scdoc2mdoc

commit 7f298175918c16b3bc467b47bde976def0650d68
parent bcd942a02ff3911fe5babce8d6b9d814254ed3f1
Author: Drew DeVault <sir@cmpwn.com>
Date:   Sun, 27 Jan 2019 11:08:03 -0500

Use a more robust approach for in-word-underscores

Diffstat:
M
include/util.h
 | 
1
-
M
src/main.c
 | 
16
+++++++++-------
M
src/util.c
 | 
22
++++++++++------------
M
test/inline-formatting
 | 
16
++++++++++++++++

4 files changed, 35 insertions(+), 20 deletions(-)
diff --git a/include/util.h b/include/util.h
@@ -12,7 +12,6 @@ struct parser {
 	uint32_t flags;
 	const char *str;
 	int fmt_line, fmt_col;
-	uint32_t last[2];
 };
 
 enum formatting {
diff --git a/src/main.c b/src/main.c
@@ -163,11 +163,6 @@ static void parse_format(struct parser *p, enum formatting fmt) {
 		}
 		fprintf(p->output, "\\fR");
 	} else {
-		if (fmt == FORMAT_UNDERLINE && !isspace(p->last[0])) {
-			// Ignore underscores in the middle of words
-			utf8_fputch(p->output, '_');
-			return;
-		}
 		fprintf(p->output, "\\f%c", formats[fmt]);
 		p->fmt_line = p->line;
 		p->fmt_col = p->col;
@@ -199,7 +194,7 @@ static void parse_linebreak(struct parser *p) {
 }
 
 static void parse_text(struct parser *p) {
-	uint32_t ch;
+	uint32_t ch, last = ' ';
 	int i = 0;
 	while ((ch = parser_getch(p)) != UTF8_INVALID) {
 		switch (ch) {
@@ -217,7 +212,13 @@ static void parse_text(struct parser *p) {
 			parse_format(p, FORMAT_BOLD);
 			break;
 		case '_':
-			parse_format(p, FORMAT_UNDERLINE);
+			if ((p->flags & FORMAT_UNDERLINE)) {
+				parse_format(p, FORMAT_UNDERLINE);
+			} else if (!p->flags && isspace(last)) {
+				parse_format(p, FORMAT_UNDERLINE);
+			} else {
+				utf8_fputch(p->output, ch);
+			}
 			break;
 		case '+':
 			parse_linebreak(p);
@@ -233,6 +234,7 @@ static void parse_text(struct parser *p) {
 			}
 			/* fallthrough */
 		default:
+			last = ch;
 			utf8_fputch(p->output, ch);
 			break;
 		}
diff --git a/src/util.c b/src/util.c
@@ -14,26 +14,24 @@ void parser_fatal(struct parser *parser, const char *err) {
 }
 
 uint32_t parser_getch(struct parser *parser) {
-	uint32_t ch = 0;
 	if (parser->qhead) {
-		ch = parser->queue[--parser->qhead];
-	} else if (parser->str) {
+		return parser->queue[--parser->qhead];
+	}
+	if (parser->str) {
 		uint32_t ch = utf8_decode(&parser->str);
 		if (!ch || ch == UTF8_INVALID) {
 			parser->str = NULL;
 			return UTF8_INVALID;
 		}
+		return ch;
+	}
+	uint32_t ch = utf8_fgetch(parser->input);
+	if (ch == '\n') {
+		parser->col = 0;
+		++parser->line;
 	} else {
-		ch = utf8_fgetch(parser->input);
-		if (ch == '\n') {
-			parser->col = 0;
-			++parser->line;
-		} else {
-			++parser->col;
-		}
+		++parser->col;
 	}
-	parser->last[0] = parser->last[1];
-	parser->last[1] = ch;
 	return ch;
 }
 
diff --git a/test/inline-formatting b/test/inline-formatting
@@ -9,6 +9,22 @@ _hello *world*_
 EOF
 end 1
 
+begin "Ignores underscores in words"
+scdoc <<EOF | grep -v 'fR' >/dev/null
+test(8)
+
+hello_world
+EOF
+end 0
+
+begin "Allows underscores in bolded words"
+scdoc <<EOF | grep '^\\fBhello_world\\fR' >/dev/null
+test(8)
+
+*hello_world*
+EOF
+end 0
+
 begin "Emits bold text"
 scdoc <<EOF | grep '^hello \\fBworld\\fR' >/dev/null
 test(8)