Add support for title-casing letters (bug#24603)

mina86 · mina86 · commit 13d813b1a093 · 2017-04-06T20:54:57.000+02:00
* src/casefiddle.c (struct casing_context, prepare_casing_context): Add
titlecase_char_table member.  It’s set to the ‘titlecase’ Unicode
property table if capitalisation has been requested.
(case_character): Make use of the titlecase_char_table to title-case
initial characters when capitalising.

* test/src/casefiddle-tests.el (casefiddle-tests--characters,
casefiddle-tests-casing): Update test cases which are now passing.
diff --git a/etc/NEWS b/etc/NEWS
@@ -355,7 +355,7 @@ same as in modes where the character is not whitespace.
 Instead of only checking the modification time, Emacs now also checks
 the file's actual content before prompting the user.
 
-** Title case characters are properly converted to upper case.
+** Title case characters are properly cased (from and into).
 'upcase', 'upcase-region' et al. convert title case characters (such
 as the single character "ǲ") into their upper case form (such as "Ǳ").
 As a downside, 'capitalize' and 'upcase-initials' produce awkward
diff --git a/src/casefiddle.c b/src/casefiddle.c
@@ -33,6 +33,9 @@ enum case_action {CASE_UP, CASE_DOWN, CASE_CAPITALIZE, CASE_CAPITALIZE_UP};
 
 /* State for casing individual characters.  */
 struct casing_context {
+  /* A char-table with title-case character mappings or nil.  Non-nil implies
+     flag is CASE_CAPITALIZE or CASE_CAPITALIZE_UP.  */
+  Lisp_Object titlecase_char_table;
   /* User-requested action. */
   enum case_action flag;
   /* If true, function operates on a buffer as opposed to a string or character.
@@ -53,6 +56,8 @@ prepare_casing_context (struct casing_context *ctx,
   ctx->flag = flag;
   ctx->inbuffer = inbuffer;
   ctx->inword = flag == CASE_DOWN;
+  ctx->titlecase_char_table = (int)flag < (int)CASE_CAPITALIZE ? Qnil :
+    uniprop_table (intern_c_string ("titlecase"));
 
   /* If the case table is flagged as modified, rescan it.  */
   if (NILP (XCHAR_TABLE (BVAR (current_buffer, downcase_table))->extras[1]))
@@ -67,10 +72,16 @@ prepare_casing_context (struct casing_context *ctx,
 static int
 case_character (struct casing_context *ctx, int ch)
 {
+  Lisp_Object prop;
+
   if (ctx->inword)
     ch = ctx->flag == CASE_CAPITALIZE_UP ? ch : downcase (ch);
+  else if (!NILP (ctx->titlecase_char_table) &&
+	   CHARACTERP (prop = CHAR_TABLE_REF (ctx->titlecase_char_table, ch)))
+    ch = XFASTINT (prop);
   else
     ch = upcase(ch);
+
   if ((int) ctx->flag >= (int) CASE_CAPITALIZE)
     ctx->inword = SYNTAX (ch) == Sword &&
       (!ctx->inbuffer || ctx->inword || !syntax_prefix_flag_p (ch));
@@ -198,8 +209,8 @@ The argument object is not altered--the value is a copy.  */)
 
 DEFUN ("capitalize", Fcapitalize, Scapitalize, 1, 1, 0,
        doc: /* Convert argument to capitalized form and return that.
-This means that each word's first character is upper case
-and the rest is lower case.
+This means that each word's first character is converted to either
+title case or upper case, and the rest to lower case.
 The argument may be a character or string.  The result has the same type.
 The argument object is not altered--the value is a copy.  */)
   (Lisp_Object obj)
@@ -211,7 +222,8 @@ The argument object is not altered--the value is a copy.  */)
 
 DEFUN ("upcase-initials", Fupcase_initials, Supcase_initials, 1, 1, 0,
        doc: /* Convert the initial of each word in the argument to upper case.
-Do not change the other letters of each word.
+This means that each word's first character is converted to either
+title case or upper case, and the rest are left unchanged.
 The argument may be a character or string.  The result has the same type.
 The argument object is not altered--the value is a copy.  */)
   (Lisp_Object obj)
@@ -375,8 +387,8 @@ point and the mark is operated on.  */)
 
 DEFUN ("capitalize-region", Fcapitalize_region, Scapitalize_region, 2, 2, "r",
        doc: /* Convert the region to capitalized form.
-Capitalized form means each word's first character is upper case
-and the rest of it is lower case.
+This means that each word's first character is converted to either
+title case or upper case, and the rest to lower case.
 In programs, give two arguments, the starting and ending
 character positions to operate on.  */)
   (Lisp_Object beg, Lisp_Object end)
@@ -390,7 +402,8 @@ character positions to operate on.  */)
 DEFUN ("upcase-initials-region", Fupcase_initials_region,
        Supcase_initials_region, 2, 2, "r",
        doc: /* Upcase the initial of each word in the region.
-Subsequent letters of each word are not changed.
+This means that each word's first character is converted to either
+title case or upper case, and the rest are left unchanged.
 In programs, give two arguments, the starting and ending
 character positions to operate on.  */)
   (Lisp_Object beg, Lisp_Object end)
diff --git a/test/src/casefiddle-tests.el b/test/src/casefiddle-tests.el
@@ -63,13 +63,9 @@
     (?Ł ?Ł ?ł ?Ł)
     (?ł ?Ł ?ł ?Ł)
 
-    ;; FIXME(bug#24603): Commented ones are what we want.
-    ;;(?Ǆ ?Ǆ ?ǆ ?ǅ)
-    (?Ǆ ?Ǆ ?ǆ ?Ǆ)
-    ;;(?ǅ ?Ǆ ?ǆ ?ǅ)
-    (?ǅ ?Ǆ ?ǆ ?Ǆ)
-    ;;(?ǆ ?Ǆ ?ǆ ?ǅ)
-    (?ǆ ?Ǆ ?ǆ ?Ǆ)
+    (?Ǆ ?Ǆ ?ǆ ?ǅ)
+    (?ǅ ?Ǆ ?ǆ ?ǅ)
+    (?ǆ ?Ǆ ?ǆ ?ǅ)
 
     (?Σ ?Σ ?σ ?Σ)
     (?σ ?Σ ?σ ?Σ)
@@ -186,19 +182,19 @@
       ;; input     upper     lower    capitalize up-initials
       '(("Foo baR" "FOO BAR" "foo bar" "Foo Bar" "Foo BaR")
         ("Ⅷ ⅷ" "Ⅷ Ⅷ" "ⅷ ⅷ" "Ⅷ Ⅷ" "Ⅷ Ⅷ")
+        ;; "ǅUNGLA" is an unfortunate result but it’s really best we can
+        ;; do while still being consistent.  Hopefully, users only ever
+        ;; use upcase-initials on camelCase identifiers not real words.
+        ("ǄUNGLA" "ǄUNGLA" "ǆungla" "ǅungla" "ǅUNGLA")
+        ("ǅungla" "ǄUNGLA" "ǆungla" "ǅungla" "ǅungla")
+        ("ǆungla" "ǄUNGLA" "ǆungla" "ǅungla" "ǅungla")
         ;; FIXME(bug#24603): Everything below is broken at the moment.
         ;; Here’s what should happen:
-        ;;("ǄUNGLA" "ǄUNGLA" "ǆungla" "ǅungla" "ǅUNGLA")
-        ;;("ǅungla" "ǄUNGLA" "ǆungla" "ǅungla" "ǅungla")
-        ;;("ǆungla" "ǄUNGLA" "ǆungla" "ǅungla" "ǅungla")
         ;;("deﬁne" "DEFINE" "deﬁne" "Deﬁne" "Deﬁne")
         ;;("ﬁsh" "FIsh" "ﬁsh" "Fish" "Fish")
         ;;("Straße" "STRASSE" "straße" "Straße" "Straße")
         ;;("ΌΣΟΣ" "ΌΣΟΣ" "όσος" "Όσος" "Όσος")
         ;; And here’s what is actually happening:
-        ("ǄUNGLA" "ǄUNGLA" "ǆungla" "Ǆungla" "ǄUNGLA")
-        ("ǅungla" "ǄUNGLA" "ǆungla" "Ǆungla" "Ǆungla")
-        ("ǆungla" "ǄUNGLA" "ǆungla" "Ǆungla" "Ǆungla")
         ("deﬁne" "DEﬁNE" "deﬁne" "Deﬁne" "Deﬁne")
         ("ﬁsh" "ﬁSH" "ﬁsh" "ﬁsh" "ﬁsh")
         ("Straße" "STRAßE" "straße" "Straße" "Straße")
@@ -243,4 +239,21 @@
                 "\xef\xff\xef Zażółć GĘŚlą \xcf\xcf")))))))
 
 
+(ert-deftest casefiddle-tests-char-casing ()
+  ;;             input upcase downcase [titlecase]
+  (dolist (test '((?a ?A ?a) (?A ?A ?a)
+                  (?ł ?Ł ?ł) (?Ł ?Ł ?ł)
+                  (?ß ?ß ?ß) (?ẞ ?ẞ ?ß)
+                  (?ⅷ ?Ⅷ ?ⅷ) (?Ⅷ ?Ⅷ ?ⅷ)
+                  (?Ǆ ?Ǆ ?ǆ ?ǅ) (?ǅ ?Ǆ ?ǆ ?ǅ) (?ǆ ?Ǆ ?ǆ ?ǅ)))
+    (let ((ch (car test))
+          (up (nth 1 test))
+          (lo (nth 2 test))
+          (tc (or (nth 3 test) (nth 1 test))))
+      (should (eq up (upcase ch)))
+      (should (eq lo (downcase ch)))
+      (should (eq tc (capitalize ch)))
+      (should (eq tc (upcase-initials ch))))))
+
+
 ;;; casefiddle-tests.el ends here
