In HTML5 anchors should not be closed when encountering divs, for example, but should be closed when encountering buttons, for example. Many sites that use HTML5-style anchors end up having links displayed with zero-length link texts. The proposed patch correct this behaviour by detecting whether the document is HTML5, then suppressing the close-anchor action in CLOSE_A if it's an HTML5 document. A new macro handles the HTML5-specific cases where anchors are not already always closed.

This also fixes a bug in the tokenizing FSM in etc.c that prevented the !doctype element from being recognized; the fix is necessary because HTML5 detection depends on checking the !doctype element.

10 files changed, 72 insertions, 6 deletions

diff --git a/etc.c b/etc.c
index 37c4f15..16d1295 100644
--- a/etc.c
+++ b/etc.c
@@ -727,6 +727,11 @@ next_status(char c, int *status)
 	case '>':
 	    *status = R_ST_NORMAL;
 	    break;
+	case 'D':
+	case 'd':
+	    /* could be a !doctype */
+	    *status = R_ST_TAG;
+	    break;
 	default:
 	    *status = R_ST_IRRTAG;
 	}
diff --git a/file.c b/file.c
index c0fc044..3d772db 100644
--- a/file.c
+++ b/file.c
@@ -1,4 +1,5 @@
 /* $Id: file.c,v 1.266 2012/05/22 09:45:56 inu Exp $ */
+/* vi: set sw=4 ts=8 ai sm noet : */
 #include "fm.h"
 #include <sys/types.h>
 #include "myctype.h"
@@ -4322,9 +4323,18 @@ process_idattr(struct readbuffer *obuf, int cmd, struct parsed_tag *tag)
       obuf->flag &= ~RB_P;\
     }
 
-#define CLOSE_A \
-    CLOSE_P; \
-    close_anchor(h_env, obuf);
+#define HTML5_CLOSE_A do { \
+	if (obuf->flag & RB_HTML5) { \
+	    close_anchor(h_env, obuf); \
+	} \
+    } while (0)
+
+#define CLOSE_A do { \
+	CLOSE_P; \
+	if (!(obuf->flag & RB_HTML5)) { \
+	    close_anchor(h_env, obuf); \
+	} \
+    } while (0)
 
 #define CLOSE_DT \
     if (obuf->flag & RB_IN_DT) { \
@@ -4930,6 +4940,8 @@ HTMLtagproc1(struct parsed_tag *tag, struct html_feed_environ *h_env)
 	close_anchor(h_env, obuf);
 	return 1;
     case HTML_IMG:
+	if (parsedtag_exists(tag, ATTR_USEMAP))
+	    HTML5_CLOSE_A;
 	tmp = process_img(tag, h_env->limit);
 	HTMLlineproc1(tmp->ptr, h_env);
 	return 1;
@@ -5125,6 +5137,7 @@ HTMLtagproc1(struct parsed_tag *tag, struct html_feed_environ *h_env)
            HTMLlineproc1(tmp->ptr, h_env);
        return 1;
     case HTML_BUTTON:
+       HTML5_CLOSE_A;
        tmp = process_button(tag);
        if (tmp)
            HTMLlineproc1(tmp->ptr, h_env);
@@ -5180,6 +5193,11 @@ HTMLtagproc1(struct parsed_tag *tag, struct html_feed_environ *h_env)
 			     NULL);
 	HTMLlineproc1(tmp->ptr, h_env);
 	return 1;
+    case HTML_DOCTYPE:
+	if (!parsedtag_exists(tag, ATTR_PUBLIC)) {
+	    obuf->flag |= RB_HTML5;
+	}
+	return 1;
     case HTML_META:
 	p = q = r = NULL;
 	parsedtag_get_value(tag, ATTR_HTTP_EQUIV, &p);
@@ -5378,6 +5396,7 @@ HTMLtagproc1(struct parsed_tag *tag, struct html_feed_environ *h_env)
 	}
 	return 1;
     case HTML_EMBED:
+	HTML5_CLOSE_A;
 	if (view_unseenobject) {
 	    if (parsedtag_get_value(tag, ATTR_SRC, &p)) {
 		Str s;
diff --git a/fm.h b/fm.h
index 4a17ecc..330fd09 100644
--- a/fm.h
+++ b/fm.h
@@ -661,6 +661,7 @@ struct readbuffer {
 #endif				/* FORMAT_NICE */
 #define RB_DEL		0x100000
 #define RB_S		0x200000
+#define RB_HTML5	0x400000
 
 #define RB_GET_ALIGN(obuf) ((obuf)->flag&RB_ALIGN)
 #define RB_SET_ALIGN(obuf,align) {(obuf)->flag &= ~RB_ALIGN; (obuf)->flag |= (align); }
@@ -673,7 +674,7 @@ struct readbuffer {
    RB_SET_ALIGN(obuf,(obuf)->flag_stack[--(obuf)->flag_sp]); \
 }
 
-/* status flags */
+/* state of token scanning finite state machine */
 #define R_ST_NORMAL 0		/* normal */
 #define R_ST_TAG0   1		/* within tag, just after < */
 #define R_ST_TAG    2		/* within tag */
diff --git a/html.c b/html.c
index 90cf66e..35a5787 100644
--- a/html.c
+++ b/html.c
@@ -37,6 +37,8 @@ unsigned char ALST_TABLE[] =
     ATTR_CELLPADDING, ATTR_VSPACE, ATTR_CORE
 };
 #define MAXA_TABLE	MAXA_CORE + 6
+unsigned char ALST_DOCTYPE[] = { ATTR_PUBLIC }; /* only (html and) public should be checked */
+#define MAXA_DOCTYPE	1
 unsigned char ALST_META[] = { ATTR_HTTP_EQUIV, ATTR_CONTENT, ATTR_CHARSET, ATTR_CORE };
 #define MAXA_META	MAXA_CORE + 3
 unsigned char ALST_FRAME[] = { ATTR_SRC, ATTR_NAME, ATTR_CORE };
@@ -221,7 +223,7 @@ TagInfo TagMAP[MAX_HTMLTAG] = {
     {"/option", NULL, 0, TFLG_END},	/*  94 HTML_N_OPTION   */
     {"head", ALST_NOP, MAXA_NOP, 0},	/*  95 HTML_HEAD       */
     {"/head", NULL, 0, TFLG_END},	/*  96 HTML_N_HEAD     */
-    {"doctype", ALST_NOP, MAXA_NOP, 0},	/*  97 HTML_DOCTYPE    */
+    {"doctype", ALST_DOCTYPE, MAXA_DOCTYPE, 0},	/*  97 HTML_DOCTYPE    */
     {"noframes", ALST_NOFRAMES, MAXA_NOFRAMES, 0},	/*  98 HTML_NOFRAMES   */
     {"/noframes", NULL, 0, TFLG_END},	/*  99 HTML_N_NOFRAMES */
 
@@ -367,7 +369,7 @@ TagAttrInfo AttrMAP[MAX_TAGATTR] = {
     {"rev", VTYPE_STR, 0},	/* 48 ATTR_REV            */
     {"title", VTYPE_STR, 0},	/* 49 ATTR_TITLE          */
     {"accesskey", VTYPE_STR, 0},	/* 50 ATTR_ACCESSKEY          */
-    {NULL, VTYPE_NONE, 0},	/* 51 Undefined           */
+    {"public", VTYPE_NONE, 0},	/* 51 ATTR_PUBLIC         */
     {NULL, VTYPE_NONE, 0},	/* 52 Undefined           */
     {NULL, VTYPE_NONE, 0},	/* 53 Undefined           */
     {NULL, VTYPE_NONE, 0},	/* 54 Undefined           */
diff --git a/html.h b/html.h
index 6da6f5f..7b173c3 100644
--- a/html.h
+++ b/html.h
@@ -318,6 +318,7 @@ typedef struct {
 #define ATTR_REV		48
 #define ATTR_TITLE		49
 #define ATTR_ACCESSKEY		50
+#define ATTR_PUBLIC		51
 
 /* Internal attribute */
 #define ATTR_XOFFSET		60
diff --git a/tests/a1.expected b/tests/a1.expected
new file mode 100644
index 0000000..5812232
--- /dev/null
+++ b/tests/a1.expected
@@ -0,0 +1,2 @@
+ 
+test
diff --git a/tests/a1.html b/tests/a1.html
new file mode 100644
index 0000000..7e89006
--- /dev/null
+++ b/tests/a1.html
@@ -0,0 +1 @@
+<a href="example"><div>test</div></a>
diff --git a/tests/a2.expected b/tests/a2.expected
new file mode 100644
index 0000000..9daeafb
--- /dev/null
+++ b/tests/a2.expected
@@ -0,0 +1 @@
+test
diff --git a/tests/a2.html b/tests/a2.html
new file mode 100644
index 0000000..197b006
--- /dev/null
+++ b/tests/a2.html
@@ -0,0 +1,3 @@
+<!doctype html>
+<meta charset=utf-8>
+<a href="example"><div>test</div></a>
diff --git a/tests/run_tests b/tests/run_tests
new file mode 100644
index 0000000..0ec3080
--- /dev/null
+++ b/tests/run_tests
@@ -0,0 +1,31 @@
+total=0
+pass=0
+fail=0
+w3m="../w3m
+-config
+/dev/null
+-o
+ignore_null_img_alt=false"
+for i in *.html; do
+	cmd="$w3m
+-I
+utf-8
+-O
+utf-8
+-T
+text/html"
+	opts="`basename "$i" .html`.opts"
+	test -f "$opts" && cmd="$cmd
+`grep -v '^#' $opts`"
+	if (set -x;IFS='
+';$cmd) < "$i" | diff -u - "`basename "$i" .html`.expected"; then
+		pass="`expr 1 + "$pass"`"
+	else
+		fail="`expr 1 + "$fail"`"
+	fi
+	total="`expr 1 + "$total"`"
+done
+echo "TOTAL: $total test(s)"
+echo "PASS : $pass"
+echo "FAIL : $fail"
+test 0 -eq "$fail"