URL: fix ASCII dependency in strcpy_url and strlen_url

Commit 3c630f9b0a partially reverted the
changes from commit dd7521bcc1 because of
the problem that strcpy_url() was modified unilaterally without also
modifying strlen_url(). As a consequence strcpy_url() was again
depending on ASCII encoding.

This change fixes strlen_url() and strcpy_url() in parallel to use a
common host-encoding independent criterion for deciding whether an URL
character must be %-escaped.

Closes #2535
This commit is contained in:
Stephan Mühlstrasser 2018-04-26 10:15:26 +02:00 committed by Daniel Stenberg
parent 0be4679ba9
commit 7f41432c19
No known key found for this signature in database
GPG key ID: 5CC908FDB71E12C2
3 changed files with 22 additions and 2 deletions

View file

@ -1446,6 +1446,16 @@ static const char *find_host_sep(const char *url)
return sep < query ? sep : query;
}
/*
* Decide in an encoding-independent manner whether a character in an
* URL must be escaped. The same criterion must be used in strlen_url()
* and strcpy_url().
*/
static bool urlchar_needs_escaping(int c)
{
return !(ISCNTRL(c) || ISSPACE(c) || ISGRAPH(c));
}
/*
* strlen_url() returns the length of the given URL if the spaces within the
* URL were properly URL encoded.
@ -1474,7 +1484,7 @@ static size_t strlen_url(const char *url, bool relative)
left = FALSE;
/* fall through */
default:
if(*ptr >= 0x80)
if(urlchar_needs_escaping(*ptr))
newlen += 2;
newlen++;
break;
@ -1519,7 +1529,7 @@ static void strcpy_url(char *output, const char *url, bool relative)
left = FALSE;
/* fall through */
default:
if(*iptr >= 0x80) {
if(urlchar_needs_escaping(*iptr)) {
snprintf(optr, 4, "%%%02x", *iptr);
optr += 3;
}