From 9d86d2f32303ae0048f6a5daa552bacceb9b12ea Mon Sep 17 00:00:00 2001 From: Vicent Marti Date: Tue, 9 Sep 2014 04:00:36 +0200 Subject: Update the spec with better entity handling --- Makefile | 4 ++-- spec.txt | 22 ++++++++++++---------- src/html/houdini_html_u.c | 2 +- 3 files changed, 15 insertions(+), 13 deletions(-) diff --git a/Makefile b/Makefile index b5e487d..5d13272 100644 --- a/Makefile +++ b/Makefile @@ -1,5 +1,5 @@ -CFLAGS=-g -pg -O3 -Wall -Wextra -std=c99 -Isrc $(OPTFLAGS) -LDFLAGS=-g -pg -O3 -Wall -Werror +CFLAGS=-g -O3 -Wall -Wextra -std=c99 -Isrc $(OPTFLAGS) +LDFLAGS=-g -O3 -Wall -Werror SRCDIR=src DATADIR=data diff --git a/spec.txt b/spec.txt index ebd6d98..112dccc 100644 --- a/spec.txt +++ b/spec.txt @@ -3762,20 +3762,20 @@ as the "unknown codepoint" character (`0xFFFD`) [Hexadecimal entities](#hexadecimal-entities) consist of `&#` + either `X` or `x` + a string of 1-8 hexadecimal digits -+ `;`. ++ `;`. They will also be parsed and turned into their corresponding UTF8 values in the AST. . - " ആ ಫ +" ആ ಫ . -

 " ആ ಫ

+

" ആ ಫ

. Here are some nonentities: . -  &x; &#; &#x; � &ThisIsWayTooLongToBeAnEntityIsntIt; &hi?; +  &x; &#; &#x; &ThisIsWayTooLongToBeAnEntityIsntIt; &hi?; . -

&nbsp &x; &#; &#x; � &ThisIsWayTooLongToBeAnEntityIsntIt; &hi?;

+

&nbsp &x; &#; &#x; &ThisIsWayTooLongToBeAnEntityIsntIt; &hi?;

. Although HTML5 does accept some entities without a trailing semicolon @@ -3808,7 +3808,7 @@ code blocks, including raw HTML, URLs, [link titles](#link-title), and . [foo](/föö "föö") . -

foo

+

foo

. . @@ -3816,7 +3816,7 @@ code blocks, including raw HTML, URLs, [link titles](#link-title), and [foo]: /föö "föö" . -

foo

+

foo

. . @@ -3824,7 +3824,7 @@ code blocks, including raw HTML, URLs, [link titles](#link-title), and foo ``` . -
foo
+
foo
 

 .
 
@@ -4817,12 +4817,14 @@ in Markdown:
 
link

 .
 
-URL-escaping and entities should be left alone inside the destination:
+URL-escaping and should be left alone inside the destination, as all URL-escaped characters
+are also valid URL characters. HTML entities in the destination will be parsed into their UTF8
+codepoints, as usual, and optionally URL-escaped when written as HTML.
 
 .
 [link](foo%20bä)
 .
-
link

+
link

 .
 
 Note that, because titles can often be parsed as destinations,
diff --git a/src/html/houdini_html_u.c b/src/html/houdini_html_u.c
index 762f980..b8e2d8d 100644
--- a/src/html/houdini_html_u.c
+++ b/src/html/houdini_html_u.c
@@ -24,7 +24,7 @@ houdini_unescape_ent(strbuf *ob, const uint8_t *src, size_t size)
 				codepoint = (codepoint * 16) + ((src[i] | 32) % 39 - 9);
 		}
 
-		if (i < size && src[i] == ';') {
+		if (i < size && src[i] == ';' && codepoint) {
 			utf8proc_encode_char(codepoint, ob);
 			return i + 1;
 		}
-- 
cgit v1.2.3