From 99ac4e7f5a389898673348449e4a522a86d07f1c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?John=20Ankarstr=C3=B6m?= <john@ankarstrom.se>
Date: Sat, 30 Jan 2021 09:24:26 +0000
Subject: Keep track of open inline format, auto-close at block break

---
 aux/emparse | 91 ++++++++++++++++++++++++++++++++++++-------------------------
 test.em     |  6 ++++
 test.html   |  9 ++++++
 3 files changed, 69 insertions(+), 37 deletions(-)

diff --git a/aux/emparse b/aux/emparse
index edafcae..929c1c7 100755
--- a/aux/emparse
+++ b/aux/emparse
@@ -60,37 +60,50 @@ function format(line) {
 		else if (match(w[i], "^\\[[0-9a-z]+\\]\\)[.,:;?!]$"))
 			w[i] = ref(substr(w[i], 2, RLENGTH-4)) substr(w[i], RLENGTH-1)
 
-		if (match(w[i], "^`"))
-			w[i] = "<tt>" substr(w[i], 2)
-		else if (match(w[i], "^\\*"))
-			w[i] = "<i>" substr(w[i], 2)
-		else if (match(w[i], "^_"))
-			w[i] = "<b>" substr(w[i], 2)
-		else if (match(w[i], "^\\(`"))
-			w[i] = "(<tt>" substr(w[i], 3)
-		else if (match(w[i], "^\\(\\*"))
-			w[i] = "(<i>" substr(w[i], 3)
-		else if (match(w[i], "^\\(_"))
-			w[i] = "(<b>" substr(w[i], 3)
-
-		if (match(w[i], "`$"))
-			w[i] = substr(w[i], 1, RSTART-1) "</tt>"
-		else if (match(w[i], "\\*$"))
-			w[i] = substr(w[i], 1, RSTART-1) "</i>"
-		else if (match(w[i], "_$"))
-			w[i] = substr(w[i], 1, RSTART-1) "</b>"
-		else if (match(w[i], "`[.,:;?!)]$"))
-			w[i] = substr(w[i], 1, RSTART-1) "</tt>" substr(w[i], RSTART+1)
-		else if (match(w[i], "\\*[.,:;?!)]$"))
-			w[i] = substr(w[i], 1, RSTART-1) "</i>" substr(w[i], RSTART+1)
-		else if (match(w[i], "_[.,:;?!)]$"))
-			w[i] = substr(w[i], 1, RSTART-1) "</b>" substr(w[i], RSTART+1)
-		else if (match(w[i], "`\\)[.,:;?!]$"))
-			w[i] = substr(w[i], 1, RSTART-1) "</tt>" substr(w[i], RSTART+1)
-		else if (match(w[i], "\\*\\)[.,:;?!]$"))
-			w[i] = substr(w[i], 1, RSTART-1) "</i>" substr(w[i], RSTART+1)
-		else if (match(w[i], "_\\)[.,:;?!]$"))
-			w[i] = substr(w[i], 1, RSTART-1) "</b>" substr(w[i], RSTART+1)
+		if (!openformat) {
+			if (match(w[i], "^`")) {
+				w[i] = "<tt>" substr(w[i], 2)
+				openformat = "tt"
+			} else if (match(w[i], "^\\*")) {
+				w[i] = "<i>" substr(w[i], 2)
+				openformat = "i"
+			} else if (match(w[i], "^_")) {
+				w[i] = "<b>" substr(w[i], 2)
+				openformat = "b"
+			} else if (match(w[i], "^\\(`")) {
+				w[i] = "(<tt>" substr(w[i], 3)
+				openformat = "tt"
+			} else if (match(w[i], "^\\(\\*")) {
+				w[i] = "(<i>" substr(w[i], 3)
+				openformat = "i"
+			} else if (match(w[i], "^\\(_")) {
+				w[i] = "(<b>" substr(w[i], 3)
+				openformat = "b"
+			}
+		}
+
+		if (openformat) {
+			orig = w[i]
+			if (match(w[i], "`$"))
+				w[i] = substr(w[i], 1, RSTART-1) "</tt>"
+			else if (match(w[i], "\\*$"))
+				w[i] = substr(w[i], 1, RSTART-1) "</i>"
+			else if (match(w[i], "_$"))
+				w[i] = substr(w[i], 1, RSTART-1) "</b>"
+			else if (match(w[i], "`[.,:;?!)]$"))
+				w[i] = substr(w[i], 1, RSTART-1) "</tt>" substr(w[i], RSTART+1)
+			else if (match(w[i], "\\*[.,:;?!)]$"))
+				w[i] = substr(w[i], 1, RSTART-1) "</i>" substr(w[i], RSTART+1)
+			else if (match(w[i], "_[.,:;?!)]$"))
+				w[i] = substr(w[i], 1, RSTART-1) "</b>" substr(w[i], RSTART+1)
+			else if (match(w[i], "`\\)[.,:;?!]$"))
+				w[i] = substr(w[i], 1, RSTART-1) "</tt>" substr(w[i], RSTART+1)
+			else if (match(w[i], "\\*\\)[.,:;?!]$"))
+				w[i] = substr(w[i], 1, RSTART-1) "</i>" substr(w[i], RSTART+1)
+			else if (match(w[i], "_\\)[.,:;?!]$"))
+				w[i] = substr(w[i], 1, RSTART-1) "</b>" substr(w[i], RSTART+1)
+			if (w[i] != orig) openformat = ""
+		}
 
 		printf "%s", w[i]
 		if (i < n) printf " "
@@ -98,35 +111,39 @@ function format(line) {
 	printf "\n"
 }
 
-function beginblock(name) {
+function opentag(name) {
 	if (name == "nl") printf "<ol class=\"reflist\" style=\"font-size: small;\">\n"
 	else printf "<%s>\n", name
 }
 
-function endblock(name) {
+function closetag(name) {
 	if (name == "nl") printf "</ol>\n"
 	else printf "</%s>\n", name
 }
 
 function newblock(name) {
-	beginblock(name)
+	opentag(name)
 	openblock = name
-	openitem = 0
 	expectblock = 0
 }
 
 function breakblock() {
+	if (openformat) {
+		printf "</%s>", openformat
+		printf "(%s:%d) warning: open <%s> closed at block break\n", ARGV[1], NR, openformat > "/dev/stderr"
+	}
 	if (openitem) {
 		printf "</li>\n"
 		while (itemlevel-- > 1)
-			endblock(openblock)
+			closetag(openblock)
 		itemlevel = 1
 	}
 	if (opendef) printf "</dd>\n"
-	if (openblock) endblock(openblock)
+	if (openblock) closetag(openblock)
 	openitem = 0
 	opendef = 0
 	openblock = 0
+	openformat = ""
 	expectblock = 1
 }
 
diff --git a/test.em b/test.em
index 79569a6..830e4bd 100644
--- a/test.em
+++ b/test.em
@@ -2,6 +2,12 @@
 
 This is an example of *em* [a].
 
+Here is some `teletype text`, some _bold text_.
+
+Here is an asterisk at the end of a word*.
+
+Here is an *unfinished inline formatting tag.
+
  - Here is an
 unordered list item.
  - Here is another.
diff --git a/test.html b/test.html
index af80b0b..b5bafb5 100644
--- a/test.html
+++ b/test.html
@@ -2,6 +2,15 @@
 <p>
 This is an example of <i>em</i> (<a href="http://example.com">link</a>).
 </p>
+<p>
+Here is some <tt>teletype text</tt>, some <b>bold text</b>.
+</p>
+<p>
+Here is an asterisk at the end of a word*.
+</p>
+<p>
+Here is an <i>unfinished inline formatting tag.
+</i></p>
 <ul>
 <li>Here is an
 unordered list item.
-- 
cgit v1.2.3