revised and improved the Unicode glyph-to-byte conversion heuristics

Previously almost all glyph-to-byte offset conversions consulted
Scintilla's line index and counted characters on the resulting line.
For instance a simple expression like `.+1J` would scan the same line
twice completely, which would be very slow on pathologically long lines.
Even insertions did that due to having to update the ^Y ranges.
If you repeat such an operation over all characters as in `<.+1:J;>`
you would have complexity O(n^2) for n = line length.
Only commands with an explicit relative nature like `C` and `A` would
use teco_view_glyph2bytes_relative() which scans beginning at dot
as long as the relative movement is less than 1024 glyphs.

Wit the new heuristics almost all glyph-to-byte and byte-to-glyph
conversions can make use of that optimization.
This requires that dot must at all times be known in glyphs as well -
the byte position is managed by Scintilla (SCI_GETCURRENTPOS).
We therefore introduced teco_current_doc_set_dot() and
teco_current_doc_get_dot() to update dot in the current buffer or
Q-Register -- it cannot be stored along with the view since
Q-Registers share a single view.
A number of auxiliary functions have been introduced for
converting relative to a known (glyphs,bytes) offset pair
and for converting absolute and relative positions with regard
to the current doc and SCI_GETCURRENTPOS position.
Of course this is error-prone since the glyph and dot positions
are interdependant - they must always be kept in sync.

With these new optimizations even pathologically long lines can
(usually) be managed even in UTF-8 documents.
It does not address slow-downs in Scintilla's line layout, yet.
grosciteco.tes for instance runs twice as fast now.

1 files changed, 57 insertions, 15 deletions

diff --git a/src/move-commands.c b/src/move-commands.c
index 45afc4e..9ef893e 100644
--- a/src/move-commands.c
+++ b/src/move-commands.c
@@ -58,12 +58,13 @@ teco_state_start_jump(teco_machine_main_t *ctx, GError **error)
 	if (!teco_expressions_pop_num_calc(&v, 0, error))
 		return;
 
-	gssize pos = teco_interface_glyphs2bytes(v);
-	if (pos >= 0) {
+	gssize next_pos = teco_interface_glyphs2bytes_absdot(v);
+	if (next_pos >= 0) {
 		if (teco_current_doc_must_undo())
 			undo__teco_interface_ssm(SCI_GOTOPOS,
 			                         teco_interface_ssm(SCI_GETCURRENTPOS, 0, 0), 0);
-		teco_interface_ssm(SCI_GOTOPOS, pos, 0);
+		teco_interface_ssm(SCI_GOTOPOS, next_pos, 0);
+		teco_current_doc_set_dot(v);
 
 		if (teco_machine_main_eval_colon(ctx) > 0)
 			teco_expressions_push(TECO_SUCCESS);
@@ -78,14 +79,15 @@ teco_state_start_jump(teco_machine_main_t *ctx, GError **error)
 static teco_bool_t
 teco_move_chars(teco_int_t n)
 {
-	sptr_t pos = teco_interface_ssm(SCI_GETCURRENTPOS, 0, 0);
-	gssize next_pos = teco_interface_glyphs2bytes_relative(pos, n);
+	gssize next_pos = teco_interface_glyphs2bytes_reldot(n);
 	if (next_pos < 0)
 		return TECO_FAILURE;
 
-	teco_interface_ssm(SCI_GOTOPOS, next_pos, 0);
 	if (teco_current_doc_must_undo())
-		undo__teco_interface_ssm(SCI_GOTOPOS, pos, 0);
+		undo__teco_interface_ssm(SCI_GOTOPOS, teco_interface_ssm(SCI_GETCURRENTPOS, 0, 0), 0);
+
+	teco_interface_ssm(SCI_GOTOPOS, next_pos, 0);
+	teco_current_doc_set_dot(teco_current_doc_get_dot() + n);
 
 	return TECO_SUCCESS;
 }
@@ -151,7 +153,15 @@ teco_move_lines(teco_int_t n)
 	if (!teco_validate_line(line))
 		return TECO_FAILURE;
 
-	teco_interface_ssm(SCI_GOTOLINE, line, 0);
+	sptr_t next_pos = teco_interface_ssm(SCI_POSITIONFROMLINE, line, 0);
+	/*
+	 * FIXME: Perhaps it would be more efficient to fall back to the
+	 * line index cache via teco_view_bytes2glyphs_unicode() since next_pos
+	 * is guaranteed to fall on line starts.
+	 */
+	teco_current_doc_set_dot(teco_interface_bytes2glyphs_absdot(next_pos));
+	teco_interface_ssm(SCI_GOTOPOS, next_pos, 0);
+
 	if (teco_current_doc_must_undo())
 		undo__teco_interface_ssm(SCI_GOTOPOS, pos, 0);
 
@@ -388,6 +398,8 @@ teco_state_start_words(teco_machine_main_t *ctx, const gchar *cmd, gint factor,
 	if (rc) {
 		if (teco_current_doc_must_undo())
 			undo__teco_interface_ssm(SCI_GOTOPOS, pos, 0);
+
+		teco_current_doc_set_dot(teco_interface_bytes2glyphs_absdot(word_pos));
 		teco_interface_ssm(SCI_GOTOPOS, word_pos, 0);
 	}
 
@@ -457,6 +469,9 @@ teco_state_start_delete_words(teco_machine_main_t *ctx, const gchar *cmd, gint f
 	if (rc && start_pos != end_pos) {
 		g_assert(start_pos < end_pos);
 
+		if (start_pos < pos)
+			teco_current_doc_set_dot(teco_interface_bytes2glyphs_absdot(start_pos));
+
 		teco_interface_ssm(SCI_BEGINUNDOACTION, 0, 0);
 		teco_interface_ssm(SCI_DELETERANGE, start_pos, end_pos-start_pos);
 		teco_interface_ssm(SCI_ENDUNDOACTION, 0, 0);
@@ -487,8 +502,10 @@ teco_state_start_kill(teco_machine_main_t *ctx, const gchar *cmd, gboolean by_li
 	if (!teco_expressions_eval(FALSE, error))
 		return FALSE;
 
+	sptr_t pos = teco_interface_ssm(SCI_GETCURRENTPOS, 0, 0);
+
 	if (teco_expressions_args() <= 1) {
-		from = teco_interface_ssm(SCI_GETCURRENTPOS, 0, 0);
+		from = pos;
 		if (by_lines) {
 			teco_int_t line;
 			if (!teco_expressions_pop_num_calc(&line, teco_num_sign, error))
@@ -496,13 +513,24 @@ teco_state_start_kill(teco_machine_main_t *ctx, const gchar *cmd, gboolean by_li
 			line += teco_interface_ssm(SCI_LINEFROMPOSITION, from, 0);
 			len = teco_interface_ssm(SCI_POSITIONFROMLINE, line, 0) - from;
 			rc = teco_bool(teco_validate_line(line));
+
+			/*
+			 * FIXME: Perhaps it would be more efficient to fall back to the
+			 * line index cache via teco_view_bytes2glyphs_unicode() since from+len
+			 * is guaranteed to fall on line starts.
+			 */
+			if (teco_is_success(rc) && len < 0)
+				teco_current_doc_set_dot(teco_interface_bytes2glyphs_reldot(len));
 		} else {
 			teco_int_t len_glyphs;
 			if (!teco_expressions_pop_num_calc(&len_glyphs, teco_num_sign, error))
 				return FALSE;
-			gssize to = teco_interface_glyphs2bytes_relative(from, len_glyphs);
+			gssize to = teco_interface_glyphs2bytes_reldot(len_glyphs);
 			rc = teco_bool(to >= 0);
 			len = to-from;
+
+			if (teco_is_success(rc) && len_glyphs < 0)
+				teco_current_doc_set_dot(teco_current_doc_get_dot() + len_glyphs);
 		}
 		if (len < 0) {
 			len *= -1;
@@ -510,11 +538,18 @@ teco_state_start_kill(teco_machine_main_t *ctx, const gchar *cmd, gboolean by_li
 		}
 	} else {
 		teco_int_t to_glyphs = teco_expressions_pop_num(0);
-		gssize to = teco_interface_glyphs2bytes(to_glyphs);
+		gssize to = teco_interface_glyphs2bytes_absdot(to_glyphs);
 		teco_int_t from_glyphs = teco_expressions_pop_num(0);
-		from = teco_interface_glyphs2bytes(from_glyphs);
+		from = teco_interface_glyphs2bytes_absdot(from_glyphs);
 		len = to - from;
 		rc = teco_bool(len >= 0 && from >= 0 && to >= 0);
+
+		if (teco_is_success(rc)) {
+			if (to < pos)
+				teco_current_doc_set_dot(teco_current_doc_get_dot() - (to_glyphs-from_glyphs));
+			else if (from < pos)
+				teco_current_doc_set_dot(from_glyphs);
+		}
 	}
 
 	if (teco_machine_main_eval_colon(ctx) > 0) {
@@ -528,7 +563,6 @@ teco_state_start_kill(teco_machine_main_t *ctx, const gchar *cmd, gboolean by_li
 		return TRUE;
 
 	if (teco_current_doc_must_undo()) {
-		sptr_t pos = teco_interface_ssm(SCI_GETCURRENTPOS, 0, 0);
 		undo__teco_interface_ssm(SCI_GOTOPOS, pos, 0);
 		undo__teco_interface_ssm(SCI_UNDO, 0, 0);
 	}
@@ -629,7 +663,11 @@ teco_state_control_lines2glyphs(teco_machine_main_t *ctx, GError **error)
 		if (!teco_expressions_args()) {
 			pos = teco_interface_ssm(SCI_GETCURRENTPOS, 0, 0);
 		} else {
-			pos = teco_interface_glyphs2bytes(teco_expressions_pop_num(0));
+			/*
+			 * FIXME: Could be optimized by directly calling
+			 * SCI_LINEFROMINDEXPOSITION.
+			 */
+			pos = teco_interface_glyphs2bytes_absdot(teco_expressions_pop_num(0));
 			if (pos < 0) {
 				teco_error_range_set(error, "^Q");
 				return;
@@ -652,6 +690,10 @@ teco_state_control_lines2glyphs(teco_machine_main_t *ctx, GError **error)
 		}
 
 		sptr_t line_pos = teco_interface_ssm(SCI_POSITIONFROMLINE, line, 0);
-		teco_expressions_push(teco_interface_bytes2glyphs(line_pos) - teco_interface_bytes2glyphs(pos));
+		/*
+		 * FIXME: Could be optimized with SCI_INDEXPOSITIONFROMLINE.
+		 */
+		teco_expressions_push(teco_interface_bytes2glyphs_absdot(line_pos) -
+		                      teco_current_doc_get_dot());
 	}
 }