docbook2mdoc: Parse internal subset declarations in <!DOCTYPE> tags as

[schwarze@mandoc.bsd.lv]

Log Message:
-----------
Parse internal subset declarations in <!DOCTYPE> tags as XML;
in particular, they can contain XML comments.
Similar to parts of parse.c rev. 1.23, but simpler.

Modified Files:
--------------
    docbook2mdoc:
        statistics.c

Revision Data
-------------
Index: statistics.c
===================================================================
RCS file: /home/cvs/mdocml/docbook2mdoc/statistics.c,v
retrieving revision 1.25
retrieving revision 1.26
diff -Lstatistics.c -Lstatistics.c -u -p -r1.25 -r1.26
--- statistics.c
+++ statistics.c
@@ -211,10 +211,10 @@ parse_file(int fd, char *fname)
 	size_t		 rlen;  /* Number of bytes in b[]. */
 	size_t		 poff;  /* Parse offset in b[]. */
 	size_t		 pend;  /* Offset of the end of the current word. */
-	int		 in_tag, in_arg, in_quotes, elem_end;
+	int		 in_tag, in_arg, in_quotes, in_doctype, elem_end;
 
 	rlen = 0;
-	in_tag = in_arg = in_quotes = 0;
+	in_tag = in_arg = in_quotes = in_doctype = 0;
 	while ((rsz = read(fd, b + rlen, sizeof(b) - rlen - 1)) >= 0) {
 		if ((rlen += rsz) == 0)
 			break;
@@ -251,6 +251,11 @@ parse_file(int fd, char *fname)
 				if (elem_end)
 					stack_pop(NULL);
 			} else if (in_tag) {
+				if (in_doctype && b[pend] == '[') {
+					in_tag = in_doctype = 0;
+					pend++;
+					continue;
+				}
 				if (advance(b, rlen, &pend, " =>") && rsz > 0)
 					break;
 				elem_end = 0;
@@ -301,6 +306,9 @@ parse_file(int fd, char *fname)
 				if (b[++poff] == '/') {
 					elem_end = 1;
 					poff++;
+				} else if (strcasecmp(b + poff,
+				    "!DOCTYPE") == 0) {
+					in_doctype = 1;
 				} else if (b[poff] != '!' && b[poff] != '?') {
 					table_add(stacki > 0 ?
 					    stack[stacki - 1] : "ROOT",
--
 To unsubscribe send an email to source+unsubscribe@mandoc.bsd.lv