teco_string_get_coord() returns character offsets now (refs #5)

* This is used for error messages (TECO macro stackframes),
  so it's important to display columns in characters.
* Program counters are in bytes and therefore everywhere gsize.
  This is by glib convention.

8 files changed, 21 insertions, 16 deletions

diff --git a/src/cmdline.h b/src/cmdline.h
index 78d101c..4aa862c 100644
--- a/src/cmdline.h
+++ b/src/cmdline.h
@@ -46,7 +46,7 @@ typedef struct {
 	gsize effective_len;
 
 	/** Program counter within the command-line macro */
-	guint pc;
+	gsize pc;
 
 	/**
 	 * Specifies whether the immediate editing modifier
diff --git a/src/error.c b/src/error.c
index 120ed1d..afa2ac1 100644
--- a/src/error.c
+++ b/src/error.c
@@ -37,13 +37,6 @@ guint teco_error_return_args = 0;
  */
 guint teco_error_pos = 0, teco_error_line = 0, teco_error_column = 0;
 
-void
-teco_error_set_coord(const gchar *str, guint pos)
-{
-	teco_error_pos = pos;
-	teco_string_get_coord(str, pos, &teco_error_line, &teco_error_column);
-}
-
 typedef enum {
 	TECO_FRAME_QREG,
 	TECO_FRAME_FILE,
diff --git a/src/error.h b/src/error.h
index 7543d02..b12ec80 100644
--- a/src/error.h
+++ b/src/error.h
@@ -143,7 +143,11 @@ teco_error_return_set(GError **error, guint args)
 
 extern guint teco_error_pos, teco_error_line, teco_error_column;
 
-void teco_error_set_coord(const gchar *str, guint pos);
+static inline void
+teco_error_set_coord(const gchar *str, gsize pos)
+{
+	teco_string_get_coord(str, pos, &teco_error_pos, &teco_error_line, &teco_error_column);
+}
 
 void teco_error_display_short(const GError *error);
 void teco_error_display_full(const GError *error);
diff --git a/src/parser.h b/src/parser.h
index ae2cb9b..fe7f559 100644
--- a/src/parser.h
+++ b/src/parser.h
@@ -440,7 +440,7 @@ typedef enum {
 struct teco_machine_main_t {
 	teco_machine_t parent;
 
-	gint macro_pc;
+	gsize macro_pc;
 
 	/**
 	 * Aliases bitfield with an integer.
diff --git a/src/string-utils.c b/src/string-utils.c
index d9b12e0..b284760 100644
--- a/src/string-utils.c
+++ b/src/string-utils.c
@@ -55,13 +55,20 @@ teco_string_echo(const gchar *str, gsize len)
 	return ret;
 }
 
-/** @memberof teco_string_t */
+/**
+ * Get character coordinates for a given byte index.
+ *
+ * The given string must be valid UTF-8.
+ *
+ * @memberof teco_string_t
+ */
 void
-teco_string_get_coord(const gchar *str, guint pos, guint *line, guint *column)
+teco_string_get_coord(const gchar *str, gsize off, guint *pos, guint *line, guint *column)
 {
+	*pos = 0;
 	*line = *column = 1;
 
-	for (guint i = 0; i < pos; i++) {
+	for (guint i = 0; i < off; i = g_utf8_next_char(str+i) - str) {
 		switch (str[i]) {
 		case '\r':
 			if (str[i+1] == '\n')
@@ -75,6 +82,7 @@ teco_string_get_coord(const gchar *str, guint pos, guint *line, guint *column)
 			(*column)++;
 			break;
 		}
+		(*pos)++;
 	}
 }
 
diff --git a/src/string-utils.h b/src/string-utils.h
index 806140e..ebe25d5 100644
--- a/src/string-utils.h
+++ b/src/string-utils.h
@@ -162,7 +162,7 @@ void undo__teco_string_truncate(teco_string_t *, gsize);
 
 gchar *teco_string_echo(const gchar *str, gsize len);
 
-void teco_string_get_coord(const gchar *str, guint pos, guint *line, guint *column);
+void teco_string_get_coord(const gchar *str, gsize off, guint *pos, guint *line, guint *column);
 
 typedef gsize (*teco_string_diff_t)(const teco_string_t *a, const gchar *b, gsize b_len);
 gsize teco_string_diff(const teco_string_t *a, const gchar *b, gsize b_len);
diff --git a/src/undo.c b/src/undo.c
index dc54c7a..bc12107 100644
--- a/src/undo.c
+++ b/src/undo.c
@@ -112,7 +112,7 @@ teco_undo_push_size(teco_undo_action_t action_cb, gsize size)
 }
 
 void
-teco_undo_pop(gint pc)
+teco_undo_pop(gsize pc)
 {
 	while ((gint)teco_undo_heads->len > pc) {
 		teco_undo_token_t *top =
diff --git a/src/undo.h b/src/undo.h
index 9715c7a..1d1d6fb 100644
--- a/src/undo.h
+++ b/src/undo.h
@@ -243,5 +243,5 @@ TECO_DECLARE_UNDO_SCALAR(gconstpointer);
 
 /** @} */
 
-void teco_undo_pop(gint pc);
+void teco_undo_pop(gsize pc);
 void teco_undo_clear(void);