aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/mandoc.c
diff options
context:
space:
mode:
authorIngo Schwarze <schwarze@openbsd.org>2014-10-28 13:24:44 +0000
committerIngo Schwarze <schwarze@openbsd.org>2014-10-28 13:24:44 +0000
commit747bf8062cca375c58d984004718a2351a88daed (patch)
treec6a09911f1ddd799d8fc34009ce98e10bab9ad77 /mandoc.c
parent7f43f205b47aa5a68d76c5c14c3d5bb261df1cff (diff)
downloadmandoc-747bf8062cca375c58d984004718a2351a88daed.tar.gz
mandoc-747bf8062cca375c58d984004718a2351a88daed.tar.zst
mandoc-747bf8062cca375c58d984004718a2351a88daed.zip
Tighten Unicode escape name parsing.
Accept only 0xXXXX, 0xYXXXX, 0x10XXXX with Y != 0. This simplifies mchars_num2uc().
Diffstat (limited to 'mandoc.c')
-rw-r--r--mandoc.c15
1 files changed, 10 insertions, 5 deletions
diff --git a/mandoc.c b/mandoc.c
index e82093b9..2ec179ea 100644
--- a/mandoc.c
+++ b/mandoc.c
@@ -1,4 +1,4 @@
-/* $Id: mandoc.c,v 1.87 2014/10/13 17:17:45 schwarze Exp $ */
+/* $Id: mandoc.c,v 1.88 2014/10/28 13:24:44 schwarze Exp $ */
/*
* Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
* Copyright (c) 2011, 2012, 2013, 2014 Ingo Schwarze <schwarze@openbsd.org>
@@ -334,13 +334,18 @@ mandoc_escape(const char **end, const char **start, int *sz)
if (1 == *sz && 'c' == **start)
gly = ESCAPE_NOSPACE;
/*
- * Unicode escapes are defined in groff as \[uXXXX]
+ * Unicode escapes are defined in groff as \[u0000]
* to \[u10FFFF], where the contained value must be
* a valid Unicode codepoint. Here, however, only
- * check the length and the validity of all digits.
+ * check the length and range.
*/
- else if (*sz > 4 && *sz < 8 && **start == 'u' &&
- (int)strspn(*start + 1, "0123456789ABCDEFabcdef")
+ if (**start != 'u' || *sz < 5 || *sz > 7)
+ break;
+ if (*sz == 7 && ((*start)[1] != '1' || (*start)[2] != '0'))
+ break;
+ if (*sz == 6 && (*start)[1] == '0')
+ break;
+ if ((int)strspn(*start + 1, "0123456789ABCDEFabcdef")
+ 1 == *sz)
gly = ESCAPE_UNICODE;
break;