Readline completion of module names in import statements

@vadmium

BPO	25419
Nosy	@vadmium, @serhiy-storchaka, @MojoVampire
Dependencies	bpo-16182: readline: Wrong tab completion scope indices in Unicode terminals
Files	complete-import.patch complete-import.v2.patch

^{Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.}

Show more details

GitHub fields:

assignee = None
closed_at = None
created_at = <Date 2015-10-16.11:16:48.120>
labels = ['type-feature', 'library']
title = 'Readline completion of module names in import statements'
updated_at = <Date 2016-05-25.08:03:01.876>
user = 'https://github.com/vadmium'

bugs.python.org fields:

activity = <Date 2016-05-25.08:03:01.876>
actor = 'martin.panter'
assignee = 'none'
closed = False
closed_date = None
closer = None
components = ['Library (Lib)']
creation = <Date 2015-10-16.11:16:48.120>
creator = 'martin.panter'
dependencies = ['16182']
files = ['40794', '42986']
hgrepos = []
issue_num = 25419
keywords = ['patch']
message_count = 6.0
messages = ['253072', '253095', '253104', '253115', '253132', '266320']
nosy_count = 3.0
nosy_names = ['martin.panter', 'serhiy.storchaka', 'josh.r']
pr_nums = []
priority = 'normal'
resolution = None
stage = 'patch review'
status = 'open'
superseder = None
type = 'enhancement'
url = 'https://bugs.python.org/issue25419'
versions = ['Python 3.6']

Linked PRs

Merge remote-tracking branch 'origin/main' into call-len

Shouldn't we totally disable completion on import attributes?

Since completions don't come from module attributes but current namespace, I find it pretty misleading, and unlikely to be useful.

Eg. on 3.14 / current main,

>>> from typing import Na<TAB>
>>> from typing import NameError(

while the user probably wanted to import NamedTuple: since we deferred completion of attributes, I believe it would be best to provide no suggestion at all.

while the user probably wanted to import NamedTuple: since we deferred completion of attributes, I believe it would be best to provide no suggestion at all.

That sounds reasonable to me! Would you like to send a PR for that? We'd most likely need to modify ModuleCompleter.get_completions to somehow distinguish between:

an invalid import statement which shouldn't disable completions (ImportParser.parse returns None) and
a valid import statement with no available completions which should disable completions

Currently we just return an empty list for both.

That sounds reasonable to me! Would you like to send a PR for that? We'd most likely need to modify ModuleCompleter.get_completions

Sure, I'll have a go at it, thanks for the pointers!

Feel free to ping me if you need help!

Done, it was pretty straightforward indeed! 🚀

Two other things I noted while working on it:

Submodule completion doesn't seem to always work, eg. from collections. doesn't suggest collections.abc (whereas http.server works). Do you know why / if this is can be fixed? Looks like pkgutils.iter_modules returns nothing, but I didn't investigate further.
Wouldn't it be cool if from typing <TAB> completed the import keyword? ImportParser.parse() returns None in such a case, and I don't know if there is precedent for completing after a space, might be a litte bit too magical?

Thanks! I'll review it tomorrow.

This is expected, collections.abc is not a real module, same for os.path so pkgutil won't find it. One thing we could do is hardcode these modules but I haven't had the time to look into it.
Yup! I'm already working on that along with general improvements to the parser :)

gh-69605: Disable PyREPL module autocomplete fallback on regular comp…

pythongh-69605: Disable PyREPL module autocomplete fallback on regula…

[3.14] gh-69605: Disable PyREPL module autocomplete fallback on regul…

pythongh-69605: Add PyREPL import autocomplete feature to 'What's New' (

pythongh-69605: Disable PyREPL module autocomplete fallback on regula…

Closing since this was actually completed.

That reminded me of this issue:

This is expected, collections.abc is not a real module, same for os.path so pkgutil won't find it. One thing we could do is hardcode these modules but I haven't had the time to look into it.

I went ahead and opened a small PR to do just that, let me know if it should become a separate issue!

Let's just reopen it.

gh-69605: Hardcode some stdlib submodules in PyREPL module completion…

pythongh-69605: Hardcode some stdlib submodules in PyREPL module comp…

[3.14] gh-69605: Hardcode some stdlib submodules in PyREPL module com…

gh-69605: Check for already imported modules in PyREPL module complet…

pythongh-69605: In test_pyrepl.test_already_imported*, invalidate FS …

gh-69605: In test_pyrepl.test_already_imported*, invalidate FS import…

pythongh-69605: Check for already imported modules in PyREPL module c…

pythongh-69605: In test_pyrepl.test_already_imported*, invalidate FS …

[3.14] gh-69605: Check for already imported modules in PyREPL module …

Nice, I think this issue can be closed again!

pythongh-69605: Check for already imported modules in PyREPL module c…

pythongh-69605: In test_pyrepl.test_already_imported*, invalidate FS …

gh-69605: Add math.integer to PyREPL module completer hardcoded list (#…

@tomasr8

Wouldn't it be cool if from typing completed the import keyword? ImportParser.parse() returns None in such a case, and I don't know if there is precedent for completing after a space, might be a litte bit too magical?

Yup! I'm already working on that along with general improvements to the parser :)

Hey @tomasr8, is this still on your radar? I'd love to see this land someday, can have a try at it if needed!

@loic-simon

Wouldn't it be cool if from typing completed the import keyword? ImportParser.parse() returns None in such a case, and I don't know if there is precedent for completing after a space, might be a litte bit too magical?

Yup! I'm already working on that along with general improvements to the parser :)

Hey @tomasr8, is this still on your radar? I'd love to see this land someday, can have a try at it if needed!

Hey @loic-simon I was indeed working on that at some point (I even have a branch for it, but can't find it ATM), but feel free to pick it up! The import parser might also need some refactoring while you're at it :) I'll try to review it in a timely manner 😆

pythongh-69605: Add math.integer to PyREPL module completer hardcoded…

pythongh-69605: Disable PyREPL module autocomplete fallback on regula…

gh-69605: fix PyREPL completions inserted for import statements with …

[3.15] gh-69605: fix PyREPL completions inserted for import statement…

gh-69605: PyREPL: insert "import" after "from foo <tab>" (#148445)

Navigation Menu

Repository navigation

Readline completion of module names in import statements #69605

Linked PRs

Metadata

Assignees

Labels

Projects

Milestone

Relationships

Development

Readline completion of module names in import statements #69605

Description

Linked PRs

Activity

loic-simon commented on May 17, 2025

tomasr8 commented on May 17, 2025

loic-simon commented on May 18, 2025

tomasr8 commented on May 18, 2025

loic-simon commented on May 18, 2025

tomasr8 commented on May 18, 2025

picnixz commented on Aug 30, 2025

loic-simon commented on Aug 30, 2025

picnixz commented on Aug 30, 2025

loic-simon commented on Jan 5, 2026

loic-simon commented on Apr 10, 2026

tomasr8 commented on Apr 11, 2026

Metadata

Metadata

Assignees

Labels

Projects

Milestone

Relationships

Development

Issue actions