From: Fletcher T. Penney <fletcher@fletcherpenney.net>
Date: Sun, 12 Feb 2017 19:56:03 +0000 (-0500)
Subject: FIXED: Improve reliability or link scanner
X-Git-Tag: 0.3.0a^2~19
X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=141976a36a911975a4511700645bc08cc6ab0e76;p=multimarkdown

FIXED: Improve reliability or link scanner
---

diff --git a/src/html.c b/src/html.c
index 6d08ee7..689ff8d 100644
--- a/src/html.c
+++ b/src/html.c
@@ -943,9 +943,7 @@ void mmd_export_token_html(DString * out, const char * source, token * t, size_t
 			print("</code>");
 			break;
 		case PAIR_ANGLE:
-			temp_token = t;
-
-			temp_char = url_accept(source, &temp_token, true);
+			temp_char = url_accept(source, t->start + 1, t->len - 2, NULL, true);
 
 			if (temp_char) {
 				if (scan_email(temp_char))
diff --git a/src/latex.c b/src/latex.c
index 457e913..1d04073 100644
--- a/src/latex.c
+++ b/src/latex.c
@@ -660,9 +660,7 @@ void mmd_export_token_latex(DString * out, const char * source, token * t, scrat
 			print_char(' ');
 			break;
 		case PAIR_ANGLE:
-			temp_token = t;
-
-			temp_char = url_accept(source, &temp_token, true);
+			temp_char = url_accept(source, t->start + 1, t->len - 2, NULL, true);
 
 			if (temp_char) {
 				if (scan_email(temp_char)) {
diff --git a/src/scanners.c b/src/scanners.c
index 15ddd0c..9284b6c 100644
--- a/src/scanners.c
+++ b/src/scanners.c
@@ -1,4 +1,4 @@
-/* Generated by re2c 0.14.3 on Sat Feb 11 13:20:22 2017 */
+/* Generated by re2c 0.14.3 on Sun Feb 12 13:52:25 2017 */
 /**
 
 	MultiMarkdown 6 -- Lightweight markup processor to produce HTML, LaTeX, and more.
@@ -9468,7 +9468,7 @@ yy730:
 }
 
 
-size_t scan_setext(const char * c) {
+size_t scan_title(const char * c) {
 	const char * marker = NULL;
 	const char * start = c;
 
@@ -9478,9 +9478,9 @@ size_t scan_setext(const char * c) {
 	yych = *c;
 	switch (yych) {
 	case '\n':	goto yy733;
-	case ' ':	goto yy734;
-	case '-':	goto yy736;
-	case '=':	goto yy735;
+	case '"':	goto yy734;
+	case '\'':	goto yy735;
+	case '(':	goto yy736;
 	default:	goto yy737;
 	}
 yy733:
@@ -9488,22 +9488,26 @@ yy733:
 yy734:
 	yych = *(marker = ++c);
 	switch (yych) {
-	case ' ':	goto yy749;
-	case '-':	goto yy750;
-	case '=':	goto yy751;
-	default:	goto yy733;
+	case 0x00:
+	case '\n':
+	case '\r':	goto yy733;
+	default:	goto yy746;
 	}
 yy735:
 	yych = *(marker = ++c);
 	switch (yych) {
-	case '=':	goto yy744;
-	default:	goto yy733;
+	case 0x00:
+	case '\n':
+	case '\r':	goto yy733;
+	default:	goto yy744;
 	}
 yy736:
 	yych = *(marker = ++c);
 	switch (yych) {
-	case '-':	goto yy738;
-	default:	goto yy733;
+	case 0x00:
+	case '\n':
+	case '\r':	goto yy733;
+	default:	goto yy739;
 	}
 yy737:
 	yych = *++c;
@@ -9511,72 +9515,155 @@ yy737:
 yy738:
 	++c;
 	yych = *c;
+yy739:
 	switch (yych) {
 	case 0x00:
-	case '\n':	goto yy741;
-	case '\r':	goto yy743;
-	case '-':	goto yy738;
-	default:	goto yy740;
+	case '\n':
+	case '\r':	goto yy740;
+	case ')':	goto yy741;
+	default:	goto yy738;
 	}
 yy740:
 	c = marker;
 	goto yy733;
 yy741:
 	++c;
-yy742:
 	{ return (size_t)( c - start ); }
 yy743:
+	++c;
+	yych = *c;
+yy744:
+	switch (yych) {
+	case 0x00:
+	case '\n':
+	case '\r':	goto yy740;
+	case '\'':	goto yy741;
+	default:	goto yy743;
+	}
+yy745:
+	++c;
+	yych = *c;
+yy746:
+	switch (yych) {
+	case 0x00:
+	case '\n':
+	case '\r':	goto yy740;
+	case '"':	goto yy741;
+	default:	goto yy745;
+	}
+}
+	
+}
+
+size_t scan_setext(const char * c) {
+	const char * marker = NULL;
+	const char * start = c;
+
+
+{
+	char yych;
+	yych = *c;
+	switch (yych) {
+	case '\n':	goto yy749;
+	case ' ':	goto yy750;
+	case '-':	goto yy752;
+	case '=':	goto yy751;
+	default:	goto yy753;
+	}
+yy749:
+	{ return 0; }
+yy750:
+	yych = *(marker = ++c);
+	switch (yych) {
+	case ' ':	goto yy765;
+	case '-':	goto yy766;
+	case '=':	goto yy767;
+	default:	goto yy749;
+	}
+yy751:
+	yych = *(marker = ++c);
+	switch (yych) {
+	case '=':	goto yy760;
+	default:	goto yy749;
+	}
+yy752:
+	yych = *(marker = ++c);
+	switch (yych) {
+	case '-':	goto yy754;
+	default:	goto yy749;
+	}
+yy753:
 	yych = *++c;
+	goto yy749;
+yy754:
+	++c;
+	yych = *c;
 	switch (yych) {
-	case '\n':	goto yy741;
-	default:	goto yy742;
+	case 0x00:
+	case '\n':	goto yy757;
+	case '\r':	goto yy759;
+	case '-':	goto yy754;
+	default:	goto yy756;
 	}
-yy744:
+yy756:
+	c = marker;
+	goto yy749;
+yy757:
+	++c;
+yy758:
+	{ return (size_t)( c - start ); }
+yy759:
+	yych = *++c;
+	switch (yych) {
+	case '\n':	goto yy757;
+	default:	goto yy758;
+	}
+yy760:
 	++c;
 	yych = *c;
 	switch (yych) {
 	case 0x00:
-	case '\n':	goto yy746;
-	case '\r':	goto yy748;
-	case '=':	goto yy744;
-	default:	goto yy740;
+	case '\n':	goto yy762;
+	case '\r':	goto yy764;
+	case '=':	goto yy760;
+	default:	goto yy756;
 	}
-yy746:
+yy762:
 	++c;
-yy747:
+yy763:
 	{ return (size_t)( c - start ); }
-yy748:
+yy764:
 	yych = *++c;
 	switch (yych) {
-	case '\n':	goto yy746;
-	default:	goto yy747;
+	case '\n':	goto yy762;
+	default:	goto yy763;
 	}
-yy749:
+yy765:
 	yych = *++c;
 	switch (yych) {
-	case ' ':	goto yy752;
-	case '-':	goto yy750;
-	case '=':	goto yy751;
-	default:	goto yy740;
+	case ' ':	goto yy768;
+	case '-':	goto yy766;
+	case '=':	goto yy767;
+	default:	goto yy756;
 	}
-yy750:
+yy766:
 	yych = *++c;
 	switch (yych) {
-	case '-':	goto yy738;
-	default:	goto yy740;
+	case '-':	goto yy754;
+	default:	goto yy756;
 	}
-yy751:
+yy767:
 	yych = *++c;
 	switch (yych) {
-	case '=':	goto yy744;
-	default:	goto yy740;
+	case '=':	goto yy760;
+	default:	goto yy756;
 	}
-yy752:
+yy768:
 	++c;
 	switch ((yych = *c)) {
-	case '-':	goto yy750;
-	case '=':	goto yy751;
-	default:	goto yy740;
+	case '-':	goto yy766;
+	case '=':	goto yy767;
+	default:	goto yy756;
 	}
 }
 	
diff --git a/src/scanners.h b/src/scanners.h
index d952f72..c42ecb0 100644
--- a/src/scanners.h
+++ b/src/scanners.h
@@ -90,6 +90,7 @@ size_t scan_ref_link_no_attributes(const char * c);
 size_t scan_setext(const char * c);
 size_t scan_spnl(const char * c);
 size_t scan_table_separator(const char * c);
+size_t scan_title(const char * c);
 size_t scan_url(const char * c);
 size_t scan_value(const char * c);
 
diff --git a/src/scanners.re b/src/scanners.re
index 9f1394c..8306b16 100644
--- a/src/scanners.re
+++ b/src/scanners.re
@@ -402,6 +402,16 @@ size_t scan_destination(const char * c) {
 }
 
 
+size_t scan_title(const char * c) {
+	const char * marker = NULL;
+	const char * start = c;
+
+/*!re2c
+	title	{ return (size_t)( c - start ); }
+	.?			{ return 0; }
+*/	
+}
+
 size_t scan_setext(const char * c) {
 	const char * marker = NULL;
 	const char * start = c;
diff --git a/src/writer.c b/src/writer.c
index 276e957..f39eaae 100644
--- a/src/writer.c
+++ b/src/writer.c
@@ -722,90 +722,80 @@ char * destination_accept(const char * source, token ** remainder, bool validate
 }
 
 
-char * url_accept(const char * source, token ** remainder, bool validate) {
+char * url_accept(const char * source, size_t start, size_t max_len, size_t * end_pos, bool validate) {
 	char * url = NULL;
 	char * clean = NULL;
-	token * t = NULL;
-	token * first = NULL;
-	token * last = NULL;
+	size_t scan_len;
 
-	switch ((*remainder)->type) {
-		case PAIR_PAREN:
-		case PAIR_ANGLE:
-		case PAIR_QUOTE_SINGLE:
-		case PAIR_QUOTE_DOUBLE:
-			t = token_chain_accept_multiple(remainder, 2, PAIR_ANGLE, PAIR_PAREN);
-			url = text_inside_pair(source, t);
-			break;
-		case SLASH:
-		case TEXT_PLAIN:
-			first = *remainder;
-			
-			// Grab parts for URL
-			while (token_chain_accept_multiple(remainder, 7, AMPERSAND, COLON, EQUAL, SLASH, TEXT_PERIOD, TEXT_PLAIN, UL));
+	scan_len = scan_destination(&source[start]);
 
-			last = (*remainder)->prev;
+	if (scan_len) {
+		if (scan_len > max_len)
+			scan_len = max_len;
 
-			// Is there a space in a URL concatenated with a title or attribute?
-			// e.g. [foo]: http://foo.bar/ class="foo"
-			// Since only one space between URL and class, they are joined.
+		if (end_pos)
+			*end_pos = start + scan_len;
 
-			if (last->type == TEXT_PLAIN) {
-				// Trim leading whitespace
-				token_trim_leading_whitespace(last, source);
-				token_split_on_char(last, source, ' ');
-				*remainder = last->next;
-			}
+		// Is this <foo>?
+		if ((source[start] == '<') &&
+			(source[start + scan_len - 1] == '>')) {
+			// Strip '<' and '>'
+			start++;
+			scan_len -= 2;
+		}
 
-			url = strndup(&source[first->start], last->start + last->len - first->start);
-			break;
-	}
+		url = strndup(&source[start], scan_len);
 
-	// Is this a valid URL?
-	clean = clean_string(url, false);
-	
-	if (validate && !validate_url(clean)) {
-		free(clean);
-		clean = NULL;
+		clean = clean_string(url, false);
+
+		if (validate && !validate_url(clean)) {
+			free(clean);
+			clean = NULL;
+		}
+
+		free(url);
 	}
 
-	free(url);
 	return clean;
 }
 
 
 /// Extract url string from `(foo)` or `(<foo>)` or `(foo "bar")`
 void extract_from_paren(token * paren, const char * source, char ** url, char ** title, char ** attributes) {
-	token * t;
+   size_t scan_len;
+    size_t pos = paren->child->next->start;
+    
+    
 	size_t attr_len;
 
-	token * remainder = paren->child->next;
+	// Skip whitespace
+	while (char_is_whitespace(source[pos]))
+		pos++;
 
-	if (remainder) {
-		// Skip whitespace
-		whitespace_accept(&remainder);
+	// Grab URL
+	*url = url_accept(source, pos, paren->start + paren->len - 1 - pos, &pos, false);
 
-		// Grab URL
-		*url = url_accept(source, &remainder, false);
+	// Skip whitespace
+	while (char_is_whitespace(source[pos]))
+		pos++;
 
-		// Skip whitespace
-		whitespace_accept(&remainder);
+	// Grab title, if present
+	scan_len = scan_title(&source[pos]);
 
-		// Grab title, if present
-		t = token_chain_accept_multiple(&remainder, 3, PAIR_QUOTE_DOUBLE, PAIR_QUOTE_SINGLE, PAIR_PAREN);
+	if (scan_len) {
+		*title = strndup(&source[pos + 1], scan_len - 2);
+		pos += scan_len;
+	}
 
-		if (t) {
-			*title = text_inside_pair(source, t);
-		}
+	// Skip whitespace
+	while (char_is_whitespace(source[pos]))
+		pos++;
 
-		// Grab attributes, if present
-		if (t) {
-			attr_len = scan_attributes(&source[t->start + t->len]);
-			
-			if (attr_len) {
-				*attributes = strndup(&source[t->start + t->len], attr_len);
-			}
-		}
+	// Grab attributes, if present
+	attr_len = scan_attributes(&source[pos]);
+	
+	if (attr_len) {
+		*attributes = strndup(&source[pos], attr_len);
 	}
 }
 
diff --git a/src/writer.h b/src/writer.h
index 85b92d5..b20b1f3 100644
--- a/src/writer.h
+++ b/src/writer.h
@@ -189,7 +189,7 @@ void print_token_raw(DString * out, const char * source, token * t);
 
 void print_token_tree_raw(DString * out, const char * source, token * t);
 
-char * url_accept(const char * source, token ** remainder, bool validate);
+char * url_accept(const char * source, size_t start, size_t max_len, size_t * end_pos, bool validate);
 
 void footnote_from_bracket(const char * source, scratch_pad * scratch, token * t, short * num);
 void citation_from_bracket(const char * source, scratch_pad * scratch, token * t, short * num);
diff --git a/tests/MMD6Tests/Inline Links.html b/tests/MMD6Tests/Inline Links.html
index 664c4b7..6f71ab4 100644
--- a/tests/MMD6Tests/Inline Links.html	
+++ b/tests/MMD6Tests/Inline Links.html	
@@ -30,3 +30,15 @@
 
 <p>[URL and title]
 (/url/file.txt &#8220;<em>title</em>&#8221;).</p>
+
+<p><a href="/url/file.txt" title="title">URL and title</a>.</p>
+
+<p>15</p>
+
+<p><a href="/url/file.txt" title="title">URL and title</a>.</p>
+
+<p><a href="/url/file.txt" title="title">URL and title</a>.</p>
+
+<p><a href="/url/file.txt" title="title">URL and title</a>.</p>
+
+<p><a href="/url/file.txt" title="title">URL and title</a>.</p>
diff --git a/tests/MMD6Tests/Inline Links.htmlc b/tests/MMD6Tests/Inline Links.htmlc
index 302544f..2394629 100644
--- a/tests/MMD6Tests/Inline Links.htmlc	
+++ b/tests/MMD6Tests/Inline Links.htmlc	
@@ -30,3 +30,15 @@
 
 <p>[URL and title]
 (/url/file.txt &quot;<em>title</em>&quot;).</p>
+
+<p><a href="/url/file.txt" title="title">URL and title</a>.</p>
+
+<p>15</p>
+
+<p><a href="/url/file.txt" title="title">URL and title</a>.</p>
+
+<p><a href="/url/file.txt" title="title">URL and title</a>.</p>
+
+<p><a href="/url/file.txt" title="title">URL and title</a>.</p>
+
+<p><a href="/url/file.txt" title="title">URL and title</a>.</p>
diff --git a/tests/MMD6Tests/Inline Links.text b/tests/MMD6Tests/Inline Links.text
index a6a4095..16d4d38 100644
--- a/tests/MMD6Tests/Inline Links.text	
+++ b/tests/MMD6Tests/Inline Links.text	
@@ -30,3 +30,15 @@ Just a [URL](http://url/file.txt).
 
 [URL and title]
 (/url/file.txt "*title*").
+
+[URL and title]( /url/file.txt "title").
+
+15
+
+[URL and title](    /url/file.txt "title").
+
+[URL and title](</url/file.txt> "title").
+
+[URL and title]( </url/file.txt> "title").
+
+[URL and title](  </url/file.txt> "title").