hotedit

A simple library for opening your user's favorite text editor

The hotedit() API looks at a few places in the shell environment for a useful text editor, invokes it with your text, and finally returns the edited result to your program (or raises an exception if an error occurred invoking the editor).

Install

pip install hotedit

Installation for maintainers

After checking out github source, work with it using Poetry.

1. Install poetry:

curl -sSL https://install.python-poetry.org | python3 -

2. Activate your poetry shell in the git repo you checked out, and install dependencies:

poetry shell
...

poetry install -E test

3. Run tests:

make test

Use

# Get a string from the Internet so we can edit it:
import requests
URL = "https://pastebin.com/raw/Df9NAmYc"
response = requests.get(URL)

"""
Call `hotedit()` to launch the text editor to edit what was in the pastebin
document.

The first argument `initial` is the string to be shown in your editing buffer.
This creates a temp file, later deleted.

Other useful arguments:
  - validate_unchanged (default False): When True, raise `Unchanged` exception when
    the user did not make any edits.
  - delete_temp (default True): When False, doesn't delete the temp file (for 
    troubleshooting).
  - find_editor (default `hotedit.editor.determine_editor`): a zero-argument
    callable that returns the path to an editor executable (see below)
"""
from hotedit import hotedit
edited = hotedit(response.text)
# `edited` is the string returned from editing.

print("Your edited text:")
for line in edited.splitlines():
    print(f"> {line}")

# At this point we might upload the new string back to pastebin.

How it determines your editor

Hotedit looks at your shell environment to determine which editor you're using.

It follows some very simple rules.

1. Check the following in order for a useful editor:

   1. git config core.editor
   2. EDITOR environment variable
   3. VISUAL environment variable

2. If one of these is a string, that string is returned as the path to an editor.

3. hotedit() itself will raise an exception if the path returned is not an editor executable, but it does not keep trying other options if this happens.

This means if core.editor is unset, $EDITOR is set to a path that's missing, and $VISUAL is set to a valid path to an editor, hotedit will use $EDITOR and fail, without checking $VISUAL.

Overriding editor search with find_editor

If you don't like hotedit's search order, you can pass your own find_editor function. Commonly, you want to try a different location first, then try the default options. Here's an implementation of that:

import os
import hotedit, hotedit.editor

def my_find_editor():
    if os.environ.get("MY_APP_TEXT_EDITOR"):
        return os.environ["MY_APP_TEXT_EDITOR"]
    return hotedit.editor.determine_editor()

edited = hotedit.hotedit("[MY_APP config]...", find_editor=my_find_editor)
...

Suggestions for setting your editor

Many editor values just work, especially if they are commonly used in the terminal. For example, you can just set EDITOR=nano or EDITOR=vim.

Editors that open in their own window, or make use of an already-open window, almost always provide a command-line option for exactly the use cases hotedit is used for. For example, try the following:

(Open a new gvim window, or open a tab in an existing gvim window, respectively):

EDITOR="gvim -f"  # or
EDITOR="gvim --remote-tab-wait"

(Open a tab in VS Code):

EDITOR="code -w"

If your users are likely to be using git, the best experience for them is usually to invoke an editor the same way git would invoke one to ask for a commit message. While it's not up to the hotedit library to tell you how to live your life, here's how you would do that:

# VS code, for example
git config core.editor "code -w"

Maintainer section: releasing

To cut a release of this software, automated tests must pass. Check under Actions for the latest commit.

Create an RC branch and test

• We use the Gitflow process. For a release, this means that you should have a v1.2.3-rc branch under your develop branch. Like this:

    main  
    └── develop  
        └── v1.2.3-rc
  

• Update this file.

  1. Confirm that the docs make sense for the current release.
  2. Check links!
  3. Update the Changelog section at the bottom.

• Perform whatever tests are necessary.

Tag and cut the release with Github Actions

• Once you have tested in this branch, create a tag in the v1.2.3-rc branch:

  git tag -a -m v1.2.3 v1.2.3
  git push --tags
  

• Navigate to https://github.com/corydodt/hotedit/actions and run the action labeled ... release.

  • You will be asked to choose a branch. Choose your rc branch, e.g. v1.2.3-rc

  • If you run this action without creating a tag on v1.2.3-rc first, the action will fail with an error and nothing will happen.

  If you have correctly tagged a commit and chosen the right branch, this will run and create a new release on the Releases page.

• Edit the release on that page

Merge up

• Finish up by merging your -rc branch into
  1. main and then
  2. develop.

Changelog

versions 0.9.1.post2

• Initial public release.
• Automated builds, automated tests.