Recognize alphanumeric and underscore characters as words

This is equivalent to the ALTWERASE algorithm.

Based on previous work done by Jenz Guenther in PR #231 but further
enhanced to support UTF-8.

Makefile.am

@@ -16,7 +16,7 @@ TESTS=tests/01.t tests/02.t tests/03.t tests/04.t tests/05.t tests/06.t \
       tests/37.t tests/38.t tests/39.t tests/40.t tests/41.t tests/42.t \
       tests/43.t tests/44.t tests/45.t tests/46.t tests/47.t tests/48.t \
       tests/49.t tests/50.t tests/51.t tests/52.t tests/53.t tests/54.t \
-      tests/55.t tests/56.t tests/57.t tests/58.t
+      tests/55.t tests/56.t tests/57.t tests/58.t tests/59.t tests/60.t
 TEST_EXTENSIONS=.t
 T_LOG_COMPILER=$(top_srcdir)/tests/pick-test.sh
 AM_COLOR_TESTS=no

pick.1

@@ -74,8 +74,10 @@ Delete one character to the left of the cursor in the search query input field.
 .It Ic Delete | Ctrl-D
 Delete the character under the cursor in the search query input field.
 .It Ic Ctrl-W
-Delete to the beginning of the closest word to the left in the search query
-input field.
+Delete the word to the left of the cursor.
+A word is recognized as a sequence of characters for which either
+.Xr isalnum 3
+is true or the character is an underscore.
 .It Ic Ctrl-U
 Delete to the beginning of the line in the search query input field.
 .It Ic Ctrl-K

pick.c

@@ -16,6 +16,7 @@
 #include <termios.h>
 #include <unistd.h>
 #include <wchar.h>
+#include <wctype.h>
 
 #ifdef HAVE_NCURSESW_H
 #include <ncursesw/curses.h>
@@ -74,6 +75,7 @@ static __dead void		 handle_sigint(int);
 static void			 handle_sigwinch(int);
 static int			 isu8cont(unsigned char);
 static int			 isu8start(unsigned char);
+static int			 isword(const char *);
 static size_t			 min_match(const char *, size_t, ssize_t *,
 				    ssize_t *);
 static int			 print_choices(int, int);
@@ -285,7 +287,7 @@ selected_choice(void)
 {
 	size_t	cursor_position, i, j, length, xscroll;
 	char	buf[6];
-	int	choices_count, word_position;
+	int	choices_count;
 	int	selection = 0;
 	int	yscroll = 0;
 
@@ -413,22 +415,21 @@ selected_choice(void)
 			if (cursor_position == 0)
 				break;
 
-			for (word_position = cursor_position;;) {
-				while (isu8cont(query[--word_position]))
+			for (i = cursor_position; i > 0;) {
+				while (isu8cont(query[--i]))
 					continue;
-				if (word_position < 1)
+				if (isword(query + i))
 					break;
-				if (query[word_position] != ' '
-				    && query[word_position - 1] == ' ')
+			}
+			for (j = i; i > 0; i = j) {
+				while (isu8cont(query[--j]))
+					continue;
+				if (!isword(query + j))
 					break;
 			}
-			delete_between(
-			    query,
-			    query_length,
-			    word_position,
-			    cursor_position);
-			query_length -= cursor_position - word_position;
-			cursor_position = word_position;
+			delete_between(query, query_length, i, cursor_position);
+			query_length -= cursor_position - i;
+			cursor_position = i;
 			filter_choices();
 			selection = yscroll = 0;
 			break;
@@ -1037,3 +1038,19 @@ isu8start(unsigned char c)
 {
 	return MB_CUR_MAX > 1 && (c & (0x80 | 0x40)) == (0x80 | 0x40);
 }
+
+int
+isword(const char *s)
+{
+	wchar_t	wc;
+
+	switch (mbtowc(&wc, s, MB_CUR_MAX)) {
+	case -1:
+		mbtowc(NULL, NULL, MB_CUR_MAX);
+		/* FALLTHROUGH */
+	case 0:
+		return 0;
+	}
+
+	return iswalnum(wc) || wc == L'_';
+}

tests/13.t

@@ -1,7 +1,7 @@
 description: UTF-8 delete word
-keys: aa bå \027 bb \n # CTRL_W ENTER
+keys: aa\\ Åå \027 aa\\ aa \n # CTRL_W ENTER
 stdin:
-aa bb
-aa bå
+aa Åå aa
+aa Åå aa aa
 stdout:
-aa bb
+aa Åå aa aa

tests/59.t

@@ -0,0 +1,7 @@
+description: non alphanumeric characters are not recognized as part of a word
+keys: a/a \027 a/a \n # CTRL_W ENTER
+stdin:
+a/a
+a/a/a
+stdout:
+a/a/a

tests/60.t

@@ -0,0 +1,7 @@
+description: underscore is recognized as part of a word
+keys: aa_aa \027 bb \n # CTRL_W ENTER
+stdin:
+aa_aa_bb
+bb
+stdout:
+bb