Migrate from using readline to prompt-toolkit #1553
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Tons of tests failing and some even getting stuck.
…f the correct arguments to app.complete() This fixes all of the test_argparse_comleter.py tests. There are still failing tests in test_cmd2.py, test_history.py, and test_run_pyscript.py as well as a test in test_cmd2.py that gets stuck.
TODO: - prompt-toolkit history isn't properly initialized with history from a persistent history file, as shown by the remaining failing history test
… of mocking the built-in input function. There are still 3 failing and 1 skipped test in test_cmd2.py Additionally, some tests in test_run_pyscript.py are getting stuck.All tests in other files are passing.
…ork with prompt-toolkit
Also: - Fixed make clean so it cleans up code coverage file artifacts
Also added a bottom toolboar for displaying these type hints.
tleonhardt
added
enhancement
major
dependencies
Contributor
|
🤖 Hi @tleonhardt, I've received your request, and I'm working on it now! You can track my progress in the logs for more details. |
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #1553 +/- ##
==========================================
+ Coverage 98.94% 99.12% +0.17%
==========================================
Files 21 21
Lines 4942 4901 -41
==========================================
- Hits 4890 4858 -32
+ Misses 52 43 -9
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. |
Contributor
|
🤖 I'm sorry @tleonhardt, but I was unable to process your request. Please see the logs for more details. |
… PromptSession and ensuring consistent input/output usage
…rs for better compatibility
…ws runners for better compatibility" This reverts commit 59c5ffa.
Updated Cmd._bottom_toolbar to display the application name in green on the left and the current ISO timestamp in blue on the right when include_bottom_toolbar is enabled. Used padding to achieve right-alignment of the timestamp. Updated associated tests to match the new return format.
Modified getting_started.py to spawn a background thread that triggers a UI redraw twice a second. This ensures that dynamic content in the bottom toolbar, such as the timestamp, stays current while waiting for user input.
Updated Cmd._bottom_toolbar to format the current time with 0.01s precision (two decimal places for fractional seconds). Maintained the ISO-like format including the timezone offset.
Updated documentation throughout the docs/ directory to reflect the migration from GNU Readline to prompt-toolkit and the addition of the new bottom toolbar feature. Added an upgrade guide section for 4.x, updated the history and prompt feature pages, and adjusted mkdocs configuration to allow documentation of the _bottom_toolbar method.
tleonhardt
changed the title
…pletion Added a unit test to verify that visible commands, aliases, and macros are correctly returned as CompletionItem objects for tab completion. Specifically addressed the case where a command has no docstring.
Member
Author
|
@kmvanbrunt This is ready for your review when you get a chance. @anselor @kotfu @theagilehacker @joqual If any of you have feedback, I'd also greatly appreciate it. |
Contributor
|
I need to dedicate a couple of evenings to run tests on windows.
…On Sat, Jan 17, 2026 at 12:20 AM Todd Leonhardt ***@***.***> wrote:
*tleonhardt* left a comment (python-cmd2/cmd2#1553)
<#1553 (comment)>
@kmvanbrunt <https://github.com/kmvanbrunt> This is ready for your review
when you get a chance.
@anselor <https://github.com/anselor> @kotfu <https://github.com/kotfu>
@theagilehacker <https://github.com/theagilehacker> @joqual
<https://github.com/joqual> If any of you have feedback, I'd also greatly
appreciate it.
—
Reply to this email directly, view it on GitHub
<#1553 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AECZY4BHD7XAADZB3BKKHT34HHBBPAVCNFSM6AAAAACQIXJGZSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTONRSGY4DGOJXGI>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
Modified ArgparseCompleter to only fallback to flag completion when the current positional argument is full (reached its max nargs) instead of just reaching its minimum. This fixes an issue where commands with optional positional arguments (like 'history') would automatically complete '-' and show flags instead of the more useful positional hint text when tab-completed with no input.
Modified ArgparseCompleter to return only one completion result per action, grouping all matching option strings (short and long forms) together in the display text. This prevents duplicate entries for the same flag from appearing separately in the tab completion menu, providing a cleaner and more concise user experience.
Added a unit test 'test_autcomp_fallback_to_flags_nargs0' to cover the logic in 'ArgparseCompleter' that falls back to flag completion when a positional argument has reached its maximum number of values and the current input is empty. This was achieved by manually patching the 'nargs' attribute of a positional action to 0.
Added a unit test 'test_multiline_complete_statement_eof' to cover the logic in '_complete_statement' where an EOF is encountered during multiline command input. The test verifies that 'eof' is correctly handled by converting it to a newline terminator and printing it.
…lict Updated examples/async_call.py to use a synchronous main entry point. This prevents a RuntimeError caused by prompt-toolkit attempting to start its own asyncio loop while one was already running in the same thread. The example's core functionality of bridging sync and async code remains intact through its use of a background thread for asynchronous tasks.
Also improved some docstring whitespace in examples/async_call.py
Warn that cmd2 starts its own asyncio event loop due to prompt-toolkit using it natively.
When raising a CompletionError with apply_style=True during tab completion
in prompt_toolkit mode, the error message was printed directly to stdout using
ANSI codes. prompt_toolkit's patch_stdout would capture this but fail to
interpret the ANSI codes, resulting in raw escape sequences being displayed.
This commit fixes the issue by:
1. Modifying cmd2.complete to capture the styled error message into
self.completion_header instead of printing directly to stdout.
2. Updating Cmd2Completer.get_completions (in pt_utils.py) to print
completion_header using print_formatted_text wrapped in ANSI,
which correctly renders the styled text above the prompt.
3. Updating tests/conftest.py complete_tester to print completion_header
so existing tests (like test_completion_error) continue to pass by
capturing the error output.
The previous commit introduced accessing self.cmd_app.completion_header in Cmd2Completer.get_completions, which caused AttributeError in tests using MockCmd because it lacked this attribute. This commit adds completion_header to MockCmd and includes a new test case test_get_completions_with_header to verify that the completion header is correctly printed when present.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR migrates from using GNU Readline for user input, tab-completion, and history to using prompt-toolkit for the same purpose.
prompt-toolkitis a pure-Pythonreadlinereplacement that is fully cross-platform compatible and doesn't rely on the presence of underlying C dynamic libraries. It also opens the door to some advanced features in the future.All use of
readlineis removed and thecmd2.rl_utilsmodule has been deleted. There is a newcmd2.pt_utilsmodule in its place. Currently all linting is passing and the documentation builds. This is essentially final draft that needs more testing. There are likely a some things that need to be considered related to asynicio support, signal handlers, and blocking popen calls.All tests are passing on all platforms. I have done some manual testing on both macOS and Linux. But I don't have a Windows VM to do any manual testing on Windows however.
NOTE FOR REVIEWERS: The substantive code changes are in the 3 following files (all in the
cmd2directory):argparse_completer.py(modified)cmd2.py(modified)pt_utils.py(new)TODO:
alias create <TAB>- currently showing nothinggetting_started.pyexample to show how to use the optional bottom toolbar featureexamples/async_call.pyexample to account forprompt-toolkitstarting anasyncioevent loop