| Age | Commit message (Collapse) | Author |
|
In some cases search is triggered not from the minibuffer, in which
case minibuffer-selected-window will not accurately capture the window
in which search highlights should be created.
|
|
Reverts part of 1d6ba80.
|
|
|
|
Currently evil-goto-definition-imenu will only look in top level entries and
ignores everything in a submenu. Unfortunately this means that many definitions
that could be found are being skipped.
|
|
Honor isearch filter predicates
|
|
Revert "Enable undo-tree-mode only when evil-mode is turned on"
|
|
|
|
Enable undo-tree-mode only when evil-mode is turned on
|
|
|
|
|
|
When setting a new global map, the tab-bar entry must be copied to the new
map from the old one to avoid hiding the tab bar.
* evil-common.el (evil-read-key): Preserve the tab-bar entry from the old
global map.
|
|
|
|
|
|
Fix behavior of evil-scroll-up and evil-scroll-down again
|
|
This is handled implicitly by search-func now.
The variable isearch-filter-predicates was reverted and never saw
light of day in an official Emacs release.
|
|
|
|
|
|
The PR #1154 is supposed to fix the issue about evil-scroll-up
and evil-scroll-down. It works nicely in most scenarios.
However, when the header line is really tall,
evil-header-line-height will give an incorrect result,
making the bug that is supposed to be fixed by the mentioned PR
reappear again.
This PR switches evil-header-line-height to Emacs' built-in
window-header-line-height which corrects the mistake.
|
|
|
|
Filtering our matches before passing them to isearch prevents isearch
from entering an infinite loop if wrapped search is turned on and all
the matches are filtered out.
|
|
This reverts commit e47db185aa3e6a15eba6e34847f0b28aa7055db8.
|
|
Better integration of window manipulation commands
|
|
Fix #1180
|
|
|
|
|
|
This enables use of e.g. C-u p to paste four times.
Fixes #1199
|
|
|
|
|
|
|
|
Update docs
|
|
|
|
|
|
|
|
Documentation overhaul
|
|
|
|
|
|
Commands like `evil-window-rotate-upwards` and `evil-window-move-very-top` only
consider the buffer in each window when manipulating the window tree, but if a
window has specific window properties that need to be maintained (like those in
`window-persistent-parameters`) that information is lost.
Refactor the window manipulation commands to use `window-state-get` and
`window-state-put` to retrieved and restore the state of the window tree so
that window properties in each window are maintained.
Also, integrate the window manipulation commands with side windows, i.e. windows
with a non-nil `window-side` window parameter, by only manipulating the part of
the window tree that are not side windows.
* Add the macro `evil-save-side-windows` which removes all side windows on
entering the macro environment and restores the side windows leaving.
* Add the function `evil-move-window` which does the work of
`evil-window-move-far-left`, `evil-window-move-very-top`,
`evil-window-move-far-right`, and `evil-window-move-very-bottom`. These
functions were all identical, the only difference being which window was used
to restore the `selected-window`.
* Use `window-state-(get|put)` to get the window tree and to restore it. This
makes `evil-get-buffer-tree` and `evil-restore-window-tree` unnecessary and
they are removed. Note that `window-state-(get|put)` also handles window
parameters.
* In addition to ignoring side windows when moving a window, ignore them in
`evil-window-rotate-downwards` and `evil-window-rotate-upwards`.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Fix bounds of wrapped search
|
|
Do not perform the second search if the original search bound was
set, as it must necessarily be bypassed.
Use the original starting point for the search as the bound for the
second search.
Fixes #843
Thanks to Miciah Dashiel Butler Masters
|
|
Add evil-want-C-u-delete option
|
|
|
|
|
|
|
|
Version 1.2.15 was accidentally tagged as 1.12.15, and released as
such by MELPA Stable, so now I guess we have to run with it.
|