/// <summary>
/// The unsafeForAttributesMode tells the method to perform more aggressive
/// matching of "basic" entities, like IE does on non-markup HTML text.
/// However we can't do this kind of matching for attributes, since it
/// breaks URLs. When in doubt, use false.
///
/// Example:
///
/// UnEscapeEntities("£a", true) => "£a"
/// UnEscapeEntities("£a", false) => "£a"
/// </summary>
public static string UnEscapeEntities(string html, UnEscapeMode unEscapeMode)
{
if (html == null)
{
return(null);
}
StringBuilder output = new StringBuilder(html.Length);
int len = html.Length;
for (int i = 0; i < len; i++)
{
char c0 = html[i];
if (c0 == '&')
{
if (i + 1 < len)
{
char c1 = html[i + 1];
switch (c1)
{
case '#':
{
if (i + 2 < len)
{
char c2 = html[i + 2];
switch (c2)
{
case 'x':
case 'X':
{
// do hexadecimal match
bool semicolonTerminated = false;
int charVal = 0;
int j;
for (j = i + 3; j < len; j++)
{
int hexVal = ToHexValue(html[j]);
if (hexVal == -1)
{
// skip one more char if currently on semicolon
if (html[j] == ';')
{
semicolonTerminated = true;
}
break;
}
charVal *= 16;
charVal += hexVal;
}
if (semicolonTerminated && charVal != 0)
{
i = j;
output.Append((char)charVal);
continue;
}
// if total is 0, continue
break;
}
case '0':
case '1':
case '2':
case '3':
case '4':
case '5':
case '6':
case '7':
case '8':
case '9':
{
// do decimal match
int charVal = 0;
int j;
for (j = i + 2; j < len; j++)
{
char c = html[j];
if (c < '0' || c > '9')
{
if (c == ';')
{
++j;
}
break;
}
int cVal = c - '0';
charVal *= 10;
charVal += cVal;
}
if (charVal != 0)
{
i = j - 1;
output.Append((char)charVal);
continue;
}
// if total is 0, continue
break;
}
}
}
break;
}
default:
{
int j;
int end = Math.Min(len, i + 12);
for (j = i + 1; j < end; j++)
{
char c = html[j];
if (c == ';' || (!(c >= 'a' && c <= 'z') && !(c >= 'A' && c <= 'Z') && !(c >= '0' && c <= '9')))
{
break;
}
}
string entityRef = html.Substring(i + 1, j - (i + 1));
if (unEscapeMode != UnEscapeMode.Attribute)
{
// k = number of characters in entityRef that we are using
int k, code = -1;
for (k = 1; k < entityRef.Length; k++)
{
if (-1 != (code = EntityEscaper.Code(entityRef.Substring(0, k), true)))
{
break;
}
}
if (code == -1)
{
code = EntityEscaper.Code(entityRef, false);
}
if (code != -1)
{
output.Append((char)code);
i += 1 + k;
if (i < end && html[i] == ';')
{
++i;
}
--i;
continue;
}
}
else
{
int code = EntityEscaper.Code(entityRef, false);
if (code != -1)
{
output.Append((char)code);
i += 1 + entityRef.Length;
if (i < end && html[i] == ';')
{
++i;
}
--i;
continue;
}
}
break;
}
}
}
}
output.Append(c0);
}
return(output.ToString());
}
/// <summary>
/// The unsafeForAttributesMode tells the method to perform more aggressive
/// matching of "basic" entities, like IE does on non-markup HTML text.
/// However we can't do this kind of matching for attributes, since it
/// breaks URLs. When in doubt, use false.
///
/// Example:
///
/// UnEscapeEntities("£a", true) => "£a"
/// UnEscapeEntities("£a", false) => "£a"
/// </summary>
public static string UnEscapeEntities(string html, UnEscapeMode unEscapeMode)
{
if (html == null)
return null;
StringBuilder output = new StringBuilder(html.Length);
int len = html.Length;
for (int i = 0; i < len; i++)
{
char c0 = html[i];
if (c0 == '&')
{
if (i + 1 < len)
{
char c1 = html[i + 1];
switch (c1)
{
case '#':
{
if (i + 2 < len)
{
char c2 = html[i + 2];
switch (c2)
{
case 'x':
case 'X':
{
// do hexadecimal match
bool semicolonTerminated = false;
int charVal = 0;
int j;
for (j = i + 3; j < len; j++)
{
int hexVal = ToHexValue(html[j]);
if (hexVal == -1)
{
// skip one more char if currently on semicolon
if (html[j] == ';')
semicolonTerminated = true;
break;
}
charVal *= 16;
charVal += hexVal;
}
if (semicolonTerminated && charVal != 0)
{
i = j;
output.Append((char)charVal);
continue;
}
// if total is 0, continue
break;
}
case '0':
case '1':
case '2':
case '3':
case '4':
case '5':
case '6':
case '7':
case '8':
case '9':
{
// do decimal match
int charVal = 0;
int j;
for (j = i + 2; j < len; j++)
{
char c = html[j];
if (c < '0' || c > '9')
{
if (c == ';')
++j;
break;
}
int cVal = c - '0';
charVal *= 10;
charVal += cVal;
}
if (charVal != 0)
{
i = j - 1;
output.Append((char)charVal);
continue;
}
// if total is 0, continue
break;
}
}
}
break;
}
default:
{
int j;
int end = Math.Min(len, i + 12);
for (j = i + 1; j < end; j++)
{
char c = html[j];
if (c == ';' || (!(c >= 'a' && c <= 'z') && !(c >= 'A' && c <= 'Z') && !(c >= '0' && c <= '9')))
{
break;
}
}
string entityRef = html.Substring(i + 1, j - (i + 1));
if (unEscapeMode != UnEscapeMode.Attribute)
{
// k = number of characters in entityRef that we are using
int k, code = -1;
for (k = 1; k < entityRef.Length; k++)
{
if (-1 != (code = EntityEscaper.Code(entityRef.Substring(0, k), true)))
break;
}
if (code == -1)
{
code = EntityEscaper.Code(entityRef, false);
}
if (code != -1)
{
output.Append((char)code);
i += 1 + k;
if (i < end && html[i] == ';')
++i;
--i;
continue;
}
}
else
{
int code = EntityEscaper.Code(entityRef, false);
if (code != -1)
{
output.Append((char)code);
i += 1 + entityRef.Length;
if (i < end && html[i] == ';')
++i;
--i;
continue;
}
}
break;
}
}
}
}
output.Append(c0);
}
return output.ToString();
}
