| Age | Commit message (Collapse) | Author | Files | Lines |
|
|
|
|
|
The core vis code was originally written under the assumption that there
always exists at least one window. However, when being called from the
Lua configuration file during start up this is not yet the case.
In general, Lua code should always be placed within appropriate event
handlers e.g. vis.events.INIT for global configuration. Invoking API
functions depending on an active window from top level statements is
not supported. Where before the editor simply crashed, these changes
turn such code sections into NOPs.
Fix #561
|
|
Special case <C-v><Enter> to still insert a carriage return as discussed
in #656 and changed in 2cfc9c867bdfd4cc3ae3246f31cf636633fe1a5f.
Due to limitations of the current implementation <C-v> is not generic,
i.e. combining it as r<C-v><Enter> will not work.
Fixes #765
|
|
|
|
|
|
Check for end-of-text.
|
|
Valid values are `read`, `mmap` or `auto`.
|
|
|
|
Fix #656
|
|
When the fore and background colors are the same, swapping them has
no effect. Instead use the specified cell attributes. Previously the
CELL_ATTR_REVERSE used in the default selection style was ignored.
In general the default style definitions for non-Lua builds could
probably be improved further.
Fix #635
|
|
Previously `ci<` would have no immediate effect because in operator
pending mode `i<` was wrongly treated as a powwible prefix of `i<Tab>`.
Fix #624
|
|
Use visual mode and :| to filter text through external commands.
The mapping was already reused for selection complement.
|
|
In practice this was never an issue also it is guaranteed that the
terminating zero byte is already there.
Fixes coverity issue 157023.
|
|
This should make it easier to see which window is focused.
|
|
This should also fix coverity issue 157024.
|
|
This partially reverts f9e2b884c15919757651db8b10c033a344a19e75
further jumps after leaving visual mode should not break `gv`.
|
|
|
|
This reverts commit 54ca598fcccff8844bfbe494cd3b325b2fea487f.
The key handling functions are called too often. This for example
causes problems for `r` (replace char) in combination with :langmap.
|
|
|
|
|
|
We now use ' to refer to marks. Mark a is set using 'am and restored
using 'aM while this is slightly harder to type than ma and 'a it is
consistent with register usage for yank/put and allows a default
mark to be used which is handy for quick selection manipulation
primitives.
|
|
The key binding remain the same, but the selections are now stored on
a per-buffer basis.
|
|
This window local register holds the last active selections.
|
|
While the complete alias is added to the input queue,
the called key bindings should only see the keys they
have asked for.
Previously a mapping such as:
:map! normal gv \"^Sv
did not work as expected because the key binding for
the register did reject an invalid multi letter register
name.
Might also influence #581
|
|
This should keep the EOF markers visible when another option is enabled.
The whole UI option handling is a bit of a mess. In the longterm more of
the drawing code should be moved into Lua.
|
|
Conflicts:
view.c
view.h
|
|
Conflicts:
view.c
|
|
Merge overlapping selections.
Fix #582
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
This unifies cursors and selections. The cursor are now represendted
as singleton selections.
|
|
This fixes the internal representation as well as the path
displayed in the status bar. Previously opening a file in
the root directory e.g. /foo would be display as //foo.
|
|
|
|
|
|
|
|
When we have nothing to write to an external process, redirect stdin to
/dev/null instead of using a pipe which is immediately closed.
Some commands change their behavior when used in a shell pipeline.
As an example the following did not work as expected:
:< ag pattern
Fix #556
|
|
For interactive processes started using `:!` stdout is redirected
to stderr normally used by vis to draw its user interface.
For some reason the first byte written by the interactive application
is not being displayed. I suspect it has something to do with the
terminal state change. For now we are writing a dummy space (which
is never shown) ourself to ensure that the complete output is visible.
Fix #545
|
|
By default we only update the screen once per main loop iteration.
This might change in the future, but for now we force an update when a
redraw is explicitly requested.
|
|
|