scdoc2mdoc

commit 9b01ea2093def60be232a1b57a7d4b3d10b8c8b5
parent 71a5592354dd6750b75f7cd8291d1467cf9146fa
Author: Drew DeVault <sir@cmpwn.com>
Date:   Sun, 27 Jan 2019 10:48:28 -0500

Ignore underscores in the middle_of_a_word

Diffstat:
M
include/util.h
 | 
1
+
M
scdoc.5.scd
 | 
2
+-
M
src/main.c
 | 
5
+++++
M
src/util.c
 | 
22
++++++++++++----------

4 files changed, 19 insertions(+), 11 deletions(-)
diff --git a/include/util.h b/include/util.h
@@ -12,6 +12,7 @@ struct parser {
 	uint32_t flags;
 	const char *str;
 	int fmt_line, fmt_col;
+	uint32_t last[2];
 };
 
 enum formatting {
diff --git a/scdoc.5.scd b/scdoc.5.scd
@@ -45,7 +45,7 @@ like this.
 ## FORMATTING
 
 Text can be made *bold* or _underlined_ with asterisks and underscores: \*bold\*
-or \_underlined\_.
+or \_underlined\_. Underscores in the_middle_of_words will be disregarded.
 
 ## INDENTATION
 
diff --git a/src/main.c b/src/main.c
@@ -163,6 +163,11 @@ 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;
diff --git a/src/util.c b/src/util.c
@@ -14,24 +14,26 @@ void parser_fatal(struct parser *parser, const char *err) {
 }
 
 uint32_t parser_getch(struct parser *parser) {
+	uint32_t ch;
 	if (parser->qhead) {
-		return parser->queue[--parser->qhead];
-	}
-	if (parser->str) {
+		ch = parser->queue[--parser->qhead];
+	} else 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 {
-		++parser->col;
+		ch = utf8_fgetch(parser->input);
+		if (ch == '\n') {
+			parser->col = 0;
+			++parser->line;
+		} else {
+			++parser->col;
+		}
 	}
+	parser->last[0] = parser->last[1];
+	parser->last[1] = ch;
 	return ch;
 }