BREAKING CHANGE: As the name says, this is a whole new package. There
will be functional hiccups until I've achieved behavioral parity with
the old workspaces module.
When Emacs is started as a daemon and no frames exist, `emacsclient -c`
should simply switch to the main workspace. However, before
`+workspaces-associate-frame-fn` is called, the new frame is already
added to the frame list, which causes `persp-frame-list-without-daemon`
to always return a non-empty list. A new workspace will be created
instead of switching to the main one.
A "workspace already exists" error is thrown if trying to switch to
project workspace that already exists. This changes it to switch to the
workspace instead.
Fix: #8368
If you used `projectile-switch-project`, the workspaces module would
open a new workspace named after the project, but if a workspace named
after that project already exists, it is hijacked (e.g. two projects in
'some-project/src' and 'another-project/src' would get the same name:
"src"). This commit uniquifies their names so this doesn't happen, and
also embeds a +workspace-project persp parameter in the workspace so
that they can be disamiguated later (this association is lost if you
rename the workspace manually, however).
Fix: #7637
Otherwise, each time `+workspace-save` is called, all prior saved
workspaces would be deleted from `+workspaces-data-file` before
inserting the new one.
Doom removes the "none" workspace from persp-mode's default list of
perspectives, because it is a special case that doesn't behave
identically to other persps in the list, making it a pain to deal with.
This worked fine up until a change to `persp-update-names-cache` (in
Bad-ptr/persp-mode.el@0d6cacc) made it re-insert this "none" workspace
into `persp-names-cache` whenever it is called, undoing our hack; this
commit adapts to that change.
Ref: Bad-ptr/persp-mode.el@0d6caccab3
Ref: Bad-ptr/persp-mode.el@b2e68f97cbFix: #7986
Amend: 7f3412e317
For deleting saved workspaces (saved with `+workspace/save` or
`+workspace-save`).
Also binds `SPC TAB D` (for evil users) and `C-c w K` (for non-evil
users) to it.
Fix: #4399Close: #7869
Co-authored-by: sriramsk1999 <sriramsk1999@users.noreply.github.com>
To be more consistent with other uses of the verb in the Emacs
ecosystem. Also done in preparation for a new +workspaces/delete command
for #7869.
Ref: #7869
Due to upstream changes in ivy-rich, +workspace/switch-to throws a
`wrong-type-argument listp leaf` error. As I plan to phase out Ivy
support (and the Ivy module) in the long term, I'll simply remove
ivy (and ivy-rich) integration in the workspaces module, rather than
update it.
Fix: #7499Fix: #7173
This change uses completing-read to get the new name from the user when
renaming workspaces, to allow them to more easily make a small change to
the existing name of a workspace.
featurep! will be renamed modulep! in the future, so it's been
deprecated. They have identical interfaces, and can be replaced without
issue.
featurep! was never quite the right name for this macro. It implied that
it had some connection to featurep, which it doesn't (only that it was
similar in purpose; still, Doom modules are not features). To undo such
implications and be consistent with its namespace (and since we're
heading into a storm of breaking changes with the v3 release anyway),
now was the best opportunity to begin the transition.
Due to magit now using server-done instead of server-edit as of
magit/magit@5c02910, which calls server-done-hook, which the workspaces
module has a hook on to kill auto-generated, daemon-spawned frames,
causing workspaces to be prematurely killed when commiting in magit.
When a user tries to switch to a named workspace that doesn't exist,
creat it, instead of just failing.
Signed-off-by: Rudi Grinberg <me@rgrinberg.com>
