===================================================================
RCS file: /cvs/docbook2mdoc/parse.c,v
retrieving revision 1.42
retrieving revision 1.47
diff -u -p -r1.42 -r1.47
--- docbook2mdoc/parse.c	2019/04/14 14:00:16	1.42
+++ docbook2mdoc/parse.c	2019/04/20 04:15:06	1.47
@@ -1,4 +1,4 @@
-/* $Id: parse.c,v 1.42 2019/04/14 14:00:16 schwarze Exp $ */
+/* $Id: parse.c,v 1.47 2019/04/20 04:15:06 schwarze Exp $ */
 /*
  * Copyright (c) 2014 Kristaps Dzonsons <kristaps@bsd.lv>
  * Copyright (c) 2019 Ingo Schwarze <schwarze@openbsd.org>
@@ -56,6 +56,7 @@ struct	parse {
 	int		 nline;  /* Line number of next token. */
 	int		 ncol;   /* Column number of next token. */
 	int		 del;    /* Levels of nested nodes being deleted. */
+	int		 nofill; /* Levels of open no-fill displays. */
 	int		 flags;
 #define	PFLAG_WARN	 (1 << 0)  /* Print warning messages. */
 #define	PFLAG_SPC	 (1 << 1)  /* Whitespace before the next element. */
@@ -70,25 +71,32 @@ struct	alias {
 
 static	const struct alias aliases[] = {
 	{ "acronym",		NODE_IGNORE },
+	{ "affiliation",	NODE_IGNORE },
 	{ "anchor",		NODE_DELETE },
 	{ "application",	NODE_COMMAND },
 	{ "article",		NODE_SECTION },
 	{ "articleinfo",	NODE_BOOKINFO },
 	{ "book",		NODE_SECTION },
 	{ "chapter",		NODE_SECTION },
+	{ "caption",		NODE_IGNORE },
 	{ "code",		NODE_LITERAL },
 	{ "computeroutput",	NODE_LITERAL },
 	{ "!doctype",		NODE_DOCTYPE },
+	{ "figure",		NODE_IGNORE },
 	{ "firstname",		NODE_PERSONNAME },
 	{ "glossary",		NODE_VARIABLELIST },
 	{ "glossdef",		NODE_IGNORE },
 	{ "glossdiv",		NODE_IGNORE },
 	{ "glossentry",		NODE_VARLISTENTRY },
 	{ "glosslist",		NODE_VARIABLELIST },
+	{ "holder",		NODE_IGNORE },
+	{ "imageobject",	NODE_IGNORE },
 	{ "indexterm",		NODE_DELETE },
 	{ "informaltable",	NODE_TABLE },
 	{ "keycap",		NODE_KEYSYM },
 	{ "keycode",		NODE_IGNORE },
+	{ "mediaobject",	NODE_BLOCKQUOTE },
+	{ "orgname",		NODE_IGNORE },
 	{ "othercredit",	NODE_AUTHOR },
 	{ "othername",		NODE_PERSONNAME },
 	{ "part",		NODE_SECTION },
@@ -99,10 +107,13 @@ static	const struct alias aliases[] = {
 	{ "refsect2",		NODE_SECTION },
 	{ "refsect3",		NODE_SECTION },
 	{ "refsection",		NODE_SECTION },
+	{ "releaseinfo",	NODE_IGNORE },
 	{ "returnvalue",	NODE_IGNORE },
 	{ "secondary",		NODE_DELETE },
 	{ "sect1",		NODE_SECTION },
 	{ "sect2",		NODE_SECTION },
+	{ "sect3",		NODE_SECTION },
+	{ "sect4",		NODE_SECTION },
 	{ "sgmltag",		NODE_MARKUP },
 	{ "simpara",		NODE_PARA },
 	{ "structfield",	NODE_PARAMETER },
@@ -112,6 +123,7 @@ static	const struct alias aliases[] = {
 	{ "trademark",		NODE_IGNORE },
 	{ "ulink",		NODE_LINK },
 	{ "userinput",		NODE_LITERAL },
+	{ "year",		NODE_IGNORE },
 	{ NULL,			NODE_IGNORE }
 };
 
@@ -275,6 +287,7 @@ xml_text(struct parse *p, const char *word, int sz)
 		case CLASS_VOID:
 		case CLASS_TEXT:
 		case CLASS_BLOCK:
+		case CLASS_NOFILL:
 			np = NULL;
 			break;
 		case CLASS_TRANS:
@@ -516,6 +529,9 @@ xml_elem_start(struct parse *p, const char *name)
 	case CLASS_ENCL:
 		n->spc = (p->flags & PFLAG_SPC) != 0;
 		break;
+	case CLASS_NOFILL:
+		p->nofill++;
+		/* FALLTHROUGH */
 	default:
 		n->spc = 1;
 		break;
@@ -537,7 +553,7 @@ xml_attrkey(struct parse *p, const char *name)
 	const char	*value;
 	enum attrkey	 key;
 
-	if (p->del > 0 || p->ncur == NODE_IGNORE || *name == '\0')
+	if (p->del > 0 || p->ncur >= NODE_UNKNOWN || *name == '\0')
 		return;
 
 	if ((p->ncur == NODE_DOCTYPE || p->ncur == NODE_ENTITY) &&
@@ -574,7 +590,7 @@ xml_attrval(struct parse *p, const char *name)
 {
 	struct pattr	*a;
 
-	if (p->del > 0 || p->ncur == NODE_IGNORE ||
+	if (p->del > 0 || p->ncur >= NODE_UNKNOWN ||
 	    (p->flags & PFLAG_ATTR) == 0)
 		return;
 	if ((a = TAILQ_LAST(&p->cur->attrq, pattrq)) == NULL)
@@ -640,6 +656,8 @@ xml_elem_end(struct parse *p, const char *name)
 			warn_msg(p, "element not open: </%s>", name);
 			break;
 		}
+		if (pnode_class(node) == CLASS_NOFILL)
+			p->nofill--;
 
 		/*
 		 * Refrain from actually closing the document element.
@@ -747,11 +765,12 @@ parse_string(struct parse *p, char *b, size_t rlen,
     enum pstate *pstate, int refill)
 {
 	char		*cp;
+	size_t		 pws;	/* Parse offset including whitespace. */
 	size_t		 poff;  /* Parse offset in b[]. */
 	size_t		 pend;  /* Offset of the end of the current word. */
 	int		 elem_end;
 
-	pend = 0;
+	pend = pws = 0;
 	for (;;) {
 
 		/* Proceed to the next token, skipping whitespace. */
@@ -764,6 +783,8 @@ parse_string(struct parse *p, char *b, size_t rlen,
 			break;
 		if (isspace((unsigned char)b[pend])) {
 			p->flags |= PFLAG_SPC;
+			if (b[pend] == '\n')
+				pws = pend + 1;
 			increment(p, b, &pend, refill);
 			continue;
 		}
@@ -925,10 +946,13 @@ parse_string(struct parse *p, char *b, size_t rlen,
 			advance(p, b, rlen, &pend,
 			    p->ncur == NODE_DOCTYPE ? "<&]\n" : "<&\n",
 			    refill);
+			if (p->nofill)
+				poff = pws;
 			xml_text(p, b + poff, pend - poff);
 			if (b[pend] == '\n')
 				pnode_closetext(p, 0);
 		}
+		pws = pend;
 	}
 	return poff;
 }