| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
In extreme cases, all the images may be too much to handle by the
system ("Argument list too long", errno 7). In such cases, let's
Do The Right Thing™ and temporarily disable the open_all_images
setting. Without this option we'll only open the single image.
Fixes #1488.
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The new destination argument for `:paste` requires a guard on whether it
is a valid target. A bug snuck in because of copious negation : )
This is now fixed, we only allow the target to be a directory or `None`
which means the current tab's working directory.
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
|\ \ \
| | | |
| | | | |
Load plugin directories located in ~/.config/ranger/plugins
|
| |/ / |
|
| | |
| | |
| | |
| | |
| | |
| | | |
Since now this directory is recreated if needed on each use of
it, there is no need to create it just because the user has these
options enabled on startup.
|
| | |
| | |
| | |
| | | |
Fixes #1435.
|
|\| | |
|
| |\ \ |
|
| | |\ \ |
|
| | |\ \ \
| | | | |/
| | | |/| |
|
| | |\ \ \ |
|
| | | | | | |
|
| | | | | | |
|
| | | | | | |
|
| | | | | | |
|
| | | | | | |
|
| | |_|_|/
| |/| | |
| | | | |
| | | | | |
Fixes #1350
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
ranger/core/loader.py:
CommandLoader.generate() gets text for preview panel from scope.sh
as the blocks of byrearray (max 512 bytes). This blocks immediately
decoded to utf-8. But in case when muli-byte unicode characters is
present they may be split and python will can't correctly to decode
it. For example: b'abc\xd0', b'\x9a...'
b'\xd0\x9a' - is cyrillic letter 'K' and it can't be decoded
separately (UnicodeDecodeError occurs).
So, received data blocks (bytearrays) must be collected and
only after that decoded as one whole byrearray.
Fixes #906
Fixes #967
Fixes #1166
|
| | | | |
| | | | |
| | | | |
| | | | | |
Fixes #1347
|
| | |_|/
| |/| |
| | | |
| | | | |
Fixes #1347
|
| | |/
| |/| |
|
| | | |
|
| | | |
|
| | | |
|
| |\ \ |
|
| | | | |
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Let's have the users report how they want shifting to work.
You can currently create tabs at negative indices so it feels wrong not
to be able to shift tabs there or have shifts of tabs at negative
indices have a weird result.
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
offset will shift the tab number as seen
pos will shift the tab to the tav number given
preserves as much as possible the tab numbers
|
| | | |
| | | |
| | | |
| | | | |
shift selected tab right/left with ALT-p or ALT-o
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
offset will shift the tab number as seen
pos will shift the tab to the tav number given
preserves as much as possible the tab numbers
|
| | | |
| | | |
| | | |
| | | | |
shift selected tab right/left with ALT-p or ALT-o
|
| |\ \ \ |
|
| | | | | |
|
| | | | | |
|
| | | |/
| | |/|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
If `selection == False` don't operate on the entire selection (the
marked files) but only on the file under the cursor.
Inconsistent, I noticed `move to=100 percentage=true` works as expected
but `move right=1 selection=false` doesn't. You need to pass `False`,
most other values test as `True` in python, so `true` works *but*
surprisingly imo, `false == True`.
Fixes #1233
|
|/ / /
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The implementation for the `%any_path` macro relies on the bookmark
being defined, however when expanding the macro for uses of `%any` that
don't involve existing bookmarks you get error messages even though the
failure's expected.
Simplest solution is to just fail silently, like the existing `%any`
macro already did. Lazy evaluation would silence the error for all uses
of `%any`, since the value for `%any_path` would never be used, but that
would require altering all the generation and consumption of macro
values. Another solution is to only resolve the `%any_path` macro when
it's necessary but that would involve changing the way `wildcards` are
passed distinguishing between `any` and `any_path` macros.
Fixes #1419
|
| | | |
|
|/ /
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The `%any` macro is very useful for commands that know about bookmarks.
Bookmarks are just short names for paths really, so it'd be nice if they
could be used with any commands that take paths as arguments.
Now they can! `%any_pathX` where `X` refers to the which `<any>` the
macro corresponds to (`%any_path` is synonymous with `%any_path0`) is
replaced with the path of the bookmark entered if that bookmark exists.
Fixes #1277
|
|\ \ |
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Was erroneously checking the entire selection for being *a* directory
but a list of 1 or more files/directories is obviously not a directory.
Fixes #1269
|
|\ \ \
| | | |
| | | | |
Implement the filter stack
|
| | | |
| | | |
| | | |
| | | | |
Closes #1268.
|
| | | |
| | | |
| | | |
| | | | |
Inspired by https://github.com/Fuco1/dired-hacks#dired-filter
|
|\ \ \ \
| | |/ /
| |/| | |
Actually check whether selection is a directory for rangerpath patch
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
`Actions.move()` used `enter_dir()` to determine whether the selection
was a directory. Now we explicitly check whether it's a directory.
Fixes #1177
|