configuration file. Configuration must be a child class of BaseConfig in config.py. It provides a default configuration class: DefaultConfig in config.py that can also be configured easily via various flag parameters. But users are free to write another configuration class or simply inherit from DefaultConfig

• Provides two ways of dry running the code locally viz. CONNECT_METHOD.DRY_RUN_FILE to dry run on a single local file and CONNECT.DRY_RUN_DIR to dry run on an entire local directory/project.

• user_driver.py provides examples on how to use. It can be modified accordingly to run the code.

Other Salient Features

• All code in Python 3
• Using following pre-commit hooks to keep code clean and nice during development
• black
• flake8
• isort
• requirements-txt-fixer
• Almost 100% test coverage for whole repository and we intend to keep it that way
• Integrated GitHub flows to keep repo clean and updated
• CodeQL
• OSSAR
• Python application