|
@@ -101,7 +101,24 @@ Debug info is printed to the *Messages* buffer."
|
|
(when ido-cr+-debug-mode
|
|
(when ido-cr+-debug-mode
|
|
(apply #'message (concat "ido-completing-read+: " format-string) args)))
|
|
(apply #'message (concat "ido-completing-read+: " format-string) args)))
|
|
|
|
|
|
-;;; Core code
|
|
|
|
|
|
+;;; Ido variables
|
|
|
|
+
|
|
|
|
+;; For unknown reasons, these variables need to be re-declared here to
|
|
|
|
+;; silence byte-compiler warnings, despite already being declared in
|
|
|
|
+;; ido.el.
|
|
|
|
+
|
|
|
|
+;;;###autoload
|
|
|
|
+(defvar ido-context-switch-command nil
|
|
|
|
+ "Variable holding the command used for switching to another completion mode.
|
|
|
|
+
|
|
|
|
+This variable is originally declared in `ido.el', but it is not
|
|
|
|
+given a value (or a docstring). This documentation comes from a
|
|
|
|
+re-declaration in `ido-completing-read+.el' that initializes it
|
|
|
|
+to nil, which should suppress some byte-compilation warnings in
|
|
|
|
+Emacs 25. Setting another package's variable is not safe in
|
|
|
|
+general, but in this case it should be, because ido always
|
|
|
|
+let-binds this variable before using it, so the initial value
|
|
|
|
+shouldn't matter.")
|
|
|
|
|
|
(defvar ido-cur-list nil
|
|
(defvar ido-cur-list nil
|
|
"Internal ido variable.
|
|
"Internal ido variable.
|
|
@@ -115,6 +132,8 @@ general, but in this case it should be, because ido always
|
|
let-binds this variable before using it, so the initial value
|
|
let-binds this variable before using it, so the initial value
|
|
shouldn't matter.")
|
|
shouldn't matter.")
|
|
|
|
|
|
|
|
+;;; Core code
|
|
|
|
+
|
|
;;;###autoload
|
|
;;;###autoload
|
|
(defvar ido-cr+-minibuffer-depth -1
|
|
(defvar ido-cr+-minibuffer-depth -1
|
|
"Minibuffer depth of the most recent ido-cr+ activation.
|
|
"Minibuffer depth of the most recent ido-cr+ activation.
|
|
@@ -589,19 +608,6 @@ See `ido-cr+-current-command'."
|
|
ad-do-it))
|
|
ad-do-it))
|
|
|
|
|
|
;; Fallback on magic C-f and C-b
|
|
;; Fallback on magic C-f and C-b
|
|
-;;;###autoload
|
|
|
|
-(defvar ido-context-switch-command nil
|
|
|
|
- "Variable holding the command used for switching to another completion mode.
|
|
|
|
-
|
|
|
|
-This variable is originally declared in `ido.el', but it is not
|
|
|
|
-given a value (or a docstring). This documentation comes from a
|
|
|
|
-re-declaration in `ido-completing-read+.el' that initializes it
|
|
|
|
-to nil, which should suppress some byte-compilation warnings in
|
|
|
|
-Emacs 25. Setting another package's variable is not safe in
|
|
|
|
-general, but in this case it should be, because ido always
|
|
|
|
-let-binds this variable before using it, so the initial value
|
|
|
|
-shouldn't matter.")
|
|
|
|
-
|
|
|
|
(defadvice ido-magic-forward-char (before ido-cr+-fallback activate)
|
|
(defadvice ido-magic-forward-char (before ido-cr+-fallback activate)
|
|
"Allow falling back in ido-completing-read+."
|
|
"Allow falling back in ido-completing-read+."
|
|
(when (ido-cr+-active)
|
|
(when (ido-cr+-active)
|
|
@@ -632,7 +638,7 @@ sets up C-j to be equivalent to TAB in the same situation."
|
|
;; Require-match is non-nil
|
|
;; Require-match is non-nil
|
|
(with-no-warnings ido-require-match)
|
|
(with-no-warnings ido-require-match)
|
|
;; Current text is not a complete choice
|
|
;; Current text is not a complete choice
|
|
- (not (member ido-text (with-no-warnings ido-cur-list))))
|
|
|
|
|
|
+ (not (member ido-text ido-cur-list)))
|
|
(progn
|
|
(progn
|
|
(ido-cr+--debug-message
|
|
(ido-cr+--debug-message
|
|
"Overriding C-j behavior for require-match: performing completion instead of exiting with current text. (This might still exit with a match if `ido-confirm-unique-completion' is nil)")
|
|
"Overriding C-j behavior for require-match: performing completion instead of exiting with current text. (This might still exit with a match if `ido-confirm-unique-completion' is nil)")
|