libgrapheme

commit f517655a98a155694cf57c180531724baa081c26
parent 07ba2622e073850bbdd6acd8dff88b391cc5ad5c
Author: Laslo Hunhold <dev@frign.de>
Date:   Mon, 21 Nov 2022 09:46:38 +0100

Implement bidirectional rule L1.4

For this, we first make use of our paragraph level slot in each
data point and store it for each. This way, even if the data
buffer is arbitrarily split up, we always know what the current
paragraph level is.

Secondly, we add the rule L1.4 itself, which is very similar to
the existing implementation of rules L1.1-L1.3.

Diffstat:
M
src/bidirectional.c
 | 
45
+++++++++++++++++++++++++++++++++++++++++----

1 file changed, 41 insertions(+), 4 deletions(-)
diff --git a/src/bidirectional.c b/src/bidirectional.c
@@ -679,6 +679,10 @@ preprocess_paragraph(enum grapheme_bidirectional_override override,
 
 	for (bufoff = 0; bufoff < buflen; bufoff++) {
 		prop = (uint_least8_t)get_state(STATE_PROP, buf[bufoff]);
+
+		/* set paragraph level, which we need for line-level-processing */
+		set_state(STATE_PARAGRAPH_LEVEL, paragraph_level,
+		          &(buf[bufoff]));
 again:
 		if (prop == BIDI_PROP_RLE) {
 			/* X2 */
@@ -997,7 +1001,7 @@ again:
 	}
 	if (runsince != SIZE_MAX) {
 		/*
-		 * this is the end of the line and we
+		 * this is the end of the paragraph and we
 		 * are in a run
 		 */
 		for (i = runsince; i < buflen; i++) {
@@ -1133,10 +1137,43 @@ void
 grapheme_bidirectional_get_line_embedding_levels(
 	const int_least32_t *linedata, size_t linelen, int_least8_t *linelevel)
 {
-	size_t i;
+	enum bidi_property prop;
+	size_t i, runsince;
 
-	/* write the levels into the level-array */
+	/* rule L1.4 */
+	runsince = SIZE_MAX;
 	for (i = 0; i < linelen; i++) {
-		linelevel[i] = get_state(STATE_LEVEL, linedata[i]);
+		prop = (uint_least8_t)get_state(STATE_RAWPROP, linedata[i]);
+
+		/* write level into level array */
+		if ((linelevel[i] = (int_least8_t)get_state(
+		                    STATE_LEVEL, linedata[i])) == -1) {
+			/* ignored character */
+			continue;
+		}
+
+		if (prop == BIDI_PROP_WS || prop == BIDI_PROP_FSI ||
+		    prop == BIDI_PROP_LRI || prop == BIDI_PROP_RLI ||
+		    prop == BIDI_PROP_PDI) {
+			if (runsince == SIZE_MAX) {
+				/* a new run has begun */
+				runsince = i;
+			}
+		} else {
+			/* sequence ended */
+			runsince = SIZE_MAX;
+		}
+	}
+	if (runsince != SIZE_MAX) {
+		/*
+		 * we hit the end of the line but were in a run;
+		 * reset the line levels to the paragraph level
+		 */
+		for (i = runsince; i < linelen; i++) {
+			if (linelevel[i] != -1) {
+				linelevel[i] = get_state(
+					STATE_PARAGRAPH_LEVEL, linedata[i]);
+			}
+		}
 	}
 }