Distinct between inner and outer word text objects

5 files changed, 98 insertions, 19 deletions

diff --git a/README b/README
index 411f470..108ff4d 100644
--- a/README
+++ b/README
@@ -371,7 +371,7 @@ and their current support in vis.
    p  paragraph
    [,], (,), {,}, <,>, ", ', `         block enclosed by these symbols
 
-  For word, sentence and paragraph there is no difference between the
+  For sentence and paragraph there is no difference between the
   inner and normal variants.
 
  Modes
diff --git a/config.def.h b/config.def.h
index 16fd674..2954ea9 100644
--- a/config.def.h
+++ b/config.def.h
@@ -160,8 +160,8 @@ static KeyBinding vis_movements[] = {
 };
 
 static KeyBinding vis_textobjs[] = {
-	{ { NONE('a'), NONE('w')  }, textobj, { .i = TEXT_OBJ_WORD                 } },
-	{ { NONE('a'), NONE('W')  }, textobj, { .i = TEXT_OBJ_LONGWORD             } },
+	{ { NONE('a'), NONE('w')  }, textobj, { .i = TEXT_OBJ_OUTER_WORD           } },
+	{ { NONE('a'), NONE('W')  }, textobj, { .i = TEXT_OBJ_OUTER_LONGWORD       } },
 	{ { NONE('a'), NONE('s')  }, textobj, { .i = TEXT_OBJ_SENTENCE             } },
 	{ { NONE('a'), NONE('p')  }, textobj, { .i = TEXT_OBJ_PARAGRAPH            } },
 	{ { NONE('a'), NONE('[')  }, textobj, { .i = TEXT_OBJ_OUTER_SQUARE_BRACKET } },
@@ -181,8 +181,8 @@ static KeyBinding vis_textobjs[] = {
 };
 
 static KeyBinding vis_inner_textobjs[] = {
-	{ { NONE('i'), NONE('w')  }, textobj, { .i = TEXT_OBJ_WORD                 } },
-	{ { NONE('i'), NONE('W')  }, textobj, { .i = TEXT_OBJ_LONGWORD             } },
+	{ { NONE('i'), NONE('w')  }, textobj, { .i = TEXT_OBJ_INNER_WORD           } },
+	{ { NONE('i'), NONE('W')  }, textobj, { .i = TEXT_OBJ_INNER_LONGWORD       } },
 	{ { NONE('i'), NONE('s')  }, textobj, { .i = TEXT_OBJ_SENTENCE             } },
 	{ { NONE('i'), NONE('p')  }, textobj, { .i = TEXT_OBJ_PARAGRAPH            } },
 	{ { NONE('i'), NONE('[')  }, textobj, { .i = TEXT_OBJ_INNER_SQUARE_BRACKET } },
diff --git a/text-objects.c b/text-objects.c
index a8aef0a..44515e1 100644
--- a/text-objects.c
+++ b/text-objects.c
@@ -18,6 +18,10 @@
 #include "text-objects.h"
 #include "util.h"
 
+#define isboundry is_word_boundry
+
+/* TODO: reduce code duplication? */
+
 Filerange text_object_longword(Text *txt, size_t pos) {
 	Filerange r;
 	char c, prev = '0', next = '0';
@@ -28,6 +32,39 @@ Filerange text_object_longword(Text *txt, size_t pos) {
 		text_iterator_byte_next(&it, NULL);
 	text_iterator_byte_next(&it, &next);
 	if (isspace(c)) {
+		/* middle of two words */
+		r.start = text_char_next(txt, text_longword_end_prev(txt, pos));
+		r.end = text_longword_start_next(txt, pos);
+	} else if (isspace(prev) && isspace(next)) {
+		/* on a single character */
+		r.start = pos;
+		r.end = text_char_next(txt, pos);
+	} else if (isspace(prev)) {
+		/* at start of a word */
+		r.start = pos;
+		r.end = text_char_next(txt, text_longword_end_next(txt, pos));
+	} else if (isspace(next)) {
+		/* at end of a word */
+		r.start = text_longword_start_prev(txt, pos);
+		r.end = text_char_next(txt, pos);
+	} else {
+		/* in the middle of a word */
+		r.start = text_longword_start_prev(txt, pos);
+		r.end = text_char_next(txt, text_longword_end_next(txt, pos));
+	}
+	return r;
+}
+
+Filerange text_object_longword_outer(Text *txt, size_t pos) {
+	Filerange r;
+	char c, prev = '0', next = '0';
+	Iterator it = text_iterator_get(txt, pos);
+	if (!text_iterator_byte_get(&it, &c))
+		return text_range_empty();
+	if (text_iterator_byte_prev(&it, &prev))
+		text_iterator_byte_next(&it, NULL);
+	text_iterator_byte_next(&it, &next);
+	if (isspace(c)) {
 		/* middle of two words, include leading white space */
 		r.start = text_char_next(txt, text_longword_end_prev(txt, pos));
 		r.end = text_char_next(txt, text_longword_end_next(txt, pos));
@@ -52,7 +89,6 @@ Filerange text_object_longword(Text *txt, size_t pos) {
 }
 
 Filerange text_object_word(Text *txt, size_t pos) {
-	#define isboundry is_word_boundry
 	Filerange r;
 	char c, prev = '0', next = '0';
 	Iterator it = text_iterator_get(txt, pos);
@@ -62,14 +98,11 @@ Filerange text_object_word(Text *txt, size_t pos) {
 		text_iterator_byte_next(&it, NULL);
 	text_iterator_byte_next(&it, &next);
 	if (isspace(c)) {
-		/* middle of two words, include leading white space */
 		r.start = text_char_next(txt, text_word_end_prev(txt, pos));
-		r.end = text_word_end_next(txt, pos);
-		if (!text_byte_get(txt, r.end, &c) && !isboundry(c))
-			r.end = text_char_next(txt, r.end);
+		r.end = text_word_start_next(txt, pos);
 	} else if (isboundry(prev) && isboundry(next)) {
 		if (isboundry(c)) {
-			r.start = text_word_end_prev(txt, pos);
+			r.start = text_char_next(txt, text_word_end_prev(txt, pos));
 			r.end = text_char_next(txt, text_word_end_next(txt, pos));
 		} else {
 			/* on a single character */
@@ -93,6 +126,45 @@ Filerange text_object_word(Text *txt, size_t pos) {
 	return r;
 }
 
+Filerange text_object_word_outer(Text *txt, size_t pos) {
+	Filerange r;
+	char c, prev = '0', next = '0';
+	Iterator it = text_iterator_get(txt, pos);
+	if (!text_iterator_byte_get(&it, &c))
+		return text_range_empty();
+	if (text_iterator_byte_prev(&it, &prev))
+		text_iterator_byte_next(&it, NULL);
+	text_iterator_byte_next(&it, &next);
+	if (isspace(c)) {
+		/* middle of two words, include leading white space */
+		r.start = text_char_next(txt, text_word_end_prev(txt, pos));
+		r.end = text_word_end_next(txt, pos);
+	} else if (isboundry(prev) && isboundry(next)) {
+		if (isboundry(c)) {
+			r.start = text_char_next(txt, text_word_end_prev(txt, pos));
+			r.end = text_word_start_next(txt, text_word_end_next(txt, pos));
+		} else {
+			/* on a single character */
+			r.start = pos;
+			r.end = text_char_next(txt, pos);
+		}
+	} else if (isboundry(prev)) {
+		/* at start of a word */
+		r.start = pos;
+		r.end = text_word_start_next(txt, text_word_end_next(txt, pos));
+	} else if (isboundry(next)) {
+		/* at end of a word */
+		r.start = text_word_start_prev(txt, pos);
+		r.end = text_word_start_next(txt, pos);
+	} else {
+		/* in the middle of a word */
+		r.start = text_word_start_prev(txt, pos);
+		r.end = text_word_start_next(txt, text_word_end_next(txt, pos));
+	}
+
+	return r;
+}
+
 Filerange text_object_line(Text *txt, size_t pos) {
 	Filerange r;
 	r.start = text_line_begin(txt, pos);
diff --git a/text-objects.h b/text-objects.h
index cb11f41..a4babb5 100644
--- a/text-objects.h
+++ b/text-objects.h
@@ -9,12 +9,15 @@
 #include <stddef.h>
 #include "text.h"
 
-/* word which happens to be at pos, includes trailing white spaces. if at pos
- * happens to be a whitespace include all neighbouring leading whitespaces
- * and the following word. */
+/* word which happens to be at pos without any neighbouring white spaces */
 Filerange text_object_word(Text*, size_t pos);
-/* same semantics as above but for a longword (i.e. delimited by whitespaces) */
+/* includes trailing white spaces. if at pos happens to be a white space
+ * include all neighbouring leading white spaces and the following word. */
+Filerange text_object_word_outer(Text*, size_t pos);
+/* same semantics as above but for a longword (i.e. delimited by white spaces) */
 Filerange text_object_longword(Text*, size_t pos);
+Filerange text_object_longword_outer(Text*, size_t pos);
+
 Filerange text_object_line(Text*, size_t pos);
 Filerange text_object_sentence(Text*, size_t pos);
 Filerange text_object_paragraph(Text*, size_t pos);
diff --git a/vis.c b/vis.c
index 9d708ec..7f3b9ca 100644
--- a/vis.c
+++ b/vis.c
@@ -323,8 +323,10 @@ static Movement moves[] = {
 
 /* these can be passed as int argument to textobj(&(const Arg){ .i = TEXT_OBJ_* }) */
 enum {
-	TEXT_OBJ_WORD,
-	TEXT_OBJ_LONGWORD,
+	TEXT_OBJ_INNER_WORD,
+	TEXT_OBJ_OUTER_WORD,
+	TEXT_OBJ_INNER_LONGWORD,
+	TEXT_OBJ_OUTER_LONGWORD,
 	TEXT_OBJ_LINE_UP,
 	TEXT_OBJ_LINE_DOWN,
 	TEXT_OBJ_SENTENCE,
@@ -346,8 +348,10 @@ enum {
 };
 
 static TextObject textobjs[] = {
-	[TEXT_OBJ_WORD]                 = { text_object_word                  },
-	[TEXT_OBJ_LONGWORD]             = { text_object_longword              },
+	[TEXT_OBJ_INNER_WORD]           = { text_object_word                  },
+	[TEXT_OBJ_OUTER_WORD]           = { text_object_word_outer            },
+	[TEXT_OBJ_INNER_LONGWORD]       = { text_object_longword              },
+	[TEXT_OBJ_OUTER_LONGWORD]       = { text_object_longword_outer        },
 	[TEXT_OBJ_LINE_UP]              = { text_object_line                  },
 	[TEXT_OBJ_LINE_DOWN]            = { text_object_line                  },
 	[TEXT_OBJ_SENTENCE]             = { text_object_sentence              },