iterfzf 1.2.0.46.1

Demo session

See also the API reference.

Key features

• No dependency but only Python is required. Prebuilt fzf binary for each platform is bundled into wheels. Everything is ready by pip install iterfzf. (Note that not wheels of all supported platforms are uploaded to PyPI as they don’t allow minor platforms e.g. FreeBSD. The complete wheels can be found from the GitHub releases.)

• Consumes an iterable rather than a list. It makes UX way better when the input data is long but streamed from low latency network. It can begin to display items immediately after only part of items are ready, and before the complete items are ready.

• Supports Python 3.8 or higher.

iterfzf.iterfzf(iterable, *, **options)

Consumes the given iterable of strings, and displays them using fzf. If a user chooses something it immediately returns the chosen things.

The following is the full list of parameters. Pass them as keyword arguments except for iterable which comes first:

iterable (required)

The only required parameter. Every element which this iterable yields is displayed immediately after each one is produced. In other words, the passed iterable is lazily consumed.

It can be an iterable of byte strings (e.g. [b'foo', b'bar']) or of Unicode strings (e.g. [u'foo', u'bar']), but must not be mixed (e.g. [u'foo', b'bar']). If they are byte strings the function returns bytes. If they are Unicode strings it returns Unicode strings. See also the encoding parameter.

multi

True to let the user to choose more than one. A user can select items with tab/shift-tab. If multi=True the function returns a list of strings rather than a string.

False to make a user possible to choose only one. If multi=False it returns a string rather than a list.

For both modes, the function returns None if nothing is matched or a user cancelled.

False by default.

Corresponds to -m/--multi option.

print_query

If True the return type is a tuple where the first element is the query the user actually typed, and the second element is the selected output as described above and depending on the state of multi.

False by default.

Corresponds to --print-query option.

New in version 0.3.0.

encoding

The text encoding name (e.g. 'utf-8', 'ascii') to be used for encoding iterable values and decoding return values. It’s ignored when the iterable values are byte strings.

The Python’s default encoding (i.e. sys.getdefaultencoding()) is used by default.

extended

True for extended-search mode. False to turn it off.

True by default.

True corresponds to -x/--extended option, and False corresponds to +x/--no-extended option.

exact

False for fuzzy matching, and True for exact matching.

False by default.

Corresponds to -e/--exact option.

case_sensitive

True for case sensitivity, and False for case insensitivity. None, the default, for smart-case match.

True corresponds to +i option and False corresponds to -i option.

query

The query string to be filled at first. (It can be removed by a user.)

Empty string by default.

Corresponds to -q/--query option.

prompt

The prompt sequence. ' >' by default.

Corresponds to --prompt option.

preview

The preview command to execute. None by default.

Corresponds to --preview option.

mouse

False to disable mouse. True by default.

Corresponds to --no-mouse option.

ansi

True to enable ansi colors mode. None by default.

Corresponds to --ansi option.

cycle

True to enable cycling scrolling.

False by default.

Corresponds to --cycle option.

__extra__

The iterable of extra raw options/arguments to pass to fzf.

Empty by default.

Author and license

The iterfzf library is written by Hong Minhee and distributed under GPLv3 or later.

The fzf program is written by Junegunn Choi and distributed under MIT license.

Changelog

/---------- 1. iterfzf's major version
|   /------ 3. bundled fzf's major version
|   |   /-- 5. bundled fzf's patch version
|   |   |
v   v   v
1.2.3.4.5
  ^   ^
  |   |
  |   \---- 4. bundled fzf's minor version
  \-------- 2. iterfzf's minor version